Clinical Topic A-Z Clinical Speciality

Urinary tract infection (lower) - women

Urinary tract infection (lower) - women
D014552Urinary Tract Infections
D003556Cystitis
Infections and infestationsKidney disease and urologyWomen's health
2013-11-01Last revised in November 2013

Urinary tract infection (lower) - women - Summary

A urinary tract infection (UTI) is usually caused by bacteria from the gastrointestinal tract, most commonly Escherichia coli.

Acute, uncomplicated UTI is a benign condition that usually resolves in a few days.

Ascending infection can occur, leading to pyelonephritis, renal failure, and sepsis. In pregnant women, symptomatic UTI can lead to developmental delay or cerebral palsy in the infant, and fetal death.

A lower UTI can be confidently diagnosed when there are typical symptoms and signs of UTI, a urinary pathogen has been detected, and an upper UTI and other alternative diagnoses has been excluded. Typical features include:

Urinary frequency, urgency, and/or strangury.

Dysuria.

Urine that is offensive smelling, cloudy or contains blood.

Constant lower abdominal ache.

Non-specific malaise, such as aching all over, nausea, tiredness and cold sweats.

Urge incontinence.

Accurate diagnosis of UTI is difficult if the woman is frail and elderly, is institutionalized, or has a long-term indwelling urinary catheter. In these situations bacteriuria is often present but is not related to symptoms and signs.

If a woman has few or mild symptoms and signs of a UTI, and is not catheterized, a urine dipstick test should be done to look for further evidence of a UTI.

If both dipstick tests are negative, a UTI is unlikely.

If the leucocyte esterase test alone is positive, a UTI is moderately likely.

If the nitrite test is positive, with or without a positive leucocyte esterase test, a UTI is highly likely.

If a woman has:

Typical or severe symptoms of a UTI, a urine dipstick test should not be done as treatment with an antibiotic is recommended whatever the dipstick test shows.

An indwelling catheter, a urine dipstick test should not be done as the false-positive rate is high.

A urine sample should be sent for microscopy and culture:

When it is important to ensure that a suitable antibiotic is used to treat a lower UTI (e.g. in pregnant women or women with persistent symptoms).

To confirm UTI and exclude non-bacterial causes in women with recurrent lower UTI.

To screen for asymptomatic bacteriuria in pregnancy.

To confirm cure of lower UTI in pregnancy.

Management includes:

Advising the women to seek medical attention if she develops a high fever or becomes systemically unwell.

Treatment with antibiotics.

Follow up is not routinely required for uncomplicated cystitis, but should be considered for a potentially complicated infection.

In pregnancy, women with asymptomatic bacteriuria and suspected or proven UTIs should be treated promptly with a 7-day course of antibiotics and followed up.

Follow up should be arranged for all women with visible or non–visible haematuria, and women should be referred urgently for investigations of suspected urological cancer if:

Infection is not confirmed on culture.

Visible haematuria persists after infection has been successfully treated.

Non-visible haematuria persists after infection has been successfully treated in a woman over 50 years of age.

Visible or non-visible haematuria is associated with persistent or recurrent urinary tract infection in a woman aged 40 years or older.

Routine referral is recommended for women with recurrent UTI's:

Who have a risk factor for an abnormality of the urinary tract.

Who are immunocompromised or who have diabetes.

Who have a known abnormality of their renal tract who might benefit from surgical correction.

Who have not responded to preventive treatments

Have I got the right topic?

168months3060monthsFemale

This CKS topic covers the management of women with cystitis, asymptomatic bacteriuria, and lower urinary tract infection (UTI) in association with an indwelling urinary catheter.

This CKS topic does not cover the prevention of UTI following surgery or instrumentation; treatment of UTI in hospital; urethral syndrome or painful bladder syndrome; upper UTI; UTI in women with an abnormal urinary tract; or UTI in women with impaired renal function. It also does not cover the management of UTI in girls younger than 14 years of age.

There are separate CKS topics on other urinary tract conditions including Pyelonephritis - acute, Renal colic - acute, Urinary tract infection - children, and Urinary tract infection (lower) - men.

There are separate CKS topics on vaginal infections including Bacterial vaginosis, Candida - female genital, Herpes simplex - genital, and Trichomoniasis.

The target audience for this CKS topic is healthcare professionals working within the NHS in the UK, and providing first contact or primary health care.

How up-to-date is this topic?

How up-to-date is this topic?

Changes

Last revised in November 2013

August  2013 to November 2013 — reviewed. A literature search was conducted in August 2013 to identify evidence-based guidelines, UK policy, systematic reviews, and key RCTs published since the last revision of the topic.

This CKS topic incorporates recommendations from the Scottish Intercollegiate Guidelines Network (SIGN) guideline Management of suspected bacterial urinary tract infection in adults: a national clinical guideline [SIGN, 2012].

There have been changes to the structure and minor changes to the recommendations. Cranberry products are no longer recommended for prophylaxis of recurrent UTI.

Previous changes

February 2013 — minor update. The 2013 QIPP options for local implementation have been added to this topic [NICE, 2013].

October 2012 — minor update. The 2012 QIPP options for local implementation have been added to this topic [NPC, 2012].

May 2011 — minor update. The 2010/2011 QIPP options for local implementation have been added to this topic [NPC, 2011]. Issued in June 2011.

April 2011 — minor update. Changes to text regarding the safety of nitrofurantoin and trimethoprim in pregnancy. Issued in June 2011.

March 2011 — topic structure revised to ensure consistency across CKS topics — no changes to clinical recommendations have been made.

March 2011 — minor update. Minor clarification to the text. Issued in March 2011.

December 2010 — minor update. Nitrofurantoin capsules have been added as an alternative option to nitrofurantoin tablets in the Prescriptions. The Supporting evidence section on Nitrofurantoin formulations has also been updated. Issued in December 2010.

August 2010 — minor update. In pregnant women with suspected symptomatic cystitis, amoxicillin is not recommended for empirical use, but is an option if the pathogen is reported to be susceptible [HPA and Association of Medical Microbiologists, 2010]. Issued in September 2010.

December 2009 — minor update. Dose of trimethoprim recommended for post-coital prophylaxis reduced to 100 mg stat. Issued in December 2009.

April to October 2009 — converted from CKS guidance to CKS topic structure. The evidence-base has been reviewed in detail, and recommendations are more clearly justified and transparently linked to the supporting evidence.

There have been no major changes to the recommendations on treatment, although some major changes have been made to the structure.

The approach to managing suspected cystitis reflects a change in focus from basing treatment decisions on accurate diagnosis to basing treatment decisions on the assessment of the patient's risks and preferences.

October 2008 — minor update to reflect guidance from the Health Protection Agency (HPA) to avoid broad spectrum antibiotics (such as co-amoxiclav, quinolones, and cephalosporins) when narrow spectrum antibiotics remain effective, as broad spectrum antibiotics increase the risk of Clostridium difficile, MRSA, and resistant UTIs [HPA and Association of Medical Microbiologists, 2008]. Issued October 2008.

January to March 2006 — reviewed. Validated in June 2006 and issued in July 2006.

September 2008 — minor correction to the Changes section. Issued September 2008.

April 2008 — minor update. Guidance updated to be in line with the SIGN guideline on the Management of Suspected Bacterial Urinary Tract Infection in Adults. Issued in May 2008.

November 2007 — minor update to clarify the use of trimethoprim in women who are pregnant or at risk of pregnancy.

October 2005 — minor technical update. Issued in November 2005.

July 2005 — updated to incorporate the Referral guidelines for suspected cancer published by the National Institute for Health and Care Excellence. Issued in July 2005.

January 2002 — written replacing previous guidance on UTI (lower) — acute and UTI (lower) — recurrent. Validated in March 2002 and issued in April 2002.

Update

New evidence

Evidence-based guidelines

No new evidence-based guidelines published since 1 August 2013.

HTAs (Health Technology Assessments)

No new HTAs since 1 August 2013.

Economic appraisals

No new economic appraisals since 1 August 2013.

Systematic reviews and meta-analyses

No new systematic reviews and meta-analyses since 1 August 2013.

Primary evidence

No new randomized controlled trials published in the major journals since 1 August 2013.

New policies

No new national policies or guidelines since 1 August 2013.

New safety alerts

No new safety alerts since 1 August 2013.

Changes in product availability

No changes in product availability since 1 August 2013.

Goals and outcome measures

Goals

To promptly recognize urinary tract infection

To effectively relieve symptoms

To avoid unnecessary use of antibiotics

To prevent recurrence

To prevent complications

QIPP - options for local implementation

QIPP - options for local implementation

Non-steroidal anti-inflammatory drugs (NSAIDs)

Review the appropriateness of NSAID prescribing widely and on a routine basis, especially in people who are at higher risk of both gastrointestinal (GI) and cardiovascular (CV) morbidity and mortality (e.g. older patients).

If initiating an NSAID is obligatory, use ibuprofen (1200 mg per day or less) or naproxen (1000 mg per day or less).

Review patients currently prescribed NSAIDs. If continued use is necessary, consider changing to ibuprofen (1200 mg per day or less) or naproxen (1000 mg per day or less).

Review and, where appropriate, revise prescribing of etoricoxib to ensure it is in line with MHRA advice and the NICE clinical guideline on osteoarthritis [CSM, 2005; NICE, 2008d].

Co-prescribe a proton pump inhibitor (PPI) with NSAIDs for people with osteoarthritis, rheumatoid arthritis, or low back pain (for people over 45 years) in accordance with NICE guidance [NICE, 2008d; NICE, 2009a; NICE, 2009b].

Take account of drug interactions when co-prescribing NSAIDs with other medicines (see Summaries of Product Characteristics). For example, co-prescribing NSAIDs with ACE inhibitors or angiotensin receptor blockers (ARBs) may pose particular risks to renal function; this combination should be especially carefully considered and regularly monitored if continued.

Antibiotic prescribing — especially quinolones and cephalosporins

Review and, where appropriate, revise current prescribing practice and use implementation techniques to ensure prescribing is in line with Health Protection Agency (HPA) guidance.

Review the total volume of antibiotic prescribing against local and national data.

Review the use of quinolones and cephalosporin prescribing against local and national data.

Antibiotic prescribing — especially quinolones and cephalosporins

Review and, where appropriate, revise current prescribing practice and use implementation techniques to ensure prescribing is in line with Health Protection Agency (HPA) guidance.

Review the total volume of antibiotic prescribing against local and national data.

Review the use of quinolones and cephalosporin prescribing against local and national data.

Three-day courses of trimethoprim for uncomplicated urinary tract infection:

Review and, where appropriate, revise current prescribing practise and use implementation techniques to ensure prescribing of three-day courses of trimethoprim is in line with HPA guidance.

[NICE, 2013]

Background information

Definition

What is it?

In this CKS topic, the following definitions apply:

Bacteriuria: bacteria are present in the urine.

Significant bacteriuria: the count of bacteria in the urine exceeds the threshold for diagnosing the presence of infection in the urine. The threshold depends on the age and sex of the person, and on the risk of contamination during collection of the urine sample.

Asymptomatic bacteriuria: there is significant bacteriuria without symptoms and signs of infection.

Urinary tract infection (UTI): there is significant bacteriuria and characteristic symptoms and signs — see When to suspect UTI.

Lower UTI means infection of the bladder, the main concern of this CKS topic. The term lower UTI is used only with reference to UTI in women with an indwelling catheter, because in this situation the symptoms and signs of UTI are seldom typical of cystitis.

Cystitis is used as a synonym for lower UTI (although technically it means inflammation of the bladder and there are rare non-infectious causes of cystitis, such as radiation and chemicals).

Upper UTI includes pyelitis (infection of the proximal part of the ureters) and pyelonephritis (infection of the kidneys and the proximal part of the ureters).

Recurrent UTI: repeated UTI, which may be due to relapse or reinfection.

Relapse is recurrent UTI with the same strain of organism. Relapse is the likely cause if infection recurs within a short period (for example within 2 weeks) after treatment.

Reinfection is recurrent UTI with a different strain or species of organism. Reinfection is the likely cause if UTI recurs more than 2 weeks after treatment.

The number of recurrences that is regarded as clinically significant depends on the risks of infection and the impact of cystitis on the woman.

Uncomplicated UTI: infection of the urinary tract by a usual pathogen (see Causes) in a person with a normal urinary tract and with normal kidney function.

Complicated UTI: UTI when one or more factors are present that predispose the person to persistent infection, recurrent infection, or treatment failure. Examples include UTI with:

Abnormal urinary tract (for example calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, indwelling catheter, urinary obstruction, recent instrumentation).

Virulent organism (such as Staphylococcus aureus).

Impaired host defences (for example poorly controlled diabetes mellitus, immunosuppressive treatment).

Impaired renal function.

Urethral syndrome or painful bladder syndrome: symptoms of cystitis in the absence of urinary tract infection. This syndrome is also called interstitial cystitis, bladder pain syndrome, and trigonitis.

Causes

What causes it?

Urinary tract infection (UTI) is usually caused by bacteria from the gastrointestinal tract. A survey of 252 community healthcare centres in 17 countries found that, in women presenting with acute symptoms of UTI [Kahlmeter, 2003]:

Escherichia coli accounted for about 80% of UTIs.

Other causative organisms were Staphylococcus species, Proteus mirabilis, and enterococci.

Candida albicans rarely causes UTI. When it does, it is usually in hospitalized people with risk factors such as an indwelling catheter, immunosuppression, diabetes mellitus, or antibiotic treatment.

Pathophysiology

How does the urinary tract become infected?

For the urinary tract to become infected, bacteria must gain entry to the urinary tract, avoid being flushed away, adhere to the epithelial surface, and multiply.

Entry of bacteria into the urinary tract may be:

Retrograde, with ascension through the urethra into the bladder.

Via the blood stream, which is more likely in people who are immunosuppressed.

Direct, for example with insertion of a catheter into the bladder, instrumentation, or surgery.

Incomplete emptying of the bladder during micturition is a key factor predisposing to urinary tract infection.

Many bacteria adhere to the surface of the urinary tract with the aid of fimbriae.

Bacteria adhering to the urothelium employ a number of poorly understood methods to optimize their environment for rapid multiplication.

Prevalence

How common is it?

Urinary tract infection

Urinary tract infection (UTI) is one of the commonest bacterial infections managed in general practice, and is the reason for 1–3% of all GP consultations [MeReC, 1995].

About one in three women will have at least one UTI by 24 years of age [Foxman, 2002].

About one in two women will be treated for a symptomatic UTI during their lifetime [MeReC, 1995; DTB, 1998].

The annual incidence of UTI in women increases with age.

Asymptomatic bacteriuria

Asymptomatic bacteriuria is more common in women than men.

Bacteriuria during pregnancy (asymptomatic and symptomatic)

A review of UTI during pregnancy (searching Medline and Cochrane from inception to August 2003) [Le et al, 2004] reported that:

The incidence of asymptomatic bacteriuria during pregnancy was in the range 2–10%.

The incidence of acute cystitis during pregnancy was in the range 1–4%.

About 20–40% of women with asymptomatic bacteriuria develop pyelonephritis later in pregnancy.

Risk factors

What are the risk factors?

Risk factors for asymptomatic bacteriuria

What are the risk factors for asymptomatic bacteriuria in women?

Risk factors for asymptomatic bacteriuria in women include [SIGN, 2012]:

Being sexually active.

Advancing age (prevalence reported to be less than 3% in women 50–59 years of age, and up to 17% in women who are more than 75 years of age).

Coexisting diabetes (prevalence reported to be in the range 8–18%).

Receiving institutional care (prevalence reported to be in the range 6–57%).

Presence of an indwelling urinary catheter (100% with long-term catheter).

Pregnancy.

Risk factors for UTI

What are the risk factors for UTI in women?

Most urinary tract infections (UTIs) in women are not associated with a risk factor.

In women with recurrent UTIs, risk factors include [Foxman, 2002]:

Sexual intercourse — common risk factor for young women.

Atrophic urethritis and vaginitis (in postmenopausal women).

Abnormalities of urinary tract function (for example due to indwelling catheter, neuropathic bladder, vesicoureteric reflux, outflow obstruction, anatomical abnormalities, significant asymptomatic bacteriuria).

Incomplete bladder emptying (for example dysfunctional urination, or blocked or misplaced chronic indwelling catheter).

Female diaphragm, spermicide-coated condoms [Fihn et al, 1985; Fihn et al, 1998].

Previous urinary tract surgery.

Immunocompromise.

Prognosis

What is the prognosis?

Acute, uncomplicated urinary tract infection (UTI) is a benign condition that usually resolves in a few days.

A UK primary care-based study found that in women with suspected cystitis and at least moderately severe symptoms [Little et al, 2009]:

Symptoms resolved after an average (standard deviation in brackets) of:

3.32 (2.54) days in women treated with an antibiotic to which the pathogen was sensitive.

4.73 (2.91) days in women treated with an antibiotic to which the pathogen was resistant.

4.94 (3.82) days in women with infection not treated with an antibiotic.

4.30 (3.42) days in women with urethral syndrome/painful bladder syndrome (symptoms, but no urinary pathogen cultured).

Symptoms lasted longer in women who presented with more severe symptoms, who had previously had cystitis, or who had frequent somatic symptoms.

Symptoms resolved more rapidly in women who thought that their doctor had a positive approach to diagnosis and prognosis — this was independent of other variables.

Another study of cystitis in women, also based in UK general practice, had somewhat different findings [McNulty et al, 2006]:

The median time to resolution of symptoms after starting antibiotic treatment was:

4 days in women treated with an antibiotic to which the pathogen was sensitive.

7 days in women treated with an antibiotic to which the pathogen was resistant.

Cystitis recurs within a year in 25–50% of women, usually as reinfections (rather than relapses) [Lawrenson and Logie, 2001].

Complications

What are the complications?

Complications of cystitis in non-pregnant women

Ascending infection can occur, leading to pyelonephritis, renal failure, and sepsis.

The risk of renal damage and renal failure is thought to be very low, although the evidence is indirect and data are limited [Hummers-Pradier and Kochen, 2002; Nickel et al, 2005].

Complications of asymptomatic bacteriuria in pregnancy

Pyelonephritis: a systematic review found that about 20% of pregnant women with untreated asymptomatic bacteriuria developed pyelonephritis (range 2.5–36%) [Smaill and Vazquez, 2007].

Complications of symptomatic urinary tract infection in pregnancy

Developmental delay or cerebral palsy in the infant, and fetal death [Foxman, 2002].

Diagnosis

Diagnosis of urinary tract infection (lower) - women

When to suspect UTI

How do I make a working diagnosis of lower urinary tract infection?

A definitive diagnosis of urinary tract infection (UTI) requires microbiological confirmation of bacteria in the urine which generally takes a few days to be reported and is generally not required. To avoid delays, the decision to offer treatment is based on an assessment of symptoms and, where appropriate, testing the urine with dipsticks.

To make a working diagnosis of lower UTI:

Identify features of UTI.

Typical features include:

Urinary frequency, urgency, and/or strangury (the feeling of needing to pass urine despite having just done so).

Dysuria (pain or discomfort on passing urine).

Urine that is offensive smelling, cloudy, or contains blood.

Constant lower abdominal ache.

Non-specific malaise, such as aching all over, nausea, tiredness and cold sweats.

Urge incontinence.

Typical features may be absent in frail, elderly women in nursing homes (catheterized and non-catheterized), and it is important to exclude other sources of infection. However the presence of one of the following suggests a UTI:

New costovertebral tenderness.

Rigors.

New onset delirium.

Fever greater than 37.9°C or 1.5°C above baseline.

Identify people with loin pain and/or fever and if present suspect acute pyelonephritis and manage accordingly. For further information see the CKS topic on Pyelonephritis - acute.

For women with mild symptoms who are not catheterized, dipstick test the urine for leucocyte esterase and nitrite to look for further evidence of UTI.

If both dipstick tests are negative, a UTI is unlikely.

If the leucocyte esterase test alone is positive, a UTI is moderately likely.

If the nitrite test is positive, with or without a positive leucocyte esterase test, a UTI is highly likely.

For women with mild symptoms who are catheterized, make a working diagnosis of UTI based on clinical judgment without testing the urine with a dipstick.

For women with moderate or severe symptoms, with or without a catheter, make a working diagnosis of UTI without testing the urine with a dipstick.

For information on when to send urine for culture to make a definitive diagnosis, see the appropriate management scenarios.

Basis for recommendation

Basis for recommendation

Making a working diagnosis of urinary tract infection (UTI)

Typical symptoms are present

The information on typical symptoms of a UTI has been taken from the Oxford textbook of medicine [Tomson and Armitage, 2010].

A set of health technology assessment studies of diagnostic algorithms in UTI concluded that UTI cannot be confidently confirmed or excluded on the basis of clinical features alone [Little et al, 2009]. The studies found that four symptoms independently predicted UTI:

Urine is cloudy on examination: odds ratio (OR) 2.32 (95% CI 1.40 to 3.85, p = 0.001).

Urine smells offensive: OR 2.02 (95% CI 1.05 to 3.90, p = 0.034).

Dysuria is moderately severe: OR 2.76 (95% CI 1.78 to 4.28, p < 0.001).

Nocturia is moderately severe: OR 1.81 (95% CI 1.16 to 2.80, p = 0.008).

Because individual symptoms are inaccurate predictors, clinical prediction rules using these four symptoms to diagnose UTI were developed and assessed [Little et al, 2009].

The 'two or more of four symptoms' rule had a positive predictive value of 77% and negative predictive value of 54%.

The 'three or four of four symptoms' rule had a positive predictive value of 84%.

And, the 'none of four symptoms' rule had a negative predictive value of 71%.

Other variables were tested but did not reach significance. These included moderately severe daytime frequency, moderately severe urgency, moderately severe haematuria, backache, feeling unwell, abdominal pain, and lower abdominal tenderness.

Diagnosing a UTI in frail elderly women when typical symptoms are absent

The information on the frequently atypical presentation of symptoms of a UTI in elderly women has been taken from textbooks [Sobel and Kaye, 2010; Tomson and Armitage, 2010].

The Guideline Development Group (GDG) of the Scottish Intercollegiate Guidelines Network (SIGN) list four changes in symptoms and signs that they suggest should prompt a clinician to consider the presence of a UTI in elderly women in nursing homes. These are not evidence-based, and were derived from a diagnostic algorithm designed for use as part of a multifaceted intervention in a cluster randomized controlled trial [Loeb et al, 2005; SIGN, 2012].

Identifying women with fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a UTI in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Using a dipstick to inform diagnosis in women with mild symptoms

The Guideline Development Group (GDG) of the Scottish Intercollegiate Guidelines Network have recommended dipstick tests should only be used to diagnose bacteriuria in women with minimal symptoms and signs. Their recommendation is based on evidence from poor quality RCTs [SIGN, 2012]. The GDG point out that a negative dipstick test does not exclude bacteriuria.

Experts recommend urinary dipstick tests to detect nitrite and leucocyte esterase in women with few symptoms and signs of a UTI, as they give an indication of the probability of a UTI being present [COMPASS, 2012].

Not using a dipstick to inform diagnosis in women with a catheter

The recommendation not to use dipsticks in women with indwelling catheters is based on the findings from two well-conducted cohort studies, and one non-analytic study that detecting pyuria by dipstick examination is of no value in differentiating asymptomatic bacteriuria from a UTI [SIGN, 2012].

Not using a dipstick to inform diagnosis in women with moderate or severe symptoms

Expert consensus from the GDG from SIGN is that in women with moderate or severe symptoms it is reasonable to start an antibiotic empirically [SIGN, 2012].

Differential diagnosis

What else might it be?

If there are urinary symptoms associated with fever and/or loin pain suspect pyelonephritis. See the CKS topic on Pyelonephritis - acute.

If there are urinary symptoms but no bacterial growth after culturing the urine suspect:

Urethral syndrome/painful bladder syndrome/interstitial cystitis if symptoms are relieved by voiding and aggravated by drinking alcohol or drinks containing caffeine.

Drug-induced cystitis in women treated with cyclophosphamide, allopurinol, danazol, or tiaprofenic acid.

Atrophic vaginitis/urethritis if a menopausal woman has vaginal discharge or itch, and pain during sexual intercourse.

Threadworms if urination causes discomfort outside the bladder and urethra, especially if there is peri-anal or peri-vaginal itching. For further information see the CKS topic on Threadworm.

Cervicitis, urethritis, or vaginitis if dysuria is associated with vaginal discharge.

Consider Chlamydia trachomatis infection if there are white blood cells in the urine without bacterial growth.

For further information see the CKS topics on Vaginal discharge, Bacterial vaginosis, Candida - female genital, Chlamydia - uncomplicated genital, Herpes simplex - genital, and Trichomoniasis.

If there are symptoms suggestive of urinary tract infection (UTI) associated with haematuria culture the urine and treat any infection.

Suspect urological cancer if:

Infection is not confirmed.

There are recurrent or persistent UTIs associated with haematuria in people aged 40 years or older.

Visible haematuria persists after infection has been treated.

Non-visible haematuria persists after infection has been treated in people more than 50 years of age.

There is an abdominal mass, identified clinically or on imaging, that is thought to arise from the urinary tract.

If non-visible haematuria persists after infection has been treated in people less than 50 years of age, test the urine for proteinuria and measure serum creatinine levels.

Suspect renal disease in those with proteinuria or raised serum creatinine.

Suspect non-malignant urological disease, such renal calculi, if there is no proteinuria and serum creatinine is normal.

Basis for recommendation

Basis for recommendation

Differential diagnosis

Information on the differential diagnosis of a urinary tract infection is based on a review article [Parsons and Toozs-Hobson, 2005].

Managing a urinary tract infection associated with visible or non-visible haematuria

The recommendations managing symptoms suggestive of underlying urological disease (including cancer), or renal disease are based on the National Institute for Health and Clinical Excellence Guideline Referral guidelines for suspected cancer [National Collaborating Centre for Primary Care, 2005].

Management

Management

Scenario: UTI — (no visible haematuria, not pregnant or catheterized) : covers the management of a suspected urinary tract infection (UTI) in women who do not have haematuria, and who are not pregnant or catheterized. It includes the management of treatment failure following initial antibiotic treatment.

Scenario: UTI — with visible or non-visible haematuria : covers the management of all women with a UTI associated with haematuria. 

Scenario: Recurrent UTI — (no visible haematuria, not pregnant or catheterized) : covers the management of recurrent UTIs that are not associated with haematuria in women who are not pregnant or catheterized. It includes recommendations on measures to prevent recurrent UTIs and when to refer to a specialist for investigation of possible underlying causes.

Scenario: UTI in pregnancy — no visible haematuria : covers the management of a suspected UTI during pregnancy in women who do not have visible haematuria.

Scenario: Asymptomatic bacteriuria in pregnancy : covers screening for and treating asymptomatic bacteriuria in pregnancy.

Scenario: UTI with catheter — no haematuria : covers the management of lower urinary tract infection in women with an indwelling urinary catheter.

Scenario: UTI — (no visible haematuria, not pregnant or catheterized)

Scenario: Suspected UTI without visible haematuria in women who are not catheterized or pregnant

168months3060monthsFemale

When to send urine for culture

When should I culture the urine of a woman with a suspected urinary tract infection?

Send urine for culture from women with a first presentation of a urinary tract infection (UTI) if they have any of the following:

Impaired renal function.

An abnormal urinary tract (for example renal calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, urinary obstruction, or recent instrumentation).

Immunosuppression (for example because they have poorly controlled diabetes mellitus or are receiving immunosuppressive treatment).

Do not routinely send urine for culture in women with an uncomplicated urinary tract infection.

Ensure that the urine is collected and stored properly. For further information see Collecting and storing urine samples.

Collecting and storing urine samples

Collecting and storing urine samples

Collecting and storing urine samples

Collection

To minimize the risk of contamination, the peri-urethral area should be cleaned, and then about 10 mL of urine collected mid-stream (without interrupting the urine flow to start or stop the collection) into a sterile container [HPA, 2011].

In a woman with a chronic indwelling urinary catheter, using an aseptic technique, drain a few millilitres of urine, then collect the urine sample from the catheter or sampling port — do not take the specimen from the collection bag, as this is more likely to be contaminated [HPA, 2011].

Containers

Urine should be transferred from the collecting vessel to a sterile specimen bottle within 30 minutes of collection. However, some evidence suggests that false-positive cultures only become clinically important when urine is stored at ambient temperature for 2–4 hours [Sanderson, 1998; Simerville et al, 2005].

The bottle should be filled to the marked line if it contains boric acid as a preservative. This is because a low volume of urine can create a concentration of boric acid high enough to be bactericidal [HPA, 2011].

Storage

Refrigerate urine at 4°C while awaiting processing.

Urine that has been stored at 4°C for 48 hours is suitable for culture but not for microscopy, because most cells will have disintegrated [Sanderson, 1998].

Urine preserved with boric acid may be stored for up to 96 hours [HPA, 2011].

Basis for recommendation

Basis for recommendation

These recommendations have largely been based on the Scottish Intercollegiate Guidelines Network guideline Management of suspected bacterial urinary tract infection in adults [SIGN, 2012].

Sending urine for culture for complicated urinary tract infections (UTIs)

Urine culture is recommended for women with a complicated infection because the risks associated with treatment failure are increased. Increasingly, infections in the community are caused by multi-resistant bacteria and susceptibility results are essential in order to guide treatment [SIGN, 2012].

Not sending urine for culture for uncomplicated UTIs

Urine microscopy and culture are not routinely recommended for women with suspected uncomplicated UTIs because the results are not available for immediate decision-making and, by the time they are available, most women's symptoms will be resolving. Three decision analyses found that, if urine were to be routinely cultured for all women with suspected UTI, the average duration of symptoms would be reduced by between 0.04 and 0.32 days [SIGN, 2012].

Similar evidence is provided by a randomized controlled trial that compared different strategies for antibiotic treatment and concluded that the optimum antibiotic treatment strategy for women with cystitis should take account of symptom severity and the woman's preferences [Little et al, 2009].

It is unlikely that routine urine culture is cost-effective. An economic analysis estimated the cost of preventing 1 day of symptoms as £215, and the cost per QALY (quality adjusted life year gained) was £215,000 [SIGN, 2012].

Empirical antibiotic choice

What antibiotic should I prescribe for a woman with a first presentation of a suspected urinary tract infection?

Offer an antibiotic to all women with a suspected urinary tract infection (UTI).

For a woman with mild symptoms who has normal immunity, normal renal function, and a normal renal tract, treatment can be delayed if she wishes to see if symptoms will resolve without treatment, especially if the urine dipstick test is negative for nitrites and leucocyte esterase (indicating a low probability of a UTI).

For all other women start treatment without delay.

Prescribe:

Trimethoprim  200 mg twice daily orally, or

Nitrofurantoin  50 mg four times daily orally, or 100 mg (modified-release) twice daily, or

Follow local guidelines that take into account local resistance patterns if these are available.

Prescribe a 5–10-day course of treatment for women who have:

Impaired renal function.

An abnormal urinary tract (for example renal calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, urinary obstruction, or recent instrumentation).

Immunosuppression (for example because they have poorly controlled diabetes mellitus or are receiving immunosuppressive treatment).

For all other women a 3-day course of antibiotics is sufficient.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Offering antibiotic treatment

When to prescribe

For women with urinary tract infections (UTIs), there is evidence from a meta-analysis that antibiotics are more effective than placebo in eradicating bacteriuria and relieving UTI symptoms [Falagas et al, 2009].

Considering the options of an antibiotic, no antibiotic, or delayed antibiotic prescription

The strategy for antibiotic prescribing is supported by evidence from a series of studies in the UK [Little et al, 2009], and a randomized controlled trial in New Zealand [Richards et al, 2005]. Empirical antibiotic choice:

No clinically (or economically) important differences were found between five different treatment strategies in which antibiotics were offered: (i) immediately, (ii) delayed for 48 hours, (iii) according to a symptom rule, (iv) according to a dipstick test rule, or (v) according to the results of urine culture.

Women who did not meet the criteria for immediate antibiotic treatment were offered a delayed antibiotic prescription to use if their symptoms did not settle after 48 hours. In each group where women were offered a delayed prescription, a high proportion chose to use it.

Women who presented with more severe symptoms of dysuria, urgency, frequency, and nocturia recovered more slowly.

Antibiotics shortened the duration of symptoms (that were at least moderately severe) by about 1–2 days.

CKS therefore recommends offering an antibiotic when:

Presenting symptoms are moderate or severe — because antibiotics are likely to shorten the duration of symptoms by 1–2 days.

The woman has a strong preference for antibiotic treatment — because there is no evidence that treatment leads to poorer outcomes, although there is also no evidence of effectiveness in women with less severe symptoms.

It may be a complicated infection — because there is a greater risk of adverse effects from infection.

What to prescribe as first-line options

Trimethoprim and nitrofurantoin (both narrow spectrum antibiotics) are generally recommended as appropriate first-line antibiotics in the UK [SIGN, 2012; BNF 66, 2013; HPA and British Infection Association, 2013]. Despite their widespread use, there are few comparative trials comparing these two antibiotics. There is evidence from four trials which found trimethoprim and nitrofurantoin to be equally effective and generally well tolerated.

Narrow spectrum antibiotics are preferred over broad spectrum antibiotics such as co-amoxiclav, quinolones, and cephalosporins. This is in line with guidance issued by the Scottish Intercollegiate Guidelines Network (SIGN) and the Health Protection Agency which recommends avoiding the use of broad spectrum antibiotics when narrow spectrum antibiotics remain effective [SIGN, 2012; HPA and British Infection Association, 2013]. There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs. Issues of antibiotic resistance are discussed below.

CKS recommend following local guidelines that take into account local resistance patterns because they are more likely to cure the UTI.

The dosages recommended are based on the British National Formulary and are in line with doses used in clinical trials [BNF 66, 2013].

The oral route is recommended, even for severe UTIs. A Cochrane systematic review found no evidence that oral antibiotic treatment is less effective than intravenous antibiotics for treating severe UTIs [Pohl, 2007].

Duration of treatment

For people with a complicated UTI, a longer course is recommended because there is evidence from a Cochrane systematic review that a 5–10-day course produced a higher bacteriological cure rate (but more adverse effects) than a 3-day regimen [Milo et al, 2005]. The Cochrane systematic review concluded that a 5–10-day course may be considered for women in whom eradication of bacteriuria is important.

A 3-day course of empirical treatment is recommended because there is good evidence from Cochrane systematic reviews that this achieves symptomatic cure in people with uncomplicated UTI; it is more effective than single-dose treatment and as effective as 5–10-day courses [Milo et al, 2005; Lutters and Vogt-Ferrier, 2008]. This is also in line with recommendations from SIGN [SIGN, 2012], the Health Protection Agency [HPA, 2011], and an international guideline [American College of Obstetricians and Gynecologists, 2008].

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of urinary tract infection, there is a lack of good evidence to support its use.

Follow up

What follow up should I arrange for a woman with a first presentation of a suspected urinary tract infection?

Advise all women to seek medical attention if they develop fever, loin pain, or do not respond to treatment with the first-choice antibiotic. If loin pain or fever develops see the CKS topic Pyelonephritis - acute.

Basis for recommendation

Basis for recommendation

Advising women to seek medical advice if they develop fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection in a woman with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up.

Advising women to seek medical advice if they do not respond to treatment with first-choice antibiotic

Non-response may be due to a resistant organism, and the woman may require further investigation and treatment.

Managing treatment failure

How should I manage a woman with a suspected urinary tract infection who has failed to respond to a first-choice antibiotic?

If loin pain or fever develops in association with a urinary tract infection (UTI) see the CKS topic Pyelonephritis - acute.

If symptoms of a UTI persist following treatment, send urine for culture and sensitivity and:

Adjust treatment as necessary when the results become available.

Seek specialist advice if a multi-resistant organism is cultured.

If there is no bacterial growth after culturing the urine, consider an alternative cause for symptoms.

If culture and sensitivity results are not available:

Choose either:

Trimethoprim 200 mg twice daily, if the first-choice antibiotic was nitrofurantoin, or

Nitrofurantoin 50 mg four times daily, or 100 mg (modified-release) twice daily if the first choice antibiotic was trimethoprim.

Prescribe a 5–10-day antibiotic course for women who have:

Impaired renal function.

An abnormal urinary tract (for example renal calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, urinary obstruction, or recent instrumentation).

Immunosuppression (for example because they have poorly controlled diabetes mellitus or are receiving immunosuppressive treatment).

For all other women, a 3 day course of antibiotics is sufficient.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Managing a woman who develops fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection (UTI) in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up. For more information, see the CKS topic Pyelonephritis - acute.

Managing persistent symptoms

Sending urine for culture and sensitivity if symptoms persist

Women whose treatment has failed are more likely to have a resistant organism, and it is important to send an MSU to identify sensitivities [HPA, 2011].

Seeking specialist advice if a multi-resistant organism is cultured

CKS recommend seeking expert advice to guide treatment when a multi-resistant organism has been cultured. This is in line with advice from the Health Protection Agency who has reported an increasing incidence of UTIs caused by community multi-resistant extended-spectrum Beta-lactamase Escherichia Coli that require expert advice to guide management [HPA and British Infection Association, 2013].

Considering other diagnoses if infection is not confirmed

Most women with cystitis will have a urinary tract infection. Failure to culture bacteria in the urine may be a false negative result, but the possibility of an alternative cause for symptoms is significantly increased and should be considered.

Advising a different antibiotic for persistent symptoms

These recommendations are in line with guidance from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012].

The recommendation to offer a different antibiotic if symptoms persist is supported by a study of the course of uncomplicated community-acquired urinary tract infection in women [McNulty et al, 2006]. The study found that, after 5 days of antibiotic treatment, symptoms had resolved in 70% of women infected with an organism sensitive to the antibiotic, but in only 24% of women with a resistant isolate. The study also found that 50% of those who reconsulted in the first week had a resistant isolate.

Duration of antibiotic treatment in women with a complicated urinary tract infection

For people with a complicated UTI, a longer course of antibiotics is recommended because there is evidence from a Cochrane systematic review that a 5–10-day course produced a higher bacteriological cure rate (but more adverse effects) than a 3-day regimen [Milo et al, 2005]. The Cochrane systematic review concluded that a 5–10-day course of antibiotics may be considered for women in whom eradication of bacteriuria is important.

Duration of antibiotic treatment in women with an uncomplicated urinary tract infection

A 3-day course of empirical treatment is recommended because there is good evidence from Cochrane systematic reviews that this achieves symptomatic cure in people with uncomplicated UTI; it is more effective than single-dose treatment and as effective as 5–10-day courses [Milo et al, 2005; Lutters and Vogt-Ferrier, 2008]. This is also in line with recommendations from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012], the Health Protection Agency [HPA, 2011], and an international guideline [American College of Obstetricians and Gynecologists, 2008].

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of urinary tract infection, there is a lack of good evidence to support its use.

Scenario: UTI — with visible or non-visible haematuria

Scenario: Managing any suspected urinary tract infection associated with visible or non-visible haematuria

168months3060monthsFemale

Urine for culture

When should I send urine for culture and sensitivity from women with a suspected urinary tract infection associated with haematuria?

Send urine for culture and sensitivity from all women with a suspected urinary tract infection associated with visible or non-visible haematuria.

Basis for recommendation

Basis for recommendation

It is important to exclude transient causes of visible or non-visible haematuria such as a urinary tract infection before investigating for another cause of haematuria such as urological cancer. This recommendation is based on expert advice in Referral guidelines for suspected cancer in adults and children published by the National Institute for Health and Care Excellence [NICE, 2005], and in a Joint Consensus Statement on the Initial Assessment of Haematuria published by the Renal Association and British Association of Urological Surgeons [Renal Association and British Association of Urological Surgeons, 2008].

Empirical antibiotic choice

What antibiotic should I prescribe empirically for women with suspected urinary tract infection associated with visible or non-visible haematuria?

For women with haematuria who are not pregnant:

Prescribe:

Trimethoprim  200 mg twice daily orally, or

Nitrofurantoin  50 mg four times daily orally, or 100 mg (modified-release) twice daily, or

Follow local prescribing guidelines that take into account local resistance patterns if these are available.

Prescribe a 5–10-day antibiotic course for women who have:

Impaired renal function.

An abnormal urinary tract (for example renal calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, urinary obstruction, or recent instrumentation).

Immunosuppression (for example because they have poorly controlled diabetes mellitus or are receiving immunosuppressive treatment).

For other women, a 3 day course of antibiotics is sufficient.

For women with haematuria who are pregnant follow local prescribing guidelines that take into account local resistance patterns if these are available. Otherwise prescribe (in order of preference):

Nitrofurantoin 50 mg four times daily, or 100 mg (modified-release) twice daily, for 7 days.

Trimethoprim 200 mg twice daily, for 7 days (off–label use).

Give folic acid 5 mg daily if it is the first trimester of pregnancy.

Do not give trimethoprim if the woman is folate deficient, taking a folate antagonist, or has been treated with trimethoprim in the past year.

Cefalexin 500 mg twice daily, or 250 mg 6-hourly, for 7 days.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Offering antibiotic treatment to non-pregnant women

When to prescribe

For women with urinary tract infections (UTIs), there is evidence from a meta-analysis that antibiotics are more effective than placebo in eradicating bacteriuria and relieving UTI symptoms [Falagas et al, 2009].

What to prescribe as first-line options

Trimethoprim and nitrofurantoin (both narrow spectrum antibiotics) are generally recommended as appropriate first-line antibiotics in the UK [SIGN, 2012; BNF 66, 2013; HPA and British Infection Association, 2013].

Narrow spectrum antibiotics are preferred over broad spectrum antibiotics such as co-amoxiclav, quinolones, and cephalosporins. This is in line with guidance issued by the Scottish Intercollegiate Guidelines Network (SIGN), and the Health Protection Agency which recommends avoiding the use of broad spectrum antibiotics when narrow spectrum antibiotics remain effective [SIGN, 2012; HPA and British Infection Association, 2013]. There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs. Issues of antibiotic resistance are discussed below.

Despite their widespread use, there are few comparative trials comparing these two antibiotics. There is evidence from four trials that found trimethoprim and nitrofurantoin to be equally effective and generally well tolerated.

CKS recommend following local guidelines that take into account local resistance patterns because the recommended antibiotic is likely to cure the UTI.

The dosages recommended are based on the British National Formulary and are in line with doses used in clinical trials [BNF 66, 2013].

The oral route is recommended, even for severe UTIs. A Cochrane systematic review found no evidence that oral antibiotic treatment is less effective than intravenous antibiotics for treating severe UTIs [Pohl, 2007].

Duration of treatment

For people with a complicated UTI, a longer course of antibiotics is recommended because there is evidence from a Cochrane systematic review that a 5 to 10-day course produced a higher bacteriological cure rate (but more adverse effects) than a 3-day regimen [Milo et al, 2005]. The Cochrane systematic review concluded that a 5 to 10-day course may be considered for women in whom eradication of bacteriuria is important.

A 3-day course of empirical treatment is recommended because there is good evidence from Cochrane systematic reviews that this achieves symptomatic cure in people with uncomplicated UTI; it is more effective than single-dose treatment and as effective as 5 to 10-day courses [Milo et al, 2005; Lutters and Vogt-Ferrier, 2008]. This is also in line with recommendations from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012], the Health Protection Agency [HPA, 2011], and an international guideline [American College of Obstetricians and Gynecologists, 2008].

Antibiotic treatment for pregnant women

Treatment with an antibiotic is recommended for pregnant women because there is good evidence from placebo-controlled trials in non-pregnant women with cystitis that antibiotics cure the infection, and experts suggest that urinary tract infection in pregnancy may increase the risk of fetal death, and increase the risks of developmental delay and cerebral palsy in the infant [Foxman, 2002].

Antibiotic choice for pregnant women

What to prescribe

Experts recommend that the choice of antibiotic for empirical treatment should take into account local rates of resistance in uropathogens [SIGN, 2012].

A Cochrane Systematic review (search date, November 2009) that aimed to determine which antibiotic was the most effective for treating symptomatic urinary tract infection in pregnant women found only ten small trials (1125 women), none of which were large enough to detect a 10% difference in cure rates [Vazquez and Abalos, 2011]. The authors concluded that there was insufficient data to recommend any specific antibiotic regimen for the treatment of symptomatic urinary tract infection during pregnancy.

For empirical treatment, nitrofurantoin is preferred over trimethoprim because:

Nitrofurantoin has been used extensively since the 1950s and its safety profile in human pregnancy has been well documented [ABPI Medicines Compendium, 2013b].

There is indirect evidence from a large multi-centre trial (undertaken by the World Health Organization [WHO], n = 778) supporting its efficacy and safety profiles for asymptomatic bacteriuria in pregnancy. For further information, see Asymptomatic bacteriuria in pregnancy.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4), compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small, the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester, although there is no evidence that this higher dose is any more beneficial than a 400 microgram daily dose [UKTIS, 2013].

Cefalexin is less preferred because although the safety profile is well documented in pregnancy, the Health Protection Agency recommends avoiding the use of broad spectrum antibiotics (such as cephalosporins) when narrow spectrum antibiotics remain effective [HPA and British Infection Association, 2013]. This is because:

There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs.

C. difficile infection can be life-threatening in pregnant women, and there are case-reports of both maternal deaths and stillborn infants [Rouphael et al, 2008].

Amoxicillin is not recommended for empirical treatment because:

There is evidence from several urine culture studies that resistance to amoxicillin is higher than for trimethoprim [Barrett et al, 1999; Magee et al, 1999; Howard et al, 2001; Priest et al, 2001; Wazait et al, 2003; Hay et al, 2005; Newell et al, 2005].

Duration of antibiotic treatment in pregnant women

Given the possible increased risk of fetal complications with a UTI, a 7-day course of antibiotics is preferred over shorter courses because:

For pregnant women with an acute uncomplicated UTI, a Cochrane systematic review and meta-analysis of 13 trials found that a 1-day course of antibiotics was less likely to produce a cure than courses of 4–7 days but the difference was not statistically significant [Widmer et al, 2011]. The authors concluded that a 7-day regimen may be considered for pregnant women when the risks of harm from a UTI are higher than in non-pregnant women.

A recent large WHO study which was included in the Cochrane review found a higher cure rate with a 7-day course of nitrofurantoin (86%) than a 1-day regimen (76%) in pregnant women with asymptomatic bacteriuria [Lumbiganon et al, 2009]. Adverse effects were not statistically different between the two groups.

A 7-day course is supported by guidance issued by the Health Protection Agency based on expert consensus [HPA and British Infection Association, 2013].

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of urinary tract infection, there is a lack of good evidence to support its use.

Follow up/referral

What follow up should I arrange for people with a suspected urinary tract infection associated with visible or non-visible haematuria?

Advise all women to seek medical attention if they develop fever, loin pain, or do not respond to treatment. If loin pain or fever develops see the CKS topic Pyelonephritis - acute.

Arrange follow up when the results of culture and sensitivity are available.

Adjust antibiotic treatment according to sensitivities, if there are persistent symptoms of a urinary tract infection following initial treatment.

For pregnant women, treatment options, in order of preference are:

Amoxicillin : 250 mg three times daily, for 7 days.

Nitrofurantoin : 50 mg four times daily, or 100 mg (modified-release) twice daily, for 7 days. Nitrofurantoin is contraindicated in women with an eGFR of less than 60 mL/min/1.73m2.

Trimethoprim : 200 mg twice daily, for 7 days (off–label use). Give folic acid 5 mg daily at the same time in the first trimester of pregnancy. Do not give trimethoprim if the woman is folate deficient, taking a folate antagonist, or has been treated with trimethoprim in the past year.

Cefalexin (500 mg twice daily, or 250 mg 6-hourly, for 7 days) may be used but is less preferred.

In pregnant women, repeat the urine culture seven days after finishing antibiotic treatment.

In pregnant women, if a group B streptococcus is isolated, inform the antenatal care service, as prophylactic antibiotics will be offered during labour and delivery.

For all women with visible or non-visible haematuria:

If infection has been confirmed re-test the urine for blood with a dipstick after completing treatment with an appropriate antibiotic, to detect persistent haematuria. Persistence is defined as two out of three dipsticks positive for blood on separate occasions.

Refer urgently for investigations of suspected urological cancer if:

Infection is not confirmed on culture.

Visible haematuria persists after infection has been successfully treated.

Non-visible haematuria persists after infection has been successfully treated in a person more than 50 years of age.

Visible or non-visible haematuria is associated with persistent or recurrent urinary tract infection in a woman aged 40 years or older.

If non-visible haematuria persists after infection has been successfully treated in a person less than 50 years of age, test the urine for proteinuria (albumin/creatinine ratio [ACR] equal to or higher than 30 mg/mmol, or protein/creatinine ratio [PCR] equal to or higher than 50 mg/mmol) and measure serum creatinine levels.

If there is proteinuria (raised ACR, or PCR) or a declining or low eGFR (<30 ml/min/1.73m2), refer to a renal physician.

If there is no proteinuria and eGFR is normal, refer non-urgently to a urologist.

Basis for recommendation

Basis for recommendation

Advising women to seek medical advice if they develop fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection (UTI) in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up.

Arranging follow up

A urinary tract infection may cause transient visible or non-visible haematuria. A consensus statement issued by experts of the Renal Association and British Association of Urological Surgeons recommends identifying and investigating people with persistent haematuria to identify and manage people with urological cancer or renal disease [Renal Association and British Association of Urological Surgeons, 2008].

Adjusting antibiotic treatment according to sensitivities

These recommendations are in line with guidance from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012].

The recommendation to offer a different antibiotic if symptoms persist is supported by a study of the course of uncomplicated community-acquired urinary tract infection in women [McNulty et al, 2006]. The study found that, after 5 days of antibiotic treatment, symptoms had resolved in 70% of women infected with an organism sensitive to the antibiotic, and 24% of women with a resistant isolate. The study also found that 50% of those who reconsulted in the first week had a resistant isolate.

Choice of antibiotics in pregnancy

Amoxicillin is recommended by the Health Protection Agency only if the organism is susceptible, as amoxicillin resistance is common [HPA and British Infection Association, 2013]. This recommendation is based on information from the UK Teratology Information Service who found no evidence to link amoxicillin with congenital malformations [UKTIS, 2012c].

Nitrofurantoin is preferred over trimethoprim because:

Nitrofurantoin has been used extensively since the 1950s and its safety profile in human pregnancy has been well documented [ABPI Medicines Compendium, 2013b].

There is indirect evidence from a large multi-centre trial (undertaken by the World Health Organization [WHO], n = 778) supporting its efficacy and safety profiles for asymptomatic bacteriuria in pregnancy. For further information, see Asymptomatic bacteriuria in pregnancy.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4) compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small, the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester, although there is no evidence that this higher dose is any more beneficial than a 400 microgram daily dose [UKTIS, 2013].

Cefalexin is less preferred because:

Although the safety profile is well documented in pregnancy, the Health Protection Agency recommends avoiding the use of broad spectrum antibiotics (such as cephalosporins) when narrow spectrum antibiotics remain effective because [HPA and British Infection Association, 2013]:

There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs.

C. difficile infection can be life-threatening in pregnant women, and there are case-reports of both maternal deaths and stillborn infants [Rouphael et al, 2008].

Following up to ensure eradication of infection

Symptomatic bacteriuria in pregnancy has been linked to pre-term labour, premature rupture of the membranes, and pyelonephritis. The expert opinion of the GDG of SIGN is that because of these risks, a urine should be cultured to confirm eradication of infection [SIGN, 2012].

Managing a group B streptococcus urinary tract infection

The antenatal care service should be informed when the group B streptococcus (GBS), Streptococcus agalactiae, is isolated from urine because this is associated with an increased risk of neonatal disease and chorioamnionitis. Experts recommend intrapartum antibiotics as well as appropriate antibiotics at the time of diagnosis [RCOG, 2012].

Managing women with visible or non-visible haematuria

Retesting urine for blood in women with visible or non-visible haematuria

This recommendation is based on expert advice in a joint consensus statement on the initial assessment of haematuria prepared on behalf or the Renal Association and the British Association of Urological Surgeons [Renal Association and British Association of Urological Surgeons, 2008], and in the NICE guideline Chronic Kidney Disease [NICE, 2008a].

Referring to exclude urological cancer

This recommendation is based on expert advice in Referral guidelines for suspected cancer in adults and children published by the National Institute for Health and Clinical Excellence [NICE, 2005].

Managing persistent non-visible haematuria in women aged less than 50 years of age

This recommendation is based on expert opinion in Referral guidelines for suspected cancer in adults and children published by the National Institute for Health and Clinical Excellence [NICE, 2005]. The definition of significant proteinuria is based on expert opinion in a Joint consensus statement on initial assessment of haematuria prepared on behalf of the Renal Association and British Association of Urological Surgeons [Renal Association and British Association of Urological Surgeons, 2008], and expert opinion in the NICE guideline on Chronic Kidney Disease [NICE, 2008a].

Scenario: Recurrent UTI — (no visible haematuria, not pregnant or catheterized)

Scenario: Suspected recurrent UTI without haematuria in women who are not catheterized or pregnant

168months3060monthsFemale

Urine for culture

Should I send urine for culture and sensitivity for women with recurrent urinary tract infections?

Send urine for culture and sensitivity from all women with recurrent urinary tract infections.

Basis for recommendation

Basis for recommendation

Sending an MSU for culture in women with recurrent urinary tract infections

This recommendation is based on expert advice from the Health Protection Agency because women with recurrent urinary tract infections are more likely to have a resistant strain [HPA, 2011].

Empirical antibiotic choice for recurrent UTI

What antibiotic should I prescribe empirically for recurrent urinary tract infections?

Offer an antibiotic to all women with recurrent urinary tract infections (UTIs).

Initiate treatment:

For a woman with mild symptoms who has normal immunity, normal renal function, and a normal renal tract, treatment can be delayed if she wishes to see if symptoms will resolve without treatment, especially if the probability of a UTI is low (indicated by a negative urine dipstick test for nitrites, and leucocyte esterase).

For all other women start treatment without delay.

Prescribe:

Nitrofurantoin  50 mg four times daily, or 100 mg (modified-release) twice daily, or

Trimethoprim  200 mg twice daily. If the woman has been treated with trimethoprim up to a year previously, consider prescribing nitrofurantoin instead of trimethoprim, or

Follow local guidelines that take into account local resistance patterns if these are available.

Prescribe a 5–10-day antibiotic course for women who have:

Impaired renal function.

An abnormal urinary tract (for example renal calculus, vesicoureteric reflux, reflux nephropathy, neurogenic bladder, urinary obstruction, or recent instrumentation).

Immunosuppression (for example because they have poorly controlled diabetes mellitus or are receiving immunosuppressive treatment).

For all other women, a 3-day course of antibiotics is sufficient.

Consider offering a prescription for a 'stand-by' antibiotic to be used for future episodes of suspected UTIs if episodes are frequent and renal tract abnormalities such as renal calculi have been excluded following investigations by a specialist.

Advise the woman to collect and submit a mid-stream sample of urine before starting the treatment, and to seek medical advice if symptoms have not resolved within 48 hours.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Antibiotics for recurrent urinary tract infection

When to prescribe

For women with urinary tract infections (UTI), there is evidence from a meta-analysis that antibiotics are more effective than placebo in eradicating bacteriuria and relieving UTI symptoms [Falagas et al, 2009].

Considering the options of an antibiotic, no antibiotic, or delayed antibiotic prescription

The strategy for antibiotic prescribing is supported by evidence from a series of studies in the UK [Little et al, 2009], and a randomized controlled trial in New Zealand [Richards et al, 2005]. Empirical antibiotic choice:

No clinically (or economically) important differences were found between five different treatment strategies in which antibiotics were offered: (i) immediately, (ii) delayed for 48 hours, (iii) according to a symptom rule, (iv) according to a dipstick test rule, or (v) according to the results of urine culture.

Women who did not meet the criteria for immediate antibiotic treatment were offered a delayed antibiotic prescription to use if their symptoms did not settle after 48 hours. In each group where women were offered a delayed prescription, a high proportion chose to use it.

Women who presented with more severe symptoms of dysuria, urgency, frequency, and nocturia recovered more slowly.

Antibiotics shortened the duration of symptoms (that were at least moderately severe) by about 1–2 days.

CKS therefore recommends offering an antibiotic when:

Presenting symptoms are moderate or severe — because antibiotics are likely to shorten the duration of symptoms by 1–2 days.

The woman has a strong preference for antibiotic treatment — because there is no evidence that treatment leads to poorer outcomes, although there is also no evidence of effectiveness in women with less severe symptoms.

It may be a complicated infection — because there is a greater risk of adverse effects from infection.

What to prescribe

The basis for the recommended treatments is discussed in empirical antibiotic choice.

Considering nitrofurantoin when trimethoprim has been used recently

Nitrofurantoin may be preferable for empirical prescribing when the woman has recently used trimethoprim because there is evidence that uropathogens are more likely to be resistant to trimethoprim if it has been used recently (up to the past year).

Trimethoprim is not preferred when nitrofurantoin has previously been used because there is no evidence that previous treatment with nitrofurantoin increases the chance that future infections will be resistant organisms. Furthermore, laboratory studies find that nitrofurantoin-resistant Escherichia coli reproduce substantially less effectively than nitrofurantoin-sensitive E. coli (in other words, nitrofurantoin resistance imposes a high fitness cost on the organism) [Sandegren et al, 2008].

Duration of antibiotic treatment

For people with a complicated UTI, a longer course is recommended because there is evidence from a Cochrane systematic review that a 5 to 10-day course produced a higher bacteriological cure rate (but more adverse effects) than a 3-day regimen [Milo et al, 2005]. The Cochrane systematic review concluded that a 5 to 10-day course may be considered for women in whom eradication of bacteriuria is important.

A 3-day course of empirical treatment is recommended because there is good evidence from Cochrane systematic reviews that this achieves symptomatic cure in people with uncomplicated UTI; it is more effective than single-dose treatment and as effective as 5 to 10-day courses [Milo et al, 2005; Lutters and Vogt-Ferrier, 2008]. This is also in line with recommendations from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012], the Health Protection Agency [HPA, 2011], and an international guideline [American College of Obstetricians and Gynecologists, 2008].

'Stand-by' antibiotics as an alternative to preventive antibiotics

The recommendation to consider 'stand-by' antibiotics and collect a mid-stream sample of urine is based on expert opinion in a review article [Harris et al, 2008], and guidelines from the Society of Obstetricians and Gynaecologists of Canada [SOGC, 2010].

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of a urinary tract infection, there is a lack of good evidence to support its use.

Follow up

How should I follow up a woman with a suspected recurrent urinary tract infection?

Advise women to seek medical attention if they develop fever, loin pain, or do not respond to treatment. If loin pain or fever develops in association with a urinary tract infection (UTI), see the CKS topic on Pyelonephritis - acute.

If symptoms of a UTI persist following treatment:

Adjust treatment, if necessary, when the results of urine culture and sensitivity testing become available.

If there is no bacterial growth after culturing the urine consider an alternative cause for symptoms.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Advising women to seek medical advice if they develop fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection (UTI) in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up. For more information, see the CKS topic Pyelonephritis - acute.

Managing persistent symptoms of a urinary tract infection.

Advising a different antibiotic for persistent symptoms

Non-response may be due to a resistant organism, and the woman may require further investigation and treatment.

These recommendations are in line with guidance from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012].

The recommendation to offer a different antibiotic if symptoms persist is supported by a study of the course of uncomplicated community-acquired urinary tract infection in women [McNulty et al, 2006]. The study found that, after 5 days of antibiotic treatment, symptoms had resolved in 70% of women infected with an organism sensitive to the antibiotic, and 24% of women with a resistant isolate. The study also found that 50% of those who reconsulted in the first week had a resistant isolate.

Considering other diagnoses if infection is not confirmed

Most women with cystitis will have a urinary tract infection. Failure to culture bacteria in the urine may be a false negative result, but the possibility of an alternative cause for symptoms is significantly increased and should be considered.

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of urinary tract infection, there is a lack of good evidence to support its use.

When to refer

When should I refer a woman with recurrent urinary tract infections for assessment of the cause?

Urgent referral is recommended for women with recurrent urinary tract infections (UTIs) associated with haematuria (visible or non-visible) for investigations to exclude urological cancer.

Routine referral is recommended for women with recurrent UTIs:

Who have a risk factor for an abnormality of the urinary tract including women with:

A past history of urinary tract surgery or trauma.

A past history of bladder or renal calculi.

Obstructive symptoms such as straining, hesitancy, poor stream.

Urea splitting bacteria on culture of the urine such as Proteus or Yersinia.

Persistent bacteriuria despite appropriate antibiotic treatment.

A past history of abdominal or pelvic malignancy.

Symptoms of a fistula such as pneumaturia.

Who are immunocompromised or who have diabetes.

Who have a known abnormality of their renal tract who might benefit from surgical correction, such as cystocele, vesicoureteric reflux, or bladder outlet obstruction.

Who have not responded to preventive treatments.

Basis for recommendation

Basis for recommendation

Urgent referral

The recommendation to refer women with suspected urological cancer is based on criteria in guidelines from the National Institute for Health and Clinical Excellence (NICE) [National Collaborating Centre for Primary Care, 2005].

Routine referral

Referring women with a risk factors for an abnormality of the urinary tract

The recommendation to refer women with a risk factor for an underlying abnormality of the urinary tract is based on expert opinion in Guidelines for the diagnosis and management of recurrent urinary tract infection in women published by the Canadian Urological Association [Dason et al, 2011].

Referring woman who are immunocompromised or who have diabetes

The recommendation to refer for specialist assessment of risk factors for complicated UTI is based on expert opinion in Guidelines for the diagnosis and management of recurrent urinary tract infection in women published by the Canadian Urological Association [Dason et al, 2011].

Referring women who have responded poorly to preventive measures

The recommendation to seek specialist opinion when preventive measures have failed is in line with recommendations issued by the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012].

Preventing recurrent UTIs

How can recurrent urinary tract infections be prevented?

For women with recurrent urinary tract infections who do not have an indication for referral for specialist assessment:

Consider prophylactic drug treatment when there is unacceptable discomfort or disruption to their lives.

For recurrent cystitis associated with sexual intercourse, offer Trimethoprim 100 mg to be taken within 2 hours of intercourse (off-label use).

For recurrent cystitis not associated with sexual intercourse:

Offer a 6-month trial of low-dose continuous antibiotic treatment.

Suitable antibiotics are Trimethoprim 100 mg every night or Nitrofurantoin (immediate-release) 50 mg to 100 mg every night (modified-release nitrofurantoin is not licensed for prophylaxis).

Advise all women with recurrent cystitis that:

Cranberry juice and products are no longer recommended for the prevention of recurrent cystitis.

If cystitis is related to sexual intercourse, they should consider using a different contraceptive method if a diaphragm is being used.

There is no evidence that voiding before or after intercourse, hydration or changes to diet help prevent recurrent urinary tract infections.

Basis for recommendation

Basis for recommendation

Prophylactic drug treatment

Post-coital antibiotics

There is evidence from two trials included in a Cochrane systematic review that post-coital antibiotics may be more effective than placebo and as effective as continuous antibiotic treatment in preventing urinary tract infection (UTI) associated with sexual intercourse.

The recommendation to offer trimethoprim 100 mg is extrapolated from evidence provided by a small, double-blind, randomized controlled trial that co-trimoxazole 240 mg (containing trimethoprim 40 mg and sulfamethoxazole 200 mg) given within 2 hours of intercourse was more effective than a post-coital placebo [Stapleton et al, 1990]. Trimethoprim (a narrow spectrum antibiotic) has been found to be as effective as co-trimoxazole in treating UTI and produces fewer adverse effects [SIGN, 2012; BNF 66, 2013; Grabe et al, 2013]. The European Association of Urology also recommend that it is reasonable to offer the doses of antibiotics used for nightly prophylaxis for post-coital use [Grabe et al, 2013].

Although another study supported the post-coital use of ciprofloxacin (a quinolone) [Melekos et al, 1997], broad spectrum antibiotics are less preferred. This is in line with guidance issued by the Health Protection Agency which recommended avoiding the use of broad spectrum antibiotics when narrow spectrum antibiotics remain effective [HPA and British Infection Association, 2013]. There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs.

Continuous antibiotic prophylaxis

Effectiveness compared with placebo. There is weak evidence from a Cochrane systematic review that continuous antibiotics reduce recurrence of urinary tract infections more than placebo but are associated with more adverse effects [Albert et al, 2004]. There are limited data showing that antibiotics do not continue to prevent bacteriuria after treatment is stopped.

Choice of antibiotic. There is insufficient direct evidence to prefer any particular antibiotic over another. Trimethoprim and nitrofurantoin are recommended options for prophylaxis of recurrent UTI because:

Narrow spectrum antibiotics, such as trimethoprim and nitrofurantoin, are generally preferred over broad spectrum antibiotics due to concerns over increasing the risk of Clostridium difficile, MRSA, and resistant UTIs [HPA and British Infection Association, 2013].

Both trimethoprim and nitrofurantoin are licensed for the prophylaxis of recurrent UTI [ABPI Medicines Compendium, 2012d; ABPI Medicines Compendium, 2013b; ABPI Medicines Compendium, 2013c; ABPI Medicines Compendium, 2013d].

Dosage

The dosages are based on those recommended by the British National Formulary and in line with the dosages used in trials investigating prophylaxis of recurrent UTIs [Albert et al, 2004; BNF 66, 2013].

Duration of treatment

A 6-month trial is advised, as this is recommended in guidelines from the Canadian Urological society [Dason et al, 2011]. Information on long-term follow up is lacking.

Bacterial resistance

There are concerns that resistance to trimethoprim and nitrofurantoin is increasing, but there is little current published evidence on resistance patterns for trimethoprim and nitrofurantoin.

Consequently, CKS recommends that local antibiotic guidelines should be followed, taking into account local resistance patterns.

Giving advice

Not recommending cranberry products

Current evidence does not support the use of cranberry products. An updated Cochrane systematic review found no evidence that cranberry juice and products decrease the incidence of symptomatic urinary tract infections (UTIs), particularly in people with a history of lower UTIs or women with recurrent UTIs [Jepson et al, 2012]. However, the lack of a standardized cranberry product in these trials makes comparison difficult. The trials were not of sufficient quality to give a clear indication of whether cranberry products are effective or not. High drop-out rates were found in some trials due to the difficulty in drinking large volumes of cranberry juice daily.

Preventing recurrent cystitis related to sexual intercourse

These recommendations are based on expert opinion in a clinical review [Harris et al, 2008].

The incidence of UTI may be increased in women who use diaphragms — this may relate to the fit and size of the diaphragm putting pressure on the urethra. The incidence of UTI may also be increased in women who use spermicides, but the use of spermicides with condoms is no longer recommended. For further information, see the CKS topic on Contraception - barrier methods and spermicides.

Not giving lifestyle advice

The Scottish Intercollegiate Guidelines Network (SIGN) guidelines explain that because 'there is no conclusive association between lifestyle factors, such as diet, hydration, clothing, toileting activity, and sexual activity, and susceptibility to bacterial UTI in adult, non-pregnant women, there is no evidence to support healthcare professionals giving routine advice about lifestyle factors' [SIGN, 2012].

Voiding before or after intercourse is often advised but expert opinion in Guidelines for the diagnosis and management of recurrent urinary tract infection in women published Canadian Urological Association is that there is no evidence for this practice although it is unlikely to be harmful [Dason et al, 2011].

Treatments not recommended

Methenamine hippurate

Methenamine hippurate is not recommended for preventing UTI because there is only weak evidence from a Cochrane systematic review that treatment may be effective for up to 7 days [Lee et al, 2012].

Oestrogen products (for postmenopausal women)

Oestrogen products are not recommended for use as preventive treatment in primary care because there is evidence from a Cochrane systematic review that oral oestrogens are no more effective than placebo in reducing recurrent UTIs in postmenopausal women, and there is conflicting evidence from two small trials on intravaginal oestrogen [Perrotta et al, 2008].

Scenario: UTI in pregnancy — no visible haematuria

Scenario: Suspected urinary tract infection without visible haematuria during pregnancy

168months3060monthsFemale

Urine for culture

When should I send urine for culture and sensitivity from pregnant women with a suspected urinary tract infection?

Send urine for culture and sensitivity from all women with a suspected urinary tract infection during pregnancy:

Before empirical antibiotic treatment is started.

Seven days after antibiotic treatment had been completed as a test of cure.

Basis for recommendation

Basis for recommendation

When to send urine for culture

Sending urine for culture before starting empirical antibiotic treatment

Sending urine for culture is recommended by experts of the Health Protection Agency because the results can confirm the diagnosis and guide further treatment, especially if the uropathogen turns out to be resistant to the empirically chosen antibiotic [HPA and British Infection Association, 2013].

Sending urine for culture as a test of cure

Prompt and effective treatment of urinary tract infections (UTIs) is important during pregnancy because there is a greater risk of complications from UTIs (such as pyelonephritis and premature labour). Expert opinion in guidelines from the Scottish Intercollegiate Guidelines Network is that urine should be cultured to test for cure [SIGN, 2012].

Empirical antibiotic choice for pregnant women

What antibiotic should I prescribe empirically for a woman with a suspected urinary tract infection during pregnancy?

Prescribe an antibiotic to all women with a suspected urinary tract infection during pregnancy. Follow local prescribing guidelines that take into account local resistance patterns if these are available. Otherwise prescribe (in order of preference):

Nitrofurantoin 50 mg four times daily, or 100 mg (modified-release) twice daily, for 7 days.

Trimethoprim 200 mg twice daily, for 7 days (off–label use).

Give folic acid 5 mg daily if it is the first trimester of pregnancy.

Do not give trimethoprim if the woman is folate deficient, taking a folate antagonist, or has been treated with trimethoprim in the past year.

Cefalexin 500 mg twice daily, or 250 mg 6-hourly, for 7 days.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Treating infection with an antibiotic

Treatment with an antibiotic is recommended because there is good evidence from placebo-controlled trials in non-pregnant women with cystitis that antibiotics cure the infection, and experts suggest that urinary tract infection (UTI) in pregnancy may increase the risk of fetal death, and increase the risks of developmental delay and cerebral palsy in the infant [Foxman, 2002].

CKS did not recommend that women who have mild symptoms should be offered the option of waiting for the urine culture results before starting antibiotic treatment, although this option is recommended for women who are not pregnant. CKS made no recommendation because no evidence and no published expert opinion was found on this strategy.

Choosing antibiotics for empirical treatment in pregnant women

What to prescribe

Experts recommend that the choice of antibiotic for empirical treatment should take into account local rates of resistance in uropathogens [SIGN, 2012].

A Cochrane systematic review (search date, November 2009) that aimed to determine which antibiotic was the most effective for treating symptomatic urinary tract infection in pregnant women found only ten small trials (1125 women), none of which were large enough to detect a 10% difference in cure rates [Vazquez and Abalos, 2011]. The authors concluded that there was insufficient data to recommend any specific antibiotic regimen for the treatment of symptomatic urinary tract infection during pregnancy.

For empirical treatment, nitrofurantoin is preferred over trimethoprim because:

Nitrofurantoin has been used extensively since the 1950s and its safety profile in human pregnancy has been well documented [ABPI Medicines Compendium, 2013b].

There is indirect evidence from a large multicentre trial (undertaken by the World Health Organization [WHO], n = 778) supporting its efficacy and safety profiles for asymptomatic bacteriuria in pregnancy [Lumbiganon et al, 2009]. For further information, see Asymptomatic bacteriuria in pregnancy.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4), compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small, the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester, although there is no evidence that this higher dose is any more beneficial than a 400 microgram daily dose [UKTIS, 2013].

Cefalexin is less preferred because although the safety profile is well documented in pregnancy, the Health Protection Agency recommends avoiding the use of broad spectrum antibiotics (such as cephalosporins) when narrow spectrum antibiotics remain effective [HPA and British Infection Association, 2013]. This is because:

There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs.

C. difficile infection can be life-threatening in pregnant women, and there are case-reports of both maternal deaths and stillborn infants [Rouphael et al, 2008].

Amoxicillin is not recommended for empirical treatment because:

There is evidence from several urine culture studies that resistance to amoxicillin is higher than for trimethoprim [Barrett et al, 1999; Magee et al, 1999; Howard et al, 2001; Priest et al, 2001; Wazait et al, 2003; Hay et al, 2005; Newell et al, 2005].

Duration of antibiotic treatment in pregnant women

Evidence on different antibiotic regimens for treating symptomatic UTIs in pregnant women is lacking.

Given the possible increased risk of fetal complications with a UTI, a 7-day course of antibiotics is preferred over shorter courses because:

For pregnant women with an acute uncomplicated UTI, a Cochrane systematic review and meta-analysis of 13 trials found that a 1-day course of antibiotics was less likely to produce a cure than courses of 4–7 days but the difference was not statistically significant [Widmer et al, 2011]. The authors concluded that a 7-day regimen should be considered for pregnant women because the risk of harm from a UTI is higher than for non-pregnant woman.

A recent large WHO study which was included in the Cochrane review found a higher cure rate with a 7-day course of nitrofurantoin (86%) than a 1-day regimen (76%) in pregnant women with asymptomatic bacteriuria [Lumbiganon et al, 2009]. Adverse effects were not statistically different between the two groups.

A 7-day course is supported by guidance issued by the Health Protection Agency based on expert consensus [HPA and British Infection Association, 2013].

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of urinary tract infection, there is a lack of good evidence to support its use.

Follow up

What follow up should I arrange for a woman with a suspected urinary tract infection during pregnancy?

Advise women to seek medical attention if they develop fever, loin pain, or do not respond to treatment. If loin pain or fever develops in association with a urinary tract infection (UTI), see the CKS topic Pyelonephritis - acute.

If symptoms of UTI persist when sensitivities are known, treatment options are (in order of preference):

Amoxicillin : 250 mg three times daily, for 7 days.

Nitrofurantoin : 50 mg four times daily, or 100 mg (modified-release) twice daily, for 7 days.

Trimethoprim : 200 mg twice daily, for 7 days (off–label use).

Give folic acid 5 mg daily if it is the first trimester of pregnancy.

Do not give trimethoprim if the woman is folate deficient, taking a folate antagonist, or has been treated with trimethoprim in the past year.

Cefalexin (500 mg twice daily, or 250 mg 6-hourly, for 7 days) may be used but is less preferred.

Repeat the urine culture seven days after finishing antibiotic treatment.

If a group B streptococcus is isolated, inform the antenatal care service, as prophylactic antibiotics will be offered during labour and delivery.

If there is no bacterial growth after culturing the urine, consider an alternative cause for symptoms.

Basis for recommendation

Basis for recommendation

Advising women to seek medical advice if they develop fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection (UTI) in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up. For more information, see the CKS topic Pyelonephritis - acute.

Choice of antibiotics in pregnancy

Amoxicillin is recommended by the Health Protection Agency only if the organism is susceptible, as amoxicillin resistance is common [HPA and British Infection Association, 2013]. This recommendation is based on information from the UK Teratology Information Service who found no evidence to link amoxicillin with congenital malformations [UKTIS, 2012c].

Nitrofurantoin is preferred over trimethoprim because:

Nitrofurantoin has been used extensively since the 1950s and its safety profile in human pregnancy has been well documented [ABPI Medicines Compendium, 2013b].

There is indirect evidence from a large multicentre trial (undertaken by the World Health Organization [WHO], n = 778) supporting its efficacy and safety profiles for asymptomatic bacteriuria in pregnancy. For further information, see Asymptomatic bacteriuria in pregnancy.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4), compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small, the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester, although there is no evidence that this higher dose is any more beneficial than a 400 microgram daily dose [UKTIS, 2013].

Cefalexin is less preferred because:

Although the safety profile is well documented in pregnancy, the Health Protection Agency recommends avoiding the use of broad spectrum antibiotics (such as cephalosporins) when narrow spectrum antibiotics remain effective because [HPA and British Infection Association, 2013]:

There are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs.

C. difficile infection can be life-threatening in pregnant women, and there are case-reports of both maternal deaths and stillborn infants [Rouphael et al, 2008].

Following up to ensure eradication of infection

Symptomatic bacteriuria in pregnancy has been linked to pre-term labour, premature rupture of the membranes, and pyelonephritis. The expert opinion of the GDG of SIGN is that because of these risks, urine should be cultured to confirm eradication of infection [SIGN, 2012].

Managing a group B streptococcus urinary tract infection

The antenatal care service should be informed when the group B streptococcus (GBS), Streptococcus agalactiae, is isolated from urine because this is associated with an increased risk of neonatal disease and chorioamnionitis. Experts recommend intrapartum antibiotics as well as appropriate antibiotics at the time of diagnosis [RCOG, 2012].

Considering other diagnoses if infection is not confirmed

Most women with cystitis will have a urinary tract infection. Failure to culture bacteria in the urine may be a false negative result, but the possibility of an alternative cause for symptoms is significantly increased and should be considered.

When to refer

When should I refer a pregnant woman with a urinary tract infection for specialist management?

Admit pregnant women with suspected pyelonephritis. For further information see the CKS topic Pyelonephritis - acute.

Seek specialist advice if a pregnant woman:

Has symptoms that fail to respond to appropriate antibiotic treatment guided by urine culture results.

Has recurrent urinary tract infections.

Basis for recommendation

Basis for recommendation

Admitting pregnant women with suspected pyelonephritis

Experts recommend arranging admission for all pregnant women with acute pyelonephritis, for at least a short observation period, because there is a risk of preterm labour and maternal renal complications [Ramakrishnan and Scheid, 2005].

Seeking specialist advice for persistent or recurrent urinary tract infections

CKS found no relevant clinical trials or advice in national guidelines to inform the management of pregnant women with recurrent urinary tract infections (UTIs). The Guideline Development Group of the Scottish Intercollegiate Guideline Network stress the risk of complications of UTIs in pregnancy including [SIGN, 2012]:

Preterm labour and premature rupture of the membranes.

Pyelonephritis. Between 10 and 30% or women who have bacteriuria during the first trimester develop pyelonephritis in the second or third trimester, which has been associated with an increase in morbidity and rarely mortality.

CKS therefore considers it important to ensure eradication of bacteriuria in pregnant women to avoid the risks and suggests referral to a specialist if a woman fails to responds to treatment or has recurrent infections.

Scenario: Asymptomatic bacteriuria in pregnancy

Scenario: Asymptomatic bacteriuria in pregnancy

168months3060monthsFemale

Asymptomatic bacteriuria in pregnancy

How should I screen for and manage asymptomatic bacteriuria during pregnancy?

Screen for asymptomatic bacteriuria on the first antenatal visit by sending urine for culture. If asymptomatic bacteriuria is found, send a second urine sample for culture.

If the second urine culture confirms asymptomatic bacteriuria, treat for 7 days with an antibiotic to which the organism is sensitive.

Refer to local guidelines for advice on the choice of antibiotics.

Options when sensitivities are known are (in order of preference):

Amoxicillin : 250 mg three times daily, for 7 days.

Nitrofurantoin : 50 mg four times daily, or 100 mg (modified-release) twice daily, for 7 days.

Trimethoprim : 200 mg twice daily, for 7 days (off–label use).

Give folic acid 5 mg daily if it is the first trimester of pregnancy.

Do not give trimethoprim if the woman is folate deficient, taking a folate antagonist, or has been treated with trimethoprim in the past year.

Cefalexin (500 mg twice daily, or 250 mg 6-hourly, for 7 days) may be used but is less preferred.

Send urine for culture at every antenatal visit until delivery to screen for asymptomatic bacteriuria.

If a group B streptococcus is isolated, inform the antenatal care service, as prophylactic antibiotics will be offered during labour and delivery.

Basis for recommendation

Basis for recommendation

Screening for asymptomatic bacteriuria

Guidelines from the National Institute for Health and Care Excellence (NICE) on antenatal care recommend that 'Women should be offered routine screening for asymptomatic bacteriuria by midstream urine culture early in pregnancy' [NICE, 2008b]. Guidelines from the Scottish Intercollegiate Guidelines Network (SIGN) are clearer: 'Women who do not have bacteriuria in the first trimester should not have repeat urine cultures' [SIGN, 2012].

Culture of urine is recommended rather than dipstick (reagent strip) tests because there is good evidence that dipstick tests are insufficiently sensitive (in whatever combination) to be used for screening. Urine culture is regarded as the gold standard and is assumed to have (close to) 100% sensitivity for detecting bacteriuria.

Confirming asymptomatic bacteriuria

SIGN recommend confirming bacteriuria with a second urine culture because there is evidence (from a cohort study that included 4917 women who were less than 25 weeks pregnant) that found that bacteriuria was confirmed in only 57% of women with an initial positive culture [Gratacós et al, 1994; SIGN, 2012].

Treating asymptomatic bacteriuria

Asymptomatic bacteriuria in pregnancy should be treated with an antibiotic because there is consistent evidence from a Cochrane systematic review that the risk of pyelonephritis is reduced: about seven women need to be treated to prevent one episode of pyelonephritis.

There is inconsistent evidence that treatment may also reduce the incidence of low birthweight and prematurity.

Referring to local guidelines

This recommendation is based on expert opinion from SIGN [SIGN, 2012].

Options where sensitivities are known

What to prescribe

Amoxicillin is preferred because:

The manufacturer of amoxicillin states that its use in pregnancy has been well documented in clinical studies [ABPI Medicines Compendium, 2012a]. It can be given at any stage of pregnancy and has not been associated with congenital malformations or any other adverse effects on the foetus [UKTIS, 2012c].

Nitrofurantoin as an alternative if amoxicillin is not suitable.

Nitrofurantoin has been used extensively since the 1950s, and its safety profile in human pregnancy has been well documented.

There is evidence from a Cochrane systematic review which supports the use of nitrofurantoin for treating asymptomatic bacteriuria in pregnancy [Smaill and Vazquez, 2007]. Nitrofurantoin was studied in five of the 14 studies identified (none on trimethoprim). Although significant heterogeneity was present, pooled results from five trials (two used nitrofurantoin) found antibiotics to be more effective than placebo in treating asymptomatic bacteriuria in pregnancy.

The efficacy and safety profiles of nitrofurantoin are further supported in a large multicentre study undertaken by the World Health Organization (WHO) in which 778 pregnant women with asymptomatic bacteriuria were treated with nitrofurantoin [Lumbiganon et al, 2009]. Adverse effects were not statistically different between the two groups.

Trimethoprim, used carefully, has a good safety profile during pregnancy.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4), compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small, the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester, although there is no evidence that this higher dose is any more beneficial than a 400 microgram daily dose [UKTIS, 2013].

For further information, see the section on Advice on folic acid in the CKS topic on Pre-conception - advice and management.

Cefalexin is less preferred because:

Although cefalexin can be used in pregnancy [Schaefer et al, 2007], the Health Protection Agency recommends avoiding the use of broad spectrum antibiotics (such as co-amoxiclav, cephalosporins, and quinolones) when narrow spectrum antibiotics remain effective because there are concerns that broad spectrum antibiotics increase the risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs [HPA and British Infection Association, 2013].

Duration of antibiotic treatment

The Guideline Development Group of the Scottish Intercollegiate Guidelines Network recommend a 7-day course based on [SIGN, 2012]:

Standard care in Scotland, where asymptomatic bacteriuria is treated with a 3–7 day course of antibiotics.

A Cochrane systematic review that included 13 trials (most of poor methodological quality in terms of design, conduct, and analysis of results) found that a 1-day course of antibiotics was less likely to produce a cure than courses of 4–7 days but the difference was not statistically significant [Widmer et al, 2011].

A Cochrane systematic review included 4 trials that used a 3–7 day course of antibiotic to treat women with asymptomatic bacteriuria [Smaill and Vazquez, 2007]. One small trial (n = 62) in this review found that a 3–7 day course had a significantly positive effect in preventing persistent bacteriuria, and a meta-analysis of the four trials found a beneficial but non-significant effect on the development of pyelonephritis.

The use of a 7-day course is supported by evidence from a well-conducted large double-blind randomized controlled trial (carried out for the World Health Organization) that found that a 7-day regimen of nitrofurantoin was more effective than a 1-day regimen in eradicating bacteriuria [Lumbiganon et al, 2009]. This multicentre trial, in Thailand, the Philippines, Vietnam, and Argentina, studied 778 women with asymptomatic nitrofurantoin-sensitive bacteriuria. Participants were randomized to receive nitrofurantoin 100 mg twice daily for 1 or 7 days. Women taking nitrofurantoin for 1 day took identical-looking placebo capsules for the following 6 days.

Eradication of bacteriuria. After 14 days, the 1-day regimen was significantly less effective than the 7-day regimen in eradicating bacteriuria (76% compared with 86%, cure rate ratio 0.88, 95% CI 0.82 to 0.94).

CKS recommend a 7-day course until data comparing 7-day regimens with 3- or 5-day regimens becomes available.

Antibiotic dosages

These are in line with dosages recommended by the manufacturers of these antibiotics.

Sending urine for culture at every antenatal visit

The recommendation to continue to screen for asymptomatic bacteriuria at each subsequent visit after completing antibiotic treatment is consistent with practice in multinational clinical trials, such as that conducted by the World Health Organization's Asymptomatic Bacteriuria Trial Group [Lumbiganon et al, 2009].

The Guideline Development Group of SIGN also recommend regular testing for asymptomatic bacteriuria based on evidence from an observational study that treatment of asymptomatic bacteriuria modifies the risk of developing pyelonephritis [Gratacós et al, 1994; SIGN, 2012].

Managing incidentally-found group B streptococcus infection

Experts recommend that the antenatal care service should be informed when a group B streptococcus (GBS), Streptococcus agalactiae, is isolated in urine because GBS bacteriuria, even if treated, may be associated with increased risk of neonatal GBS disease requiring antibiotic prophylaxis during delivery to reduce this risk [RCOG, 2012].

Scenario: UTI with catheter — no haematuria

Scenario: Suspected urinary tract infection without haematuria in women with an indwelling urinary catheter

168months3060monthsFemale

Changing the catheter

Should I change the catheter in a catheterized woman with a suspected urinary tract infection?

Change the catheter before starting antibiotics, if the catheter has been in place for longer than 7 days.

Basis for recommendation

Basis for recommendation

Changing the catheter

There is evidence from one small trial that changing the catheter before treatment improves the chance of cure [Raz et al, 2000]. This trial aimed to investigate whether replacing indwelling urinary catheters before starting antibiotics for urinary tract infection significantly improves cure rates. People who had had an indwelling catheter in situ for a mean of 30.8 days (range 17-35 days) were included in the trial and were randomised to either have their catheter replaced or not replaced [Raz et al, 2000].

Urine culture 72 hours after starting treatment was significantly more likely to find no uropathogens if the catheter had been replaced. The Infectious Diseases Society of America considered that the evidence from this trial supported changing the catheter if it had been in place for at least 2 weeks and could not be removed permanently [Hooton et al, 2010].

It is likely that bacteria will be sequestered in a biofilm on the catheter surface, and expert opinion in guidelines from the European Association of Urology [Grabe et al, 2013], and the European and Asian guidelines [Tenke et al, 2008] on management and prevention of catheter-associated urinary tract infections is that is reasonable to replace or remove the catheter before starting an antibiotic if the indwelling catheter has been in place for more than 7 days [Grabe et al, 2013]. A biofilm develops on the surface of a catheter 1-3 days after catheterization [Hooton et al, 2010].

The recommendation about when to change the catheter is based on expert opinion.

Urine for culture

When should I send urine for culture and sensitivity from catheterized women with a suspected urinary tract infection?

Send urine for culture and sensitivity from all catheterized women with a suspected urinary tract infection.

If it is to be changed, obtain the sample from the new catheter.

Basis for recommendation

Basis for recommendation

Sending urine for culture

The recommendation to use urine culture to guide treatment is based on expert opinion in Diagnosis, Prevention, and Treatment of Catheter-Associated Urinary Tract Infection in Adults: 2009 International Clinical Practice Guidelines published by the Infectious Diseases Society of America and is intended to reduce the risks of complications and treatment failure, which are generally increased in people with an indwelling urinary catheter [Hooton et al, 2010].

Taking the urine sample for culture from a freshly placed catheter

This recommendation is based on expert opinion in guidelines from the Infectious Diseases Society of America [Hooton et al, 2010]. A biofilm forms on a catheter within 1-3 days of insertion and, may not reflect the organisms within the bladder. Therefore the urine sample should be taken from a freshly placed catheter.

Empirical antibiotic choice for catheterized women

What antibiotic should I prescribe empirically for catheterized women with a suspected urinary tract infection?

Do not treat asymptomatic bacteriuria in women with indwelling catheters.

Offer an antibiotic to all catheterized women with symptoms suggestive of a urinary tract infection (UTI).

For a woman with mild symptoms who has normal immunity and normal renal function, consider withholding antibiotics until culture and sensitivity results are available to guide treatment.

For all other women start treatment without delay. Follow local prescribing guidelines that take into account local resistance patterns, if these are available. Otherwise prescribe:

Trimethoprim  200 mg twice daily, for 7 days but avoid if the woman has taken trimethoprim within the last year, or

Nitrofurantoin  50 mg four times daily, or 100 mg (modified-release) twice daily for 7 days but avoid in women with an eGFR of less than 60 mL/minute/1.73 m3.

Do not recommend cranberry products or urine alkalinizing agents.

Basis for recommendation

Basis for recommendation

Not treating asymptomatic bacteriuria in women with indwelling catheters

The Guideline Review Group (GRG) of the Scottish Intercollegiate Guidelines Network (SIGN) made this recommendation after considering evidence from three high quality studies which found inconsistent benefit from repeated treatment of asymptomatic bacteriuria in people with long-term catheters [SIGN, 2012]. The GRG also found evidence from one high quality study that repeated treatment of asymptomatic bacteriuria increases the risk of colonization by drug-resistant bacteria, and evidence from two studies (quality not stated) that all people with a long-term indwelling urinary catheter will have bacteriuria.

Offering an antibiotic

Using urine culture to guide the choice of antibiotic

The recommendation to use urine culture to guide treatment and, if practical, to withhold treatment until culture results are available, is based on expert opinion in Diagnosis, Prevention, and Treatment of Catheter-Associated Urinary Tract Infection in Adults: 2009 International Clinical Practice Guidelines published by the Infectious Diseases Society of America and is intended to reduce the risks of complications and treatment failure, which are generally increased in people with an indwelling urinary catheter [Hooton et al, 2010].

The recommendation to change to a more appropriate antibiotic if the antibiotic was started empirically and a resistant organism is isolated on urine culture is based on expert opinion in Diagnosis, Prevention, and Treatment of Catheter-Associated Urinary Tract Infection in Adults: 2009 International Clinical Practice Guidelines published by the Infectious Diseases Society of America; this strategy is intended to reduce the risks of complications and treatment failure [Hooton et al, 2010].

Prescribing an antibiotic

What to prescribe

The basis for the recommended treatments are discussed in empirical antibiotic choice.

Avoiding frequent courses of trimethoprim

In people with urinary tract infection (UTI), there is limited evidence to indicate an association between trimethoprim resistance and previous trimethoprim treatment within the past 12 months. Resistance may be more likely the more recent the exposure to trimethoprim, and with more frequent courses of trimethoprim. Studies on nitrofurantoin resistance and previous treatment were not found.

Avoiding nitrofurantoin in women with an eGFR less than 60 ml/min/1.73m2

This recommendation is based on advice in the manufacturers' summary of product characteristics and Medicines and Healthcare Products Regulatory Agency reminder on precautions for use [ABPI Medicines Compendium, 2013b; MHRA, 2013].

Duration of treatment

Treatment for 7 days is recommended by experts from the Infectious Diseases Society of America because there is evidence that longer courses of treatment are associated with a better outcome in terms of cure, persistence and relapse compared to 3 day courses of treatment [Hooton et al, 2010]. In the opinion of these experts treatment for more than 7 days is unlikely to be of greater benefit provided the woman has had a prompt resolution of symptoms.

Treating for 7 days is also in line with recommendations from SIGN [SIGN, 2012], and the European Association of Urology [Grabe et al, 2013].

Not treating asymptomatic bacteriuria in women with indwelling catheters

The Guideline Review Group (GRG) of the Scottish Intercollegiate Guidelines Network (SIGN) made this recommendation after considering evidence from three high quality studies which found inconsistent benefit from repeated treatment of asymptomatic bacteriuria in people with long-term catheters [SIGN, 2012]. The GRG also found evidence from one high quality study that repeated treatment of asymptomatic bacteriuria increases the risk of colonization by drug-resistant bacteria, and evidence from two studies (quality not stated) that all people with a long-term indwelling urinary catheter will have bacteriuria.

Treatments not recommended

Cranberry juice or other cranberry products are not recommended as there is no good evidence to support their use for treating urinary tract infection.

Although urine alkalization has been traditionally used to relieve the symptoms of a urinary tract infection, there is a lack of good evidence to support its use.

Follow up

How should I follow up catheterized women with a suspected urinary tract infection?

Advise all women to seek medical attention if they develop fever, loin pain, or do not respond to treatment. If loin pain or fever develops in association with a urinary tract infection (UTI), see the CKS topic on Pyelonephritis - acute.

Arrange routine follow up — the timing of which will depend on the clinical situation.

Clinical judgement is required to assess the clinical significance of urine culture results from people with a long-term indwelling urinary catheter, as there is a high probability of culturing bacteria which are not causing symptoms or signs of urinary tract infection [HPA, 2011; SIGN, 2012].

If symptoms of a UTI persist following treatment:

Adjust treatment, if necessary, when the results of urine culture and sensitivity testing become available.

Seek specialist advice if a multi-resistant organism is cultured.

If there is no bacterial growth after culturing the urine, consider an alternative cause for symptoms.

Basis for recommendation

Basis for recommendation

Advising women to seek medical advice if they develop fever or loin pain

In the absence of more sensitive clinical features or practical clinical tests, experts recommend diagnosing acute pyelonephritis based on evidence of a urinary tract infection (UTI) in a person with loin pain or a temperature over 38°C [Tomson and Armitage, 2010].

Women with pyelonephritis need more intensive management and follow up. For more information, see the CKS topic Pyelonephritis - acute.

Arranging follow-up

CKS found no clinical trials or national guidelines on the optimum time for follow-up in catheterized women with a UTI, and therefore recommend that the clinical circumstances should determine the time of follow-up.

Persistent symptoms of a UTI

Adjusting treatment

These recommendations are in line with guidance from the Scottish Intercollegiate Guidelines Network (SIGN) [SIGN, 2012].

The recommendation to offer a different antibiotic if symptoms persist is supported by a study of the course of uncomplicated community-acquired urinary tract infection in women [McNulty et al, 2006]. The study found that, after 5 days of antibiotic treatment, symptoms had resolved in 70% of women infected with an organism sensitive to the antibiotic, and 24% of women with a resistant isolate. The study also found that 50% of those who reconsulted in the first week had a resistant isolate.

When the uropathogen is resistant to the empirically chosen antibiotic and the woman has responded, the recommendation to consider continuing treatment until the end of the antibiotic course is based on comments of expert reviewers of previous versions of CKS topics on urinary tract infection. If symptoms have resolved, there is likely to be little added benefit from changing the antibiotic, because, either the woman is getting better of their own accord, or the laboratory assessment of resistance does not reflect the true susceptibility of the uropathogen.

Seeking specialist advice if a multi-resistant organism is cultured

CKS recommend seeking expert advice to guide treatment when multi-resistant organisms have been cultured. This is in line with advice from the Health Protection Agency who have reported an increasing incidence of UTIs caused by community multi-resistant extended-spectrum Beta-lactamase Escherichia Coli that require expert advice to guide management [HPA and British Infection Association, 2013].

Considering other diagnoses if infection is not confirmed

Most women with cystitis will have a urinary tract infection. Failure to culture bacteria in the urine may be a false negative result, but the possibility of an alternative cause for symptoms is significantly increased and should be considered.

Preventing recurrent UTIs in catheterized women

How can I prevent recurrent urinary tract infections in catheterized women?

Ensure an indwelling urinary catheter is appropriate.

Use intermittent catheterization if appropriate and practical.

Prevent the introduction of infection by:

Obtaining urine samples from a sampling port using aseptic technique.

Changing catheters only when clinically necessary (for example, to prevent blockage), or according to the manufacturer's recommendations.

Prescribing prophylactic antibiotics only for people with a history of catheter-associated urinary tract infection following catheter change.

Do not prescribe antibiotic prophylaxis routinely.

If prophylactic antibiotics are used a narrow spectrum agent should be considered.

Do not use:

Bladder installations or washouts.

Prophylactic antibiotics when changing catheters in women with a heart valve lesion, septal defect, patent ductus, or prosthetic valve.

Topical antiseptics or antibiotics applied to the catheter, urethra, or meatus; daily washing of the meatus with soap and water is sufficient.

Basis for recommendation

Basis for recommendation

These recommendations are based on the clinical guideline Infection: prevention and control of healthcare-associated infections in primary and community care from the National Institute for Health and Clinical Excellence (NICE) [National Clinical Guideline Centre, 2012].

Minimizing the use of indwelling urinary catheters

The recommendations on training and practical ways to minimize the use of indwelling urinary catheters reflect guidelines from NICE [National Clinical Guideline Centre, 2012].

Using intermittent catheterization where appropriate

The NICE Guideline Development Group (GDG) based their recommendation to use intermittent catheterization rather than an indwelling urinary catheter on a systematic review which included 22 studies and reported a higher rate of infection with an indwelling catheter compared with intermittent catheterization [National Clinical Guideline Centre, 2012].

Preventing the introduction of infection

Using a sampling port to collect a urine culture sample

The NICE Guideline Development Group (GDG) stress the importance of maintaining a sterile continuously closed urinary drainage system to prevent catheter-associated urinary tract infections [National Clinical Guideline Centre, 2012]. Breaches in the system such as taking a urine sample increases the risk of catheter-associated infection, and their expert opinion is that urine samples must be obtained from a sampling port using an aseptic technique.

Changing a catheter only when clinically necessary

The GDG of NICE found no definitive evidence about the optimum time to change a catheter. This recommendation is therefore based on their expert opinion [National Clinical Guideline Centre, 2012].

Not prescribing antibiotics routinely for catheter change

This recommendation is based on expert opinion of the GDG of NICE. The GDG felt that in groups with a high risk of infection, the potential benefits of antibiotic use might out way the potential disadvantages [National Clinical Guideline Centre, 2012]. They recommend following local guidelines because resistance patterns could vary between localities and over time. Expert opinion from the GDG of the Scottish Intercollegiate Guidelines Network (SIGN) is that a narrow spectrum antibiotic should be considered because of the risk of Clostridium difficile infection with broad spectrum antibiotics [SIGN, 2012].

Avoiding preventative measures

Not using bladder instillations or washouts

There is a lack of good quality evidence about the effectiveness of bladder instillations and washouts. They are not recommended because in the opinion of the GDG of NICE they are likely to increase the number of infections [National Clinical Guideline Centre, 2012].

Not using prophylactic antibiotics

The GDG of NICE based the recommendation not to prescribe prophylactic antibiotics when changing catheters on evidence (from one very low quality study) that antibiotic prophylaxis did not reduce the risk of urinary tract infection [National Clinical Guideline Centre, 2012].

The recommendation not to use prophylactic antibiotics when changing catheters in women with a heart valve lesion, septal defect, patent ductus, or prosthetic valve is based on the NICE clinical guideline on prophylaxis for infective endocarditis, which found this not to be cost-effective [NICE, 2008c]. 

Not using topical antiseptics or antibiotics

The recommendation not to use topical antiseptics or antibiotics applied to the catheter, urethra, or meatus is based on findings from six studies that compared meatal cleansing with a variety of antiseptic/antimicrobial agents or soap and water and found that using antiseptics and antimicrobial agents did not reduce the rate of bacteriuria. A systematic review found that vigorous cleansing of the meatus may increase the risk of infection [National Clinical Guideline Centre, 2012].

Important aspects of prescribing information relevant to primary healthcare are covered in this section specifically for the drugs recommended in this CKS topic. For further information on contraindications, cautions, drug interactions, and adverse effects, see the electronic Medicines Compendium (eMC) (http://medicines.org.uk/emc), or the British National Formulary (BNF) (www.bnf.org).

Trimethoprim

Trimethoprim

Dosage

For treatment of urinary tract infection (UTI), prescribe trimethoprim 200 mg twice daily.

For uncomplicated UTIs, treat for 3 days.

For complicated UTIs, treat for 5–10 days.

In pregnant women, treat for 7 days.

For long-term prophylaxis, prescribe 100 mg at night.

Reduce the dose if the woman has severe renal impairment — particularly if treatment is prescribed for more than 3 days (for example, people with complicated urinary tract infection or for long-term prophylaxis).

For a glomerular filtration rate (GFR) of 15–25 mL per minute, prescribe the normal dose for 3 days, then reduce the dose by half.

For a GFR less than 15 mL per minute, prescribe half the normal dose.

Basis for recommendation

Standard dose

These doses are based on those recommended by the manufacturer of trimethoprim and the British National Formulary [ABPI Medicines Compendium, 2012c; BNF 66, 2013].

Dose in renal failure

Dose adjustment is recommended for people with severe renal impairment because of the risk of drug accumulation in these people — 40–60% of the trimethoprim dose (along with its metabolites) is excreted unchanged in the urine within 24 hours [ABPI Medicines Compendium, 2012c].

Contraindications

Avoid using trimethoprim in women with blood dyscrasias.

Basis for recommendation

Because of its potential anti-folate effect, there have been reports that trimethoprim causes blood disorders. For further information, see Adverse effects. Consequently, trimethoprim is contraindicated in people with dyscrasias [ABPI Medicines Compendium, 2012c; BNF 66, 2013].

Precautions

Trimethoprim should be prescribed with caution in the following conditions:

Severe renal impairment

As the drug is predominantly excreted by the kidney, dose adjustment may be required. For further information, see Dosage.

Folate deficiency

Because of its potential anti-folate effect, there is a risk of further exacerbating folate deficiency in people who are folate deficient or who are predisposed to folate deficiency (for example elderly people). Consequently, consider prescribing a folate supplement (if this has not already been prescribed) — particularly if trimethoprim is prescribed long term.

The risk of folate deficiency is also increased when trimethoprim is combined with certain drugs. For further information, see Drug interactions.

Pregnancy

See Pregnancy and breastfeeding with trimethoprim.

Long-term trimethoprim treatment

Because of the risk of haematological adverse effects (see Adverse effects), the British National Formulary (BNF) advises that people should be warned to seek immediate medical attention if they develops signs of blood disorders.

Although the manufacturer of trimethoprim recommends regular blood monitoring, the BNF found no practical evidence to support this.

[ABPI Medicines Compendium, 2012c; BNF 66, 2013]

Pregnancy and breastfeeding

For women who are pregnant:

Trimethoprim can be used during pregnancy (although it should not be used first-line when other alternatives are available — for example, amoxicillin, if the bacterial sensitivity is known).

For information on treating asymptomatic and symptomatic urinary tract infection in pregnancy, see Asymptomatic bacteriuria in pregnancy, Empirical antibiotic choice for women with non-visible of visible haematuria, and Empirical antibiotic choice for pregnant women for women who do not have non-visible or visible haematuria.

If trimethoprim is recommended for use during the first trimester:

Ensure the woman is taking a folic acid supplement.

Prescribe folic acid 5 mg daily.

For women who are breastfeeding

Trimethoprim can be used in women who are breastfeeding.

Basis for recommendation

These recommendations have largely been based on a reference textbook, and information from the UK Teratology Information Service (UKTIS), (formerly the National Teratology Information Service [NTIS]) [Schaefer et al, 2007; UKTIS, 2013].

Safety in pregnancy

Concerns have been expressed about the use of trimethoprim during pregnancy because it is a folic acid antagonist, and low levels of folic acid have been associated with serious birth defects.

The evidence on the risks of trimethoprim during pregnancy has been critically assessed by the UK Teratology Information Service (UKTIS), formerly the National Teratology Information Service (NTIS) [UKTIS, 2013]. A similar systematic review was conducted by the Centers for Disease Control (CDC) in the USA, to assess the safety of trimethoprim-sulfamethoxazole used for prophylaxis in HIV-infected pregnant women [Forna et al, 2006]. The NTIS and CDC concluded that the benefits outweighed the risks, which were small. Additionally the NTIS concluded that:

In women with normal folate status, who are well nourished, use of trimethoprim for a short period is unlikely to induce folate deficiency.

Although trimethoprim is commonly used to treat symptomatic UTIs, good evidence to support its use in pregnancy is lacking. In addition, National Diet and Nutrition Surveys have found that in the UK, women's dietary intake of iron, vitamin D, calcium, and folate remain below recommended levels [Ruxton and Derbyshire, 2010]. The recommendation by the UK teratology information service to avoid trimethoprim if folate deficiency is suspected is based on the results of a case control study that found that women with an established low folate status taking trimethoprim during the 2 months after the last menstrual period had an increased risk of giving birth to an infant with a neural tube defect compared with women who had not taken trimethoprim (odds ratio 4.8, 95% CI 1.5 to 16.1) [Hernández-Díaz et al, 2001]. Women who took trimethoprim or carbamazepine and who did not take folic acid supplements were more likely to give birth to an infant with a neural tube defect (OR 13.3, 95% CI 2.9 to 61.4) compared with women on one of these medications who took a daily folic acid supplement (OR 1.2, 95% CI 0.1 to 12.7). Although numbers in this study were small the UKTIS considered that they support the recommendation to prescribe high dose folic acid (5 mg) to women treated with trimethoprim during the first trimester although there is no evidence that this higher dose is any more beneficial than a 400 mcg daily dose [UKTIS, 2013].

Trimethoprim is not licensed for use during pregnancy. This is reflected in the British National Formulary which recommends that it should avoided in the first trimester due to its anti-folate effect [BNF 66, 2013].

Folic acid supplement

When trimethoprim is prescribed during the first trimester, the recommendation to co-prescribe a folic acid supplement is precautionary [Schaefer et al, 2007; UKTIS, 2013].

A dose of folic acid 5 mg daily is usually recommended although it is not known whether this higher dose confers any additional benefit over the routine antenatal dose of 400 mcg daily [UKTIS, 2013].

Advice is conflicting regarding the use of trimethoprim during breastfeeding:

The UK Teratology Information Service (UKTIS) found no reports of neonatal toxicity following exposure to trimethoprim during lactation [UKTIS, 2013].

The British National Formulary states that trimethoprim is present in milk and short-term use is not known to be harmful [BNF 66, 2013].

Although it is excreted in breast milk, most manufacturers of trimethoprim state that short-term trimethoprim treatment is not contraindicated in women who are breastfeeding [ABPI Medicines Compendium, 2011b; ABPI Medicines Compendium, 2012c; ABPI Medicines Compendium, 2013a], but one manufacturer states that effects on the infant are likely if the mother is taking therapeutic doses, and trimethoprim is contraindicated if the infant is less than 4 months of age [ABPI Medicines Compendium, 2013d].

Drug interactions

The following drug interactions have been reported with trimethoprim, when it is combined:

With methotrexate (a folate antagonist):

Several cases of bone marrow suppression have been reported (some fatal) [Baxter and Preston, 2013].

With azathioprine:

Increased risk of haematological toxicity has been reported in some people with a renal transplant who are taking azathioprine — particularly if both drugs are given over a prolonged period [Baxter and Preston, 2013].

Nevertheless, for most people, both drugs can be taken together. The combination is commonly used in practice.

The reaction is also expected for mercaptopurine (a metabolite of azathioprine).

With phenytoin:

There is a small risk of phenytoin toxicity (particularly if the serum phenytoin levels are at the top end of the range) as trimethoprim can decrease the clearance of phenytoin [Baxter and Preston, 2013]. Signs of phenytoin toxicity include blurred vision, nystagmus, ataxia, or drowsiness.

This interaction is also expected with fosphenytoin (a pro-drug of phenytoin).

With ciclosporin:

Increased nephrotoxicity has been reported. However, this interaction has not been firmly established [Baxter and Preston, 2013].

With digoxin and warfarin. However the clinical significance of these interactions is still uncertain [Baxter and Preston, 2013].

Digoxin: trimethoprim has been reported to increase digoxin levels by an average of 22% in nine elderly people after taking trimethoprim 200 mg daily for 14 days (although an increase of 75% was experienced by one person). Consider monitoring for signs of digoxin toxicity in the elderly if trimethoprim is given long term for prophylaxis [Baxter and Preston, 2013].

Warfarin: the manufacturer of trimethoprim warns that it may potentiate the anticoagulant effect of warfarin. However, no case reports or controlled trials on this interaction have been reported. Although there is some indication of increased anticoagulant effect from two cohort studies, the interaction is likely to be small (if it occurs), requiring little or no adjustment in warfarin dose [Baxter and Preston, 2013].

Adverse effects

Trimethoprim is generally well tolerated.

Nausea, vomiting, pruritus, and skin rashes have occasionally been reported. These are generally mild and reversible when trimethoprim is withdrawn.

Severe adverse drug reactions with trimethoprim are rare.

There have been rare reports of trimethoprim causing haematological adverse effects, including [Aronson, 2006]:

Macrocytic and megaloblastic anaemia: this is more likely in people with pre-existing folate deficiency.

Agranulocytosis — very rare. In people where leukocytes are monitored regularly, mild leukopenia has been reported in 0.4–10% of people taking trimethoprim or co-trimoxazole (trimethoprim plus sulfamethoxazole).

Aplastic anaemia, neutropenia, thrombocytopenia, and pancytopenia.

For people receiving long-term trimethoprim treatment, the British National Formulary advises that they should be warned to seek immediate medical attention if they develop signs of blood disorders such as fever, sore throat, rash, mouth ulcers, purpura, bruising, or bleeding [BNF 66, 2013].

[Aronson, 2006; ABPI Medicines Compendium, 2012c]

Nitrofurantoin

Nitrofurantoin

Dosage

For treatment of acute urinary tract infection (UTI), prescribe nitrofurantoin 50 mg four times daily, or 100 mg (modified-release), twice daily.

For uncomplicated UTIs, treat for 3 days.

For complicated UTIs, treat for 5–10 days.

In pregnant women, treat for 7 days.

For long-term prophylaxis, prescribe 50 mg to 100 mg at night.

Basis for recommendation

Dosage

These dosages are based on those recommended in the manufacturer's Summary of Product Characteristics for nitrofurantoin [ABPI Medicines Compendium, 2012d; ABPI Medicines Compendium, 2012b; ABPI Medicines Compendium, 2013b; ABPI Medicines Compendium, 2013c].

Duration of treatment in non-pregnant women

A 3-day course of empirical treatment is recommended because there is good evidence from Cochrane systematic reviews that this achieves symptomatic cure in people with uncomplicated UTI; it is more effective than single-dose treatment and as effective as 5–10-day courses [Milo et al, 2005; Lutters and Vogt-Ferrier, 2008]. This is also in line with recommendations from the Scottish Intercollegiate Guideline Network [SIGN, 2012], the Health Protection Agency [HPA, 2011], and an international guideline [American College of Obstetricians and Gynecologists, 2008].

For people with a complicated UTI, a longer course is recommended because there is evidence from a Cochrane systematic review that a 5–10 day course of antibiotics produced a higher bacteriological cure rate (but more adverse effects) than a 3-day regimen [Milo et al, 2005]. The Cochrane systematic review concluded that a 5–10 day course may be considered for women in whom eradication of bacteriuria is important.

Duration of antibiotic treatment in pregnant women

Evidence on different antibiotic regimens for treating symptomatic UTIs in pregnant women is lacking.

Given the possible increased risk of fetal complications with a UTI, a 7-day course of antibiotics is preferred over shorter courses because:

For pregnant women with an acute uncomplicated UTI, a Cochrane systematic review and meta-analysis of 13 trials found that a 1-day course of antibiotics was less likely to produce a cure than courses of 4–7 days, but the difference was not statistically significant. The authors concluded that a 7-day regimen should be considered for pregnant women because the risk of harm from a UTI is higher than for non-pregnant women [Widmer et al, 2011].

A recent large WHO study which was included in the Cochrane review found a higher cure rate with a 7-day course of nitrofurantoin (86%) than a 1-day regimen (76%) in pregnant women with asymptomatic bacteriuria. Adverse effects were not statistically different between the two groups [Lumbiganon et al, 2009].

A 7-day course is supported by guidance issued by the Health Protection Agency based on expert consensus [HPA and British Infection Association, 2013].

Formulation

Nitrofurantoin is available in the following oral formulations:

Immediate-release tablets and capsules (50 mg and 100 mg).

Modified-release capsules (100 mg).

Oral suspension (25 mg/5 mL).

For treatment of urinary tract infection (UTI):

Both the immediate-release and modified-release formulations may be considered.

If compliance is a problem, the modified-release formulation may be preferred with its twice-daily dosage (the other formulations need to be taken four times daily).

If nausea and vomiting is a concern, a formulation that contains macrocrystalline nitrofurantoin (such as immediate-release capsules) may reduce the likelihood of these adverse effects, but the evidence-base to support this is weak.

For long-term prophylaxis of UTIs:

Prescribe an immediate-release formulation.

The modified-release formulation is not licensed for long-term prophylaxis.

The oral suspension may be considered for people with swallowing difficulties or requiring enteral-tube feeding.

Basis for recommendation

For treatment of urinary tract infections (UTIs):

CKS identified no strong evidence to prefer one nitrofurantoin formulation over another.

There is weak evidence from two small RCTs and one retrospective case series that the macrocrystalline formulation of nitrofurantoin (Macrodantin® capsules) causes less nausea than the crystalline formulation (Furadantin® tablets) in people with a UTI. However, these results should be interpreted with caution as there are a number of methodological weaknesses with these studies.

No other studies comparing the modified-release nitrofurantoin formulation (Macrobid®) with immediate-release formulations of nitrofurantoin were found.

Given its twice-daily dosage, the modified-release preparation may be preferred if compliance is a problem.

For long-term prophylaxis:

Long-term prophylaxis only requires one dose to be taken at night. For further information, see Preventing recurrent UTIs.

Contraindications

Avoid prescribing nitrofurantoin to people with:

A creatinine clearance less than 60 mL per minute, or elevated serum creatinine [MHRA, 2013].

Acute porphyria.

A deficiency of glucose-6-phosphate dehydrogenase

This is found in 10% of black people and a variable percentage of ethnic groups of Mediterranean, near Eastern, and Asian origin. It is rare in Caucasians.

Avoid nitrofurantoin in these people as it may cause haemolysis.

Discontinue nitrofurantoin if there is any sign of haemolysis (which ceases when the drug is withdrawn).

[ABPI Medicines Compendium, 2013b; ABPI Medicines Compendium, 2013c]

Precautions

Peripheral neuropathy

Nitrofurantoin should be used with caution in people with anaemia, diabetes mellitus, electrolyte imbalance, debilitating conditions, and vitamin B (particularly folate) deficiency since these conditions may enhance the occurrence of peripheral neuropathy.

The manufacturer of nitrofurantoin advises stopping the drug at the first signs of neural involvement (paraesthesiae).

If the person develops unexplained pulmonary, hepatotoxic, haematological, or neurologic syndromes, discontinue treatment with nitrofurantoin.

[ABPI Medicines Compendium, 2013c; ABPI Medicines Compendium, 2013b]

Pregnancy and breastfeeding

Nitrofurantoin can be used in women who are pregnant or breastfeeding.

The BNF recommends that nitrofurantoin should be avoided at term, because of the risk of neonatal haemolysis.

However, the risk seems very small — significant placental transfer of nitrofurantoin does not occur, and only one case of haemolytic anaemia in a neonate has been reported following maternal treatment with nitrofurantoin shortly before delivery.

Because it is excreted in milk, avoid breastfeeding during treatment with nitrofurantoin if the newborn is glucose-6-phosphate dehydrogenase deficient.

Basis for recommendation

Safety in pregnancy

Nitrofurantoin has been used for many years for the prophylaxis and treatment of urinary tract infection and asymptomatic bacteriuria in pregnancy [Schaefer et al, 2007; UKTIS, 2012b].

The drug is concentrated in the urinary tract. Consequently, significant transfer across the placenta does not occur.

Although it is not licensed for use in pregnancy, the manufacturer of nitrofurantoin reported that the drug has been used extensively clinically since 1952 and its suitability in pregnancy has been well documented [ABPI Medicines Compendium, 2013c].

Significant placental transfer of nitrofurantoin does not occur.

There have been no reports of increased risk of congenital malformations [Schaefer et al, 2007].

Haemolytic reactions are rare. There has been one case report of haemolytic anaemia in a newborn baby (without glucose-6-phosphate dehydrogenase deficiency) after maternal treatment with nitrofurantoin shortly before delivery [Bruel et al, 2000; Schaefer et al, 2007].

Breastfeeding

Nitrofurantoin may be used in women who are breastfeeding [Schaefer et al, 2007].

Because it is excreted in milk, there is a potential to cause haemolytic anaemia in newborns who are glucose-6-phosphate dehydrogenase deficient. However, this is rare, and only one case report has been documented [Schaefer et al, 2007]. The recommendation to temporarily cease breastfeeding is pragmatic advice.

Drug interactions

The use of alkalinizing agents (such as potassium citrate) should be avoided in people taking nitrofurantoin. The antibacterial activity of the nitrofurantoin is reduced when the pH of the urine is increased [SIGN, 2012].

Although the manufacturer of nitrofurantoin advises against concomitant administration of magnesium trisilicate with nitrofurantoin (due to reduced absorption), the clinical significance is uncertain as only one very small study in 6 people has reported this effect [Baxter and Preston, 2013].

Adverse effects

Adverse effects associated with nitrofurantoin

Pulmonary: nitrofurantoin-associated pulmonary reactions are reported in less than 1% of people treated with nitrofurantoin. Common manifestations are dry cough, chest pain, dyspnoea, and hypoxemia. Skin rash, arthralgia, and elevated liver enzymes are occasionally present. Chest imaging shows patchy infiltrates and fibrosis. Treatment includes stopping the medication and prescribing a course of corticosteroids [Vahid and Wildemore, 2006].

Gastrointestinal: nausea and anorexia have been reported. Vomiting, abdominal pain, and diarrhoea are less common gastrointestinal reactions [ABPI Medicines Compendium, 2012b; ABPI Medicines Compendium, 2013b; ABPI Medicines Compendium, 2013c].

Peripheral neuropathy (including optical neuritis), with symptoms of sensory as well as motor involvement, has been reported infrequently.

Stop treatment at the first sign of neurological involvement.

Amoxicillin

Prescribing information on amoxicillin

Prescribing issues

Dosage

For adults: prescribe amoxicillin 250 mg three times daily, increasing to 500 mg three times daily for more severe infections.

For the treatment of symptomatic or asymptomatic urinary tract infections in pregnancy, a 7-day course is recommended.

Contraindications

Amoxicillin should not be taken by people who have true penicillin allergy [BNF 66, 2013].

Use in pregnancy and breastfeeding

Amoxicillin can be used in women who are pregnant or breastfeeding [Schaefer et al, 2007; ABPI Medicines Compendium, 2012a]. It has never been associated with adverse effects on the foetus [UKTIS, 2012c].

Amoxicillin is licensed for the treatment of bacteriuria in pregnancy [ABPI Medicines Compendium, 2012a].

Adverse effects

Adverse effects that are commonly reported are gastrointestinal (such as nausea, vomiting, and diarrhoea), and skin rash.

Drug interactions

Contraceptives: Additional contraceptive precautions are not required during or after courses of antibiotics that do not induce liver enzymes [FSRH, 2011].

However, women should be advised about the importance of correct contraceptive practice if they experience vomiting or diarrhoea.

Anticoagulants: documented reports of oral anticoagulant/penicillin (including amoxicillin) interaction are relatively rare [Baxter and Preston, 2013]. However, the British National Formulary advises that common experience in anticoagulant clinics is that a course of broad spectrum penicillin can alter the international normalized ratio [BNF 66, 2013].

Cefalexin

Cefalexin

Prescribing issues

Cefalexin is a first generation cephalosporin.

Dosage

For uncomplicated urinary tract infections: prescribe cefalexin 250 mg every 6 hours, or 500 mg every 12 hours [ABPI Medicines Compendium, 2011a].

For treatment of symptomatic or asymptomatic urinary tract infections (UTIs) in pregnancy, a 7-day course is recommended [HPA and British Infection Association, 2013].

Contraindications and precautions

Cefalexin should not be taken by people with known allergy to the cephalosporin group of antibiotics [BNF 66, 2013].

Cephalosporins should given cautiously to penicillin-sensitive people. The British National Formulary advises that about 10% of penicillin-sensitive people will also be allergic to cephalosporins [BNF 66, 2013].

The Health Protection Agency advises that broad spectrum antibiotics (such as co-amoxiclav, quinolones, and cephalosporins) should be avoided when narrow spectrum antibiotics remain effective, as they increase risk of Clostridium difficile, meticillin-resistant Staphylococcus aureus (MRSA), and resistant UTIs [HPA and British Infection Association, 2013].

Use in pregnancy and breastfeeding

Although cefalexin is not licensed in these groups, the manufacturer reported no evidence of teratogenicity in clinical studies [ABPI Medicines Compendium, 2011a]. The UK teratology information service state that most of the available data does not suggest an increased risk of congenital abnormalities. Although two retrospective studies (one unpublished) have identified a possible association with cardiovascular defects and oral clefts a causal association remains to be proven [UKTIS, 2012a]. Cephalosporins (such as cefalexin) can be used in women who are pregnant or breastfeeding [Schaefer et al, 2007; BNF 66, 2013].

Adverse effects

Gastrointestinal adverse effects (such as nausea, vomiting, and diarrhoea), are commonly reported [ABPI Medicines Compendium, 2011a].

The manufacturer of cefalexin advises that pseudomembranous colitis should be considered in people who develop antibiotic-associated diarrhoea [ABPI Medicines Compendium, 2011a]. It can be mild, or can be life threatening. Mild cases of pseudomembranous colitis usually respond to drug discontinuation alone.

Drug interactions

Contraceptives: Additional contraceptive precautions are not required during or after courses of antibiotics that do not induce liver enzymes [FSRH, 2011].

However, women should be advised about the importance of correct contraceptive practice if they experience vomiting or diarrhoea.

Anticoagulants: documented reports of oral anticoagulant/penicillin (including amoxicillin) interaction are relatively rare [Baxter and Preston, 2013]. However, the British National Formulary advises that common experience in anticoagulant clinics is that a course of broad spectrum penicillin can alter the international normalized ratio [BNF 66, 2013].

Evidence

Evidence

Supporting evidence

Antibiotic treatment for non-pregnant women

Evidence on antibiotic treatment for UTI in non-pregnant women

Treatment strategies

Evidence on antibiotic treatment strategies for acute cystitis

There is evidence from randomized controlled trials (RCTs) in primary care that the optimum antibiotic treatment strategy for women with cystitis should take account of symptom severity and the woman's preferences.

One RCT in 59 women with cystitis but negative urine dipstick tests found that the median time for resolution of dysuria was 3 days with trimethoprim and 5 days with placebo (p = 0.002) [Richards et al, 2005].

Five commonly used strategies for treating suspected cystitis in UK primary care were compared in one RCT [Little et al, 2009].

The study included women who sought treatment by a doctor or nurse in general practices in the south of England and who were considered on clinical grounds to have cystitis without vaginitis or urethritis. The study excluded children, women older than 70 years of age, women with possible pyelonephritis, and women with nausea, vomiting, or other severe systemic symptoms.

Women with suspected cystitis were randomized to one of five antibiotic treatment strategies:

Immediate antibiotics

Women were prescribed empirical antibiotic treatment (antibiotic choice was not guided by results of urine culture).

Trimethoprim 200 mg twice daily for 3 days was prescribed unless the woman was allergic to trimethoprim, in which case cefaclor or cefalexin was offered.

Delayed antibiotics

Women were asked to wait 48 hours, but could use empirical antibiotics at their discretion.

Symptom rule

Empirical antibiotics were given if two or more of the following symptoms and signs were present: cloudy urine, offensive-smelling urine, moderately-severe dysuria, or moderately-severe nocturia.

Women who did not meet the criteria for antibiotic treatment were offered a delayed antibiotic prescription to use if their symptoms did not settle after 48 hours.

Dipstick rule

Empirical antibiotics were given if the urine dipstick test was positive for nitrites or for both leucocytes and blood (at least a trace).

Women who did not meet the criteria for antibiotic treatment were offered a delayed antibiotic prescription to use if their symptoms did not settle after 48 hours.

MSU rule

Symptomatic treatment was given until the results of culture of a mid-stream urine specimen were available to guide antibiotic treatment (results usually available within 48 hours).

Results

There were no significant differences in symptom duration, severity or frequency of symptoms, or severity of unwell symptoms between the antibiotic management strategies, and little difference (p = 0.79) between groups for re-contact with the GP surgery recorded in the notes in the first 4 weeks. Selected details are shown in Table 1.

Two groups of symptoms were identified by factor analysis:

Frequency symptoms (increased daytime frequency of urination, increased night-time frequency, urgency, and dysuria), Cronbach's alpha 0.77. Cronbach's alpha is a statistical test for internal consistency between items in a group. As a rule of thumb, values above about 0.7 can be taken as significant [Bland and Altman, 1997].

Unwell symptoms (abdominal pain, restricted activities, and feeling unwell), Cronbach's alpha 0.80.

Antibiotic resistance and no antibiotic treatment were both significantly associated with slightly worse frequency symptoms at days 2–4. Compared with the group with a sensitive uropathogen, the differences in mean symptom scores (maximum 6) were, for the group with:

A resistant organism 0.54 (95% CI 0.22 to 0.87, p < 0.001).

No antibiotic treatment 0.60 (95% CI 0.14 to 1.05, p = 0.011).

The severity of baseline frequency symptoms was positively associated with severity of symptoms at days 2–4: beta-coefficient 0.40 (95% CI 0.33 to 0.46, p < 0.001). Unwell symptoms had a modest association.

Studies of the responses to treatment with or without an antibiotic showed that antibiotics can be expected to shorten the duration of symptoms, that are at least moderately severe, by 1–2 days.

In each group where women were offered a delayed prescription, a high proportion chose to use it.

Table 1 . Effects of different treatment strategies.
Treatment strategy Symptom duration* Women presenting again within 4 weeks
Immediate antibiotics 3.54 (95% CI 2.87 to 4.21) 10% (6/58)
Delayed antibiotics 3.96 (95% CI 3.01 to 5.20) 9% (5/58)
Symptom rule 3.93 (95% CI 3.01 to 5.10) 13% (8/64)
Dipstick rule 3.22 (95% CI 2.41 to 5.70) 12% (6/51)
MSU rule 4.28 (95% CI 3.26 to 5.70) 17% (9/52)
* Symptom duration is the mean (and 95% confidence interval) number of days that symptoms were at least moderately severe. The study reported incidence rate ratios (IRRs) for all groups except the 'immediate antibiotics' group. CKS used the IRRs to estimate the absolute durations and 95% CIs for the other groups. The 95% CI for the 'immediate antibiotics' group was estimated from the mean, standard deviation, and number of women in the group. Although this method for estimating the 95% confidence intervals is not mathematically rigorous, it is accurate enough to give a clinical impression of the varations in symptom duration, and to show how the results in the different groups overlap.
Data from: [Little et al, 2009]

Uncomplicated cystitis

Evidence on antibiotic treatment for non-pregnant women with uncomplicated cystitis

There is good evidence from a systematic review of randomized controlled trials (RCTs) that antibiotic treatment is more effective than placebo in treating uncomplicated cystitis in women. The incidence of adverse events was higher with antibiotic treatment, but there was no difference in withdrawal rate due to adverse events between women treated with an antibiotic or placebo.

The benefits and harms of antibiotic treatment for women with uncomplicated cystitis were examined in a systematic review and meta-analysis (search date: November 2008) [Falagas et al, 2009].

The systematic review included five double-blind RCTS, with from 45–1143 participants ranging from 15–84 years of age. Follow up varied from 3 days to 3 months after the end of treatment.

The included studies compared antibiotic treatment (amoxicillin, cefixime, co-trimoxazole, nitrofurantoin, ofloxacin, and pivmecillinam) with placebo.

Benefits

Compared with placebo, antibiotic treatment was found to improve clinical symptoms and bacteriological cure as well as reducing the risk of relapse or reinfection.

Clinical success (defined as complete cure or improvement of symptoms): odds ratio (OR) 4.81 (95% CI 2.51 to 9.21; four RCTs, n = 1062).

Clinical cure: OR 4.67 (95% CI 2.34 to 9.35; four RCTs, n = 1062).

Bacteriological eradication (defined as negative urine culture) at end of treatment: OR 10.67 (95% CI 2.96 to 38.43; three RCTs, n = 967).

Bacteriological eradication at the first follow-up assessment: OR 5.38 (95% CI 1.63 to 17.77; three RCTs, n = 738).

Microbiological reinfection or relapse (defined as presence of a positive urine culture at long-term follow-up assessment) was less likely in people treated with antibiotics: OR 0.27 (95% CI 0.13 to 0.55; five RCTs, n = 843).

No statistically significant difference was found between antibiotic treatment and placebo for:

Incidence of pyelonephritis (diagnosed when there was fever and flank pain): OR 0.33 (95% CI 0.04 to 2.70; two RCTs, n = 962).

Emergence of bacterial resistance: OR 1.32 (95% CI 0.50 to 3.48; three RCTs, n = 173).

Harms

Adverse events were more likely to occur with antibiotic treatment: OR 1.64 (95% CI 1.10 to 2.44; four RCTs, n = 1068).

The difference between antibiotic and placebo groups in numbers of participants withdrawing due to adverse events was not statistically significant: OR 1.57 (95% CI 0.31 to 7.93; three RCTs, n = 1007).

Duration

Evidence on duration of antibiotic treatment for uncomplicated UTIs

Good evidence from two Cochrane systematic reviews of randomized controlled trials (RCTs) suggests that 3-day courses of antibiotics are almost as effective as longer courses, and adverse effects (including discontinuation) are less likely.

A Cochrane systematic review (search date: August 2003) assessed evidence on effectiveness of different durations of antibacterial treatments in acute, uncomplicated lower urinary tract infection (UTI) in otherwise healthy women 16–65 years of age [Milo et al, 2005].

The study included 32 trials with 9605 participants, and compared outcomes for 3 days' and 5–10 days' treatment with antibiotic.

Benefits were either similar or moderately less likely with 3-day treatment.

There was no significant difference between 3-day and 5 to 10-day antibiotic regimens in terms of:

Symptomatic failure rates at short-term follow up: relative risk (RR) 1.06 (95% CI 0.88 to 1.28, p = 0.52; no statistically significant heterogeneity).

Symptomatic failure rates at long-term follow up: RR 1.09 (95% CI 0.94 to 1.27, p = 0.10; no statistically significant heterogeneity).

3-day treatment was moderately less effective than 5 to 10-day treatment in terms of:

Persistent bacteriuria, short-term: statistically significant only in the subgroup of trials that used the same antibiotic in the two treatment arms: RR 1.37 (95% CI 1.07 to 1.74, p = 0.01; no statistically significant heterogeneity).

Persistent bacteriuria, long-term: RR 1.31 (95% CI 1.08 to 1.60, p = 0.006; no statistically significant heterogeneity).

Harms were less likely with 3-day treatment.

Adverse effects (of any kind): RR 0.83 (95% CI 0.74 to 0.93, p = 0.001).

Discontinued therapy: RR 0.51 (95% CI 0.328 to 0.91, p = 0.02).

The authors concluded that longer courses could be considered for women in whom eradication of bacteria in the urine is important.

A Cochrane systematic review (search date: May 2008) assessed duration of antibiotic treatment for UTI in elderly women [Lutters and Vogt-Ferrier, 2008].

The review included 15 trials with 1644 elderly women. All trials were assessed as being of low methodological quality.

The review found no important clinical difference between short treatment (3–6 days) compared with longer treatment (7–14 days), but the numbers of studies and participants were low. Selected details are in Table 1 and Table 2.

No difference was found in the rate of adverse drug reactions (RR 0.87, 95% CI 0.26 to 2.93) or discontinuation due to adverse reactions (RR 0.11, 95% CI 0.01 to 1.97) between short and long treatments. However, the review reported one study (n = 183) which found a 3-day course reduced the mean number of adverse events/participant at day 5 (mean difference [MD] 0.70, 95% CI –1.09 to –0.31) and day 9 (MD 0.90, 95% CI –1.33 to –0.47) when compared to a 7-day course [Vogel et al, 2004]. This suggests a safety advantage for short treatment.

Table 1 . Evaluation of treatment of UTI in elderly women: persistent symptoms of UTI at short-term (less than 2 weeks) follow up.
Treatment durations compared Number of trials (participants) Effect (1st group compared to 2nd group)
Single-dose with short course 1 trial (23 participants) No significant difference RR 1.69 (95% CI 0.08 to 37.26)
Single-dose with long course 1 trial (388 participants) No significant difference RR 1.94 (95% CI 0.68 to 5.57)
Short course with long course 4 trials (395 participants) No significant difference RR 0.98 (95% CI 0.62 to 1.54)
Short-course: 3–6 days of treatment; Long-course: 7–14 days of treatment. Data from: [Lutters and Vogt-Ferrier, 2008]
Table 2 . Evaluation of treatment of UTI in elderly women: persistent bacteriuria at short-term (less than 2 weeks) follow up.
Treatment durations compared Number of trials (participants) Effect (1st group compared to 2nd group)
Single-dose with short course 5 trials (356 participants) Significantly higher RR 2.01 (95% CI 1.05 to 3.84)
Single-dose with short or long course (3 to 14 days) 8 trials (809 participants) No significant difference RR 1.51 (95% CI 0.92 to 2.49)
Short course with long course 3 trial (431 participants) No significant difference RR 0.85 (95% CI 0.29 to 2.47)
Short-course: 3–6 days of treatment; Long-course: 7–14 days of treatment. Data from: [Lutters and Vogt-Ferrier, 2008]

Routes of administration

Evidence on the routes of antibiotic administration for UTIs

There is evidence to suggest that urinary tract infections (UTI) can be treated as effectively with oral antibiotics as with antibiotics given intravenously or intramuscularly, or with regimens in which antibiotics are initially given parenterally and then switched to the oral route.

A Cochrane systematic review (search date: August 2007) of antibiotic treatments for severe UTI (including pyelonephritis) assessed the effects of different routes of administering antibiotics [Pohl, 2007].

The review included 15 trials (nine in children; four in adults; one in non-pregnant women; and one in pregnant women). The review excluded trials of treatments of mild UTI.

Routes of administration were classed as oral, parenteral (intravenous infusion or intramuscular injection), switch therapy (initial parenteral therapy followed by oral therapy), and single-dose therapy (single intramuscular dose). For the purpose of this review, only comparisons with the oral route are shown.

Oral compared with parenteral administration (one trial, 38 participants)

For people with cystitis, both routes produced similarly high cure rates.

For people with pyelonephritis, the parenteral route was more effective than the oral route in producing bacteriological cure:

At the end of therapy: relative risk (RR) of bacteriological cure 1.37 (95% CI 1.02 to 1.84).

After 4 weeks: RR 1.95 (95% CI 1.24 to 3.08).

Oral administration compared with switch therapy (five trials, 1040 participants)

No statistically significant difference was found for:

Clinical and bacteriological cure rates — during therapy, at the end of therapy, and at follow up.

Reinfection rates (at the end of therapy).

Relapse rates (at the end of therapy and at follow up).

Kidney scarring rates on dimercaptosuccinic acid (DMSA) scan after 6 months.

Mean time to cessation of fever.

Rates of adverse events.

Oral administration compared with single-dose therapy (one trial, 69 children)

No statistically significant difference was found for:

Clinical and bacteriological cure rates (during treatment).

Reinfection rates after 1 month.

Rates of adverse events.

Trimethoprim compared with nitrofurantoin

Evidence on trimethoprim compared with nitrofurantoin for UTIs

There is limited evidence from four trials that trimethoprim and nitrofurantoin are equally effective in eradicating bacteriuria. Although both antibiotics were generally well tolerated, one double-blinded study reported a higher incidence of adverse effects (in particular, skin rashes) with trimethoprim. None of the trials used a 3-day regimen.

In one Norwegian double-blind trial (159 adults with urinary tract infection [UTI], 91% women), bacteriuria cure rates after 2 weeks were similar for people treated with trimethoprim (200 mg twice daily) or nitrofurantoin (50 mg four times daily) for 10 days [Sander et al, 1981]. More people in the trimethoprim group (26%) reported adverse effects than in the nitrofurantoin group (12%) (p < 0.01). Skin rashes were more common in the trimethoprim group (12%) compared with nitrofurantoin (1.3%) (p < 0.01). Nausea/vomiting was similar in both groups (8–9%).

Three other studies were identified. These were all unblinded and of poor methodological quality (both UK trials were funded by pharmaceutical companies).

One US trial in 322 female students compared trimethoprim 300 mg once daily for 7 days, trimethoprim 200 mg once daily for 10 days, and nitrofurantoin 100 mg four times daily for 10 days [Iravani et al, 1982].

One UK trial in 342 women compared Mictral® (containing nalidixic acid) three times daily, amoxicillin 250 mg three times daily, trimethoprim 200 mg twice daily, and nitrofurantoin 100 mg four times daily [Lightstone et al, 1988].

Another UK study in 538 women compared three antibiotics (each taken twice daily for 7 days): modified-release nitrofurantoin  100 mg, trimethoprim 200 mg, and co-trimoxazole 960 mg [Spencer et al, 1994].

In all three studies, there was no difference in bacteriuria cure rates (range 85–100%). The number of people reporting adverse effects was similar for trimethoprim and nitrofurantoin.

Nitrofurantoin formulations

Evidence on different nitrofurantoin formulations for UTIs

There is weak evidence that the macrocrystalline formulation of nitrofurantoin (e.g. Macrodantin®) causes less nausea than the crystalline formulation (e.g. Furadantin®) in people with urinary tract infection (UTI). No studies comparing the modified-release formulation (Macrobid®) to other formulations of nitrofurantoin were found.

CKS found one randomized double-blind trial which compared two different crystalline formulations of nitrofurantoin [Kalowski et al, 1974].

The trial randomized 88 women and 6 men with UTI (with nitrofurantoin-sensitive organisms) to treatment with nitrofurantoin 100 mg four times daily for 2 weeks; 46 people were given the crystalline formulation and 48 people the macrocrystalline formulation. Both formulations were given as identical capsules.

No statistically significant differences were found in terms of cure rate or the number of people who withdrew from the study due to adverse effects.

Adverse effects were assessed after 1 week of treatment, and then 1 week after cessation of treatment. More people reported adverse effects with crystalline nitrofurantoin (39%) than with macrocrystalline nitrofurantoin (17%) (p < 0.05). This was mainly due to the higher number of people reporting nausea (35% compared with 13%, p < 0.05).

The study had methodological weaknesses, such as:

Minimum planned sample size and statistical power was not reported.

No information was given on how the participants were randomized.

The magnitude of the reported adverse effects is unclear as these were classed only as mild or severe (requiring cessation of treatment).

Treatment was for 14 days (while current recommendations are to treat for 3 days).

Two other studies of poor methodological quality were identified.

One unblinded randomized study (method of randomization unclear, statistical power not reported) evaluated rates of nausea and vomiting in people prescribed macrocrystalline nitrofurantoin capsules (153 participants) and crystalline nitrofurantoin tablets (134 participants) [Hailey and Glascock, 1967].

Both formulations were given at a dose of 100 mg four times a day for 10 days (current recommendations are to treat for 3 days).

More people assigned to the crystalline formulation experienced nausea and vomiting compared with the group assigned to the macrocrystalline formulation: 14% compared with 8%. However, there was no statistically significant difference (p > 0.05) between groups.

This study was preceded by a pilot cohort of 112 people with a history of intolerance to crystalline nitrofurantoin. Open-label re-challenge with macrocrystalline nitrofurantoin found that 20% of this cohort also experienced nausea and vomiting with the macrocrystalline formulation [Hailey and Glascock, 1967].

A retrospective review of case records from 219 women with taking nitrofurantoin prophylaxis for recurrent UTI suggests that the rate of nausea is lower with the macrocrystalline formulation [Brumfitt and Hamilton-Miller, 1998]. Nausea was reported in 46.5% of women taking crystalline nitrofurantoin 50 mg twice a day (20 of 43 women), in 13.6% of women taking macrocrystalline nitrofurantoin 100 mg once a day (15 of 110 women), and in 12.1% of women taking macrocrystalline nitrofurantoin 50 mg once a day (8 of 66 women).

CKS found no studies that compared the immediate–release formulations with the modified-release formulation Macrobid®, which contains both macrocrystalline nitrofurantoin and nitrofurantoin monohydrate.

UTI during pregnancy

Evidence on UTI during pregnancy

Urine dipstick tests for screening

Evidence on urine dipstick tests for screening for asymptomatic bacteriuria in pregnancy

Observational studies of the sensitivity of dipstick tests for diagnosing asymptomatic bacteriuria in pregnancy have found a wide range of sensitivities. The most rigorous study found the sensitivity of 'either nitrite or leucocyte esterase (or both) positive' to be 53%.

Dipstick test (reagent strip) for screening for asymptomatic bacteriuria

A systematic review (search date: 2003, updated in 2008) was conducted by the National Institute for Health and Care Excellence (NICE) [NICE, 2008b].

The sensitivity of dipstick tests (using one, two, or all four of: nitrite, leucocyte esterase, blood, and protein) in five studies ranged from 8–50%. A sixth study found the sensitivity of dipstick test with either nitrite or leucocyte esterase (or both) positive to be 92%.

No meta-analysis was carried out, and confidence intervals were not reported, but NICE assumed in their economic analysis that the sensitivity of the dipstick test was 72%.

A subsequent Nigerian study in 400 women found similar low sensitivities for dipstick tests used alone or in selected combinations, for example 57% for leucocyte esterase alone (confidence intervals were not reported) [Eigbefoh et al, 2008].

A rigorous international study for the World Health Organization in 3048 women found the sensitivity of the dipstick test with 'either nitrite or leucocyte esterase (or both) positive' to be 53% (95% CI 48 to 58) [Mignini et al, 2009].

Antibiotic treatment of asymptomatic bacteriuria

Evidence on antibiotic treatment of asymptomatic bacteriuria in pregnancy

A Cochrane systematic review of controlled trials found that antibiotic treatment for asymptomatic bacteriuria in pregnancy can eradicate the infection, and can reduce the risks of pyelonephritis. About seven women need to be treated to prevent one episode of pyelonephritis. Treatment of asymptomatic bacteriuria reduced the incidence of pyelonephritis by about 75%. Evidence on the effect on birthweight and pre-term delivery is weak. The methodological quality of the included studies was generally poor. A Cochrane review found a lack of conclusive evidence about which antibiotic is the most effective and safe although this review included a large trial that supported the use of a 7-day regimen of nitrofurantoin.

The effect of antibiotic treatment on asymptomatic bacteriuria during pregnancy was examined in a Cochrane systematic review (search date: up to January 2007) [Smaill and Vazquez, 2007]. Fourteen randomized or quasi-randomized controlled trials compared antibiotic treatment with control (placebo or no treatment). The methodological quality (risk of bias) of the studies was generally poor. Most of the trials used continuous antibiotic treatment (n = 9) varying from a 3 week course to a course lasting until 6 weeks post-delivery. Nitrofurantoin was used in five studies (with three studies of ampicillin and one of penicillin). None of the studies used trimethoprim.

Eradication of bacteriuria

Antibiotic treatment was more effective than control treatment in eradicating asymptomatic bacteriuria (risk ratio [RR] 0.25, 95% CI 0.14 to 0.48; five studies, 793 participants). Two studies were undertaken using nitrofurantoin. There was significant statistical heterogeneity, possibly due to differences in study design and definition of persistent bacteriuria; the direction of effect was consistent.

Pyelonephritis

The incidence of pyelonephritis was reduced with antibiotic treatment when compared with control (RR 0.23, 95% CI 0.13 to 0.41; 11 studies, 1955 participants). The number of women needed to treat (NNT) to prevent one episode of pyelonephritis is seven (95% CI 6 to 8). Treatment of asymptomatic bacteriuria reduces the incidence of pyelonephritis by about 75%.

There was significant statistical heterogeneity. However, when only placebo-controlled studies were considered (seven studies, 1266 participants), there was no statistical heterogeneity, and antibiotic treatment was more effective than placebo (RR 0.17, 95% CI 0.09 to 0.31).

More recent studies found smaller treatment effects; this association could be due to the declining incidence of pyelonephritis, or to changes in obstetrical management.

Birthweight

The incidence of low birthweight (less than 2.5 kg) was reduced with antibiotic treatment (RR 0.66, 95% CI 0.49 to 0.89; seven studies, 1502 participants). There was significant statistical heterogeneity. However, when only placebo-controlled studies were considered (four studies, 689 participants), there was no statistical heterogeneity, and no statistically significant benefit on low birthweight (RR 0.64, 95% CI 0.35 to 1.16).

Gestational age

Antibiotic treatment had no statistically significant effect on reducing preterm delivery, defined as gestational age less than 38 weeks (RR 0.37, 95% CI 0.10 to 1.36; three studies, 412 participants).

Preventing recurrent UTI

Evidence on prophylactic treatments to prevent recurrent UTI

Prophylactic antibiotics

Evidence on regular prophylactic antibiotics for preventing recurrent UTI

There is consistent evidence from 10 randomized controlled trials (RCTs), all with methodological weaknesses, that regular antibiotics are more effective than placebo in preventing recurrent urinary tract infections (UTIs), but are associated with more adverse effects. The evidence from five RCTs is insufficient to prefer one antibiotic over another. The evidence on antibiotics compared with non-antibiotic drug treatment is poor.

A Cochrane systematic review (search date: April 2004) assessed the evidence on antibiotics for preventing recurrent UTI in non-pregnant women (pre- and postmenopausal) [Albert et al, 2004].

Risk of bias

Nineteen trials (published before 1996) met the inclusion criteria. Overall, these were of poor methodological quality, lacking any information on randomization and allocation concealment. Follow up and adverse events were poorly reported in some trials.

Antibiotic compared with placebo

Ten trials (430 participants) studied cefalexin, cinoxacin, cotrimoxazole, nitrofurantoin, and norfloxacin. Duration of treatment ranged from 6 months (eight studies) to 12 months (two studies).

During antibiotic treatment, recurrences of both bacteriuria and UTI symptoms (dysuria, increased frequency) were reduced compared with placebo. The number needed to be treated (NNT) for one extra woman to benefit with antibiotic treatment was two.

Microbiological UTIs: recurrence rates (per patient-year) were 0–0.9 episodes for the antibiotic group and 0.8–3.6 episodes for placebo. The relative risk (RR) of having at least one recurrence was 0.21 (95% CI 0.13 to 0.34; 11 studies, 372 participants), favouring antibiotic use; NNT 1.85 (95% CI 1.60 to 2.20).

Symptomatic UTIs: recurrence rates (per patient-year) were 0–0.27 episodes in the antibiotic group and 1.12–3.6 episodes for placebo: RR 0.15 (95% CI 0.08 to 0.28), favouring antibiotic use; NNT 2.2 (95% CI 1.80 to 2.80).

After the prophylaxis was stopped, limited data from three pooled studies (70 participants) showed that antibiotics do not prevent recurrent bacteriuria.

Bacteriuria: recurrence rates (per patient-year) were 1.2–1.3 episodes for the two antibiotic regimens, compared with 3 episodes with placebo (RR 0.82, 95% CI 0.44 to 1.53; three studies, 70 participants).

UTI symptoms: no trials were identified with this outcome.

Adverse effects (such as vaginal itching and nausea) were higher with antibiotic treatment. No differences in withdrawal and drop-outs were observed.

Severe adverse effects requiring treatment withdrawal: RR 1.58 (95% CI 0.47 to 5.28; 11 studies, 420 participants) — trend favoured placebo (but was not statistically significant).

Other adverse effects not requiring treatment withdrawal: RR 1.78 (95% CI 1.06 to 3.00; 11 studies, 420 participants), favouring placebo.

Antibiotic compared with another antibiotic

Five studies compared nitrofurantoin with cefaclor, norfloxacin, trimethoprim, or sulfamethoxazole. One trial compared trimethoprim with cinoxacin. Prophylaxis ranged from 6 months (three studies) to 12 months (three studies).

Outcomes

Five trials showed no clear benefit of one antibiotic over another.

One trial with 72 participants found that nitrofurantoin 100 mg daily reduced the recurrence of bacteriuria compared with trimethoprim 100 mg daily (RR 3.58, 95% CI 1.33 to 9.66), and reduced the recurrence of UTI symptoms (RR 1.72, 95% CI 1.06 to 2.79).

No other clinical outcomes were pooled due to significant heterogeneity between the trials and differences in outcome measures.

For adverse effects severe enough to stop treatment, when all the trials were pooled, these appeared to be higher with nitrofurantoin than with other antibiotics. However, if the above nitrofurantoin/trimethoprim trial was excluded, no statistical difference was found.

For other adverse effects, no difference was found, but significant heterogeneity existed between the trials.

Comparison of different regimens of the same antibiotics

Two studies with 513 participants were identified.

In one trial, weekly pefloxacin 400 mg was more effective than monthly pefloxacin 400 mg for preventing bacteriuria recurrence (RR 0.31, 95% CI 0.19 to 0.52).

No statistically significant difference was found for adverse events and for post-intervention follow up.

In the second trial involving 135 sexually active women with UTIs associated with sexual intercourse, no differences in the rate of UTIs and adverse events were found between postcoital ciprofloxacin and daily ciprofloxacin during and after the treatment period. For further information, see Post-coital antibiotics.

Antibiotic compared with other non-antibiotic drug treatment

Two unblinded studies with 177 participants were identified; treatment was given for 12 months.

One trial treated 20 people with trimethoprim 100 mg daily, 25 people with povidone iodine solution, and 19 people with methenamine 1 g daily. No clinically important differences were found between the treatment groups in terms of clinical and microbiological outcomes.

The other study compared nitrofurantoin 50 mg twice daily with methenamine hippurate 1 g twice daily (with crossovers). Small differences were found with recurrence rates (per patient-year) for bacteriuria, being 0.19 and 0.57 episodes respectively. More people reported severe adverse effects with nitrofurantoin (12/43) than with methenamine (2/56).

Post-coital antibiotics

Evidence on post-coital antibiotics for preventing UTI

There is limited evidence from small trials that post-coital antibiotics are more effective than placebo and as effective as continuous antibiotic treatment in preventing urinary tract infections (UTIs) associated with sexual intercourse.

A Cochrane systematic review (search date: April 2004) assessed antibiotics for preventing recurrent UTIs [Albert et al, 2004]. Two double-blind controlled trials on the use of post-coital antibiotics met the inclusion criteria.

Post-coital co-trimoxazole compared with placebo

One trial with 25 participants was identified by the review.

Co-trimoxazole 240 mg taken within 2 hours of intercourse was found to be more effective than a post-coital placebo [Stapleton et al, 1990]. Relative risks (RR) for a microbiological (bacteriuria) and clinical (symptomatic) recurrence were the same: RR 0.15 (95% CI 0.04 to 0.58), favouring antibiotic use. Adverse effects rates did not differ significantly.

Post-coital ciprofloxacin compared with long-term prophylactic ciprofloxacin

One study with 152 participants was identified by the review.

Over a study period of 12 months, post-coital ciprofloxacin 125 mg taken immediately after intercourse was as effective as daily ciprofloxacin 125 mg at bedtime [Melekos et al, 1997]. The relative risks for both microbiological and clinical recurrence were not statistically significant (RR 0.93 [95% CI 0.13 to 6.40], and RR 1.24 [95% CI 0.29 to 5.32], respectively). No statistically significant differences were found for adverse effects, or for the rate of reinfections after 12 months of treatment.

CKS found one subsequently published study on post-coital antibiotics [Stamatiou et al, 2005]. The interventions and results were too poorly reported to appraise the authors' claim that continuous prophylaxis with low-dose trimethoprim and post-coital prophylaxis with cefalexin 500 mg as a single dose are equally effective.

Evidence on changing the catheter before starting antibiotic treatment

Evidence on changing the catheter before starting antibiotic treatment

For women with indwelling urinary catheters and lower urinary tract infection (UTI), there is no evidence from controlled trials to guide the choice of antibiotic. Limited evidence from one small trial suggests that changing the catheter before starting antibiotic treatment improves the chance of cure.

Choice of antibiotic

CKS found no relevant controlled trials.

Replacement of indwelling urinary catheter before starting antibiotic treatment

A small, prospective, randomized, open-label clinical trial found that replacing indwelling urinary catheters before starting antibiotics for UTI significantly improves cure rates [Raz et al, 2000].

The study included 21 men and 33 women who were nursing home residents (mean 72.6 years of age), and had a chronic indwelling catheter (mean length of time in situ 30.8 days +/- 3.9 days) and a clinical diagnosis of symptomatic urinary tract infection.

People were initially treated with intravenous ciprofloxacin or ofloxacin. When they had been afebrile for 24 hours they were given oral ciprofloxacin or ofloxacin to complete 14 days' treatment.

Urine culture 72 hours after starting treatment was significantly more likely to find no uropathogens if the catheter had been replaced: people with catheter replacement, 89% (24 of 27); and without catheter replacement, 30% (8 of 27) (p = 0.001).

There was also a lower rate of symptomatic clinical relapse 28 days after treatment (p = 0.015).

Antibiotic resistance

Evidence on antibiotic resistance

Resistance and previous usage

Evidence on antibiotic resistance and previous antibiotic usage

In people with urinary tract infection (UTI), there is limited evidence to indicate an association between trimethoprim resistance and previous trimethoprim treatment within the past 12 months. Resistance may be more likely with more recent the exposure to trimethoprim, and with more frequent courses of trimethoprim. Studies on nitrofurantoin resistance and previous treatment were not found.

The relationship between antibiotic resistance in community-acquired UTI and risk factors was examined in a systematic review (search date: up to 2000) [Hillier et al, 2002]. Eight studies were identified; they had a lack of specificity in terms of definition and statistical power. The systematic review concluded that the evidence for recent antibiotic use and increased risk of infection with resistant UTI pathogens to be inconclusive.

Subsequently, several studies have been published, linking trimethoprim resistance (from urine culture data) with recent trimethoprim use. These results should be interpreted with caution — for example, due to sampling bias (as urine samples are more likely to be taken from people who have not responded to initial antibiotic data). Studies on nitrofurantoin were not found.

In a Tayside nested case-controlled study (expanded from an earlier study [Steinke et al, 1999]), multivariate analysis of multiple risk factors were undertaken, comparing 827 trimethoprim-resistant UTIs with 2608 trimethoprim-sensitive UTIs [Steinke et al, 2001]. Prior trimethoprim usage (past 180 days) was an important predictor for trimethoprim resistance: odds ratio (OR) 4.35 (95% CI 3.03 to 5.70). Usage of other antibiotics (past 180 days) was also an important predictor: OR 1.32 (95% CI 1.10 to 1.60).

A subsequent cross-sectional study (also in Tayside with 8833 cases), found trimethoprim resistance was statistically associated with prior exposure to trimethoprim: OR 1.22 (95% CI 1.16 to 1.28). It was also associated with use of other antibiotics: OR 1.18 (95% CI 1.06 to 1.32) [Donnan et al, 2004]. The association was more likely if trimethoprim was prescribed more recently. When prescribed 8–15 days before, the OR was 9.19 (95% CI 6.35 to 13.3). When prescribed 4–6 months before, the OR was 1.45 (95% CI 1.03 to 2.053). No association was found if trimethoprim was prescribed over 6 months ago.

An English study (based in Avon and Gloucestershire, 618 cases) found no significant association between antibiotic resistance (to either amoxicillin or trimethoprim or both) and exposure to any antibiotic in the previous months: adjusted OR 1.12 (95% CI 0.77 to 1.65, p = 0.52) [Hay et al, 2005]. Secondary analysis found that antibiotic resistance was associated with antibiotic use within the past 2 months: OR 1.95 (95% CI 1.08 to 3.49, p = 0.03). Antibiotic resistance was also associated with increased trimethoprim use in the past 12 months: OR 1.01 (95% CI 1.01 to 1.02, p = 0.001).

Another case-controlled study (based in South Wales) compared 154 people with Escherichia coli trimethoprim-resistant urine cultures with 489 controls, and found that trimethoprim-resistant UTIs were more likely if trimethoprim was prescribed in the previous year: OR 2.39 (95% CI 1.62 to 3.53) [Hillier et al, 2007]. Further analysis found the association was non-significant if trimethoprim was prescribed over 6 months ago. Other factors linked to trimethoprim resistance were number of trimethoprim prescriptions in the previous 12 months, with the OR increasing from 2.08 (95% CI 1.34 to 3.22) for one prescription to 7.53 (95% CI 2.71 to 20.88) for three or more prescriptions, and if treatment was 7 days or more: OR 4.62 (95% CI 2.73 to 7.82). A similar pattern was found with ampicillin resistance and amoxicillin prescribing.

A prospective cohort study assessed resistance patterns and associations in women presenting to general practitioner surgeries in Norwich and Gloucester with cystitis [McNulty et al, 2006]. There was a weak association between resistance and the number of episodes of cystitis: women who had four episodes of cystitis in the previous year were more likely to be infected with a resistant organism than women who had no episodes of cystitis: OR 9.5 (95% CI 1.5 to 60.1, p = 0.09). There was no evidence of an increased likelihood of a resistant pathogen from those with one to three previous UTIs in the preceding year compared with those with no previous UTIs in the same period. There was no association between resistance and the severity of any symptom or combination of symptoms recorded at enrolment, nor with age (data not shown).

Search strategy

Scope of search

A literature search was conducted for guidelines, systematic reviews and randomized controlled trials on primary care management of urinary tract infections (UTI) in women, with additional searches for evidence in the following areas:

Management of recurrent UTI in pregnant women

Management of recurrent UTI in women with indwelling catheters

Search dates

April 2009 - August 2013.

Key search terms

Various combinations of searches were carried out. The terms listed below are the core search terms that were used for Medline.

exp Urinary Tract Infections/, urinary tract infection$.tw., exp Cystitis/, cystitis.tw., (urinary adj infection$).tw., exp Bacteriuria/, bacteriuria.tw., UTI.tw.

recurrent.tw., exp Pregnancy/, exp Pregnant women/, pregnan$.tw.

recurrent.tw., exp Catheters/, catheter$.tw.

Table 1 . Key to search terms.
Search commands Explanation
/ indicates a MeSh subject heading with all subheadings selected
.tw. indicates a search for a term in the title or abstract
exp indicates that the MeSH subject heading was exploded to include the narrower, more specific terms beneath it in the MeSH tree
$ indicates that the search term was truncated (e.g. wart$ searches for wart and warts)
Topic specific literature search sources

European Association of Urology

Sources of guidelines

National Institute for Health and Care Excellence (NICE)

Scottish Intercollegiate Guidelines Network (SIGN)

Royal College of Physicians

Royal College of General Practitioners

Royal College of Nursing

NICE Evidence

Health Protection Agency

National Guidelines Clearinghouse

Guidelines International Network

TRIP database

GAIN

Institute for Clinical Systems Improvement

National Health and Medical Research Council (Australia)

Royal Australian College of General Practitioners

British Columbia Medical Association

Canadian Medical Association

Towards Optimal Practice

University of Michigan Medical School

Michigan Quality Improvement Consortium

Patient UK Guideline links

Driver and Vehicle Licensing Agency

Medline (with guideline filter)

Sources of systematic reviews and meta-analyses

The Cochrane Library :

Systematic reviews

Protocols

Database of Abstracts of Reviews of Effects

Medline (with systematic review filter)

EMBASE (with systematic review filter)

Sources of health technology assessments and economic appraisals

The Cochrane Library :

NHS Economic Evaluations

Health Technology Assessments

Canadian Agency for Drugs and Technologies in Health

International Network of Agencies for Health Technology Assessment

NIHR Health Technology Assessment programme

Sources of randomized controlled trials

The Cochrane Library :

Central Register of Controlled Trials

Medline (with randomized controlled trial filter)

EMBASE (with randomized controlled trial filter)

Sources of evidence based reviews and evidence summaries

Central Services Agency COMPASS Therapeutic Notes

Sources of national policy

Department of Health

Health Management Information Consortium (HMIC)

Patient experiences

Healthtalkonline

BMJ - Patient Journeys

Patient.co.uk - Patient Support Groups

Sources of medicines information

The following sources are used by CKS pharmacists and are not necessarily searched by CKS information specialists for all topics. Some of these resources are not freely available and require subscriptions to access content.

British National Formulary (BNF)

electronic Medicines Compendium (eMC)

European Medicines Agency (EMEA)

LactMed

Medicines and Healthcare products Regulatory Agency (MHRA)

REPROTOX

Scottish Medicines Consortium

Stockley's Drug Interactions

TERIS

TOXBASE

Micromedex

UK Medicines Information

References

ABPI Medicines Compendium (2011a) Summary of product characteristics for Keflex tablets, capsules and granules. Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2011b) Summary of product characteristics for Trimethoprim 50 mg/5 ml Suspension. Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2012a) Summary of product characteristics for Amoxil capsules 250mg. Electronic medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2012b) Summary of product characteristics for Macrobid Capsules 100mg B.P Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2012c) Summary of product characteristics for Trimethoprim Tablets BP 200mg. Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2012d) Summary of product characteristics for Furadantin 25mg/5ml Oral Suspension OR Nitrofurantoin 25mg/5ml Oral Suspension. Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2013a) Summary of product characteristics for Trimethoprim 200mg Tablets (Accord Healthcare Limited). Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2013b) Summary of product characteristics for Macrodantin Capsules 50mg B.P Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2013c) Summary of product characteristics for Macrodantin Capsules 100mg B.P Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

ABPI Medicines Compendium (2013d) Summary of product characteristics for Trimethoprim 200mg Tablets (Kent Pharmaceuticals Ltd). Electronic Medicines CompendiumDatapharm Communications Ltd. www.medicines.org.uk [Free Full-text]

Albert, X., Huertas, I., Pereiro, I. et al. (2004) Antibiotics for preventing recurrent urinary tract infection in non-pregnant women (Cochrane Review). The Cochrane Library. Issue 3. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

American College of Obstetricians and Gynecologists (2008) ACOG Practice Bulletin no. 91: treatment of urinary tract infections in nonpregnant women. Obstetrics and Gynecology 111(3), 785-794.

Aronson, J.K. (Ed.) (2006) Meyler's side effects of drugs. The international encyclopedia of adverse drug reactions and interactions. Volume 6: T-Z. 15th edn. Amsterdam: Elsevier.

Barrett, S.P., Savage, M.A., Rebec, M.P. et al. (1999) Antibiotic sensitivity of bacteria associated with community-acquired urinary tract infection in Britain. Journal of Antimicrobial Chemotherapy 44(3), 359-365. [Abstract] [Free Full-text]

Baxter, K. and Preston, C.L. (Eds.) (2013) Stockley's drug interactions: a source book of interactions, their mechanisms, clinical importance and management. 10th edn. London: Pharmaceutical Press.

Bland, J.M. and Altman, D.G. (1997) Cronbach's alpha. BMJ 314(7080), 572. [Free Full-text]

BNF 66 (2013) British National Formulary. 66th edn. London: British Medical Association and Royal Pharmaceutical Society of Great Britain.

Bruel, H., Guillemant, V., Saladin-Thiron, C. et al. (2000) Hemolytic anemia in a newborn after maternal treatment with nitrofurantoin at the end of pregnancy [French]. Archives de Pediatrie 7(7), 745-747. [Abstract]

Brumfitt, W. and Hamilton-Miller, J.M. (1998) Efficacy and safety profile of long-term nitrofurantoin in urinary infections: 18 years' experience. Journal of Antimicrobial Chemotherapy 42(3), 363-371. [Abstract] [Free Full-text]

COMPASS (2012) COMPASS therapeutic notes on the management of bacterial urinary tract infections in primary care. Northern Ireland Health and Social Care Board. www.medicinesni.com [Free Full-text]

CSM (2005) Updated advice on the safety of selective COX-2 inhibitors. Committee on Safety of Medicines. www.mhra.gov.uk [Free Full-text]

Dason, S., Dason, J.T. and Kapoor, A. (2011) Guidelines for the diagnosis and management of recurrent urinary tract infection in women. Canadian Urological Association Journal 5(5), 316-322. [Free Full-text]

Donnan, P.T., Wei, L., Steinke, D.T. et al. (2004) Presence of bacteriuria caused by trimethoprim resistant bacteria in patients prescribed antibiotics: multilevel mode with practice and individual patient data. British Medical Journal 328(7451), 1297-1300. [Abstract] [Free Full-text]

DTB (1998) Managing urinary tract infection in women. Drug & Therapeutics Bulletin 36(4), 30-32. [Abstract]

Eigbefoh, J.O., Isabu, P., Okpere, E. and Abebe, J. (2008) The diagnostic accuracy of the rapid dipstick test to predict asymptomatic urinary tract infection of pregnancy. Journal of Obstetrics and Gynaecology 28(5), 490-495. [Abstract]

Falagas, M.E., Kotsantis, I.K., Vouloumanou, E.K. and Rafailidis, P.I. (2009) Antibiotics versus placebo in the treatment of women with uncomplicated cystitis: a meta-analysis of randomized controlled trials. Journal of Infection 58(2), 91-102. [Abstract]

Fihn, S.D., Latham, R.H., Roberts, P. et al. (1985) Association between diaphragm use and urinary tract infection. Journal of the American Medical Association 254(2), 240-245. [Abstract]

Fihn, S.D., Boyko, E.J., Chen, C.L. et al. (1998) Use of spermicide-coated condoms and other risk factors for urinary tract infection caused by Staphylococcus saprophyticus. Archives of Internal Medicine 158(3), 281-287. [Abstract] [Free Full-text]

Forna, F., McConnell, M., Kitabire, F.N. et al. (2006) Systematic review of the safety of trimethoprim-sulfamethoxazole for prophylaxis in HIV-infected pregnant women: implications for resource-limited settings. AIDS Reviews 8(1), 24-36. [Abstract]

Foxman, B. (2002) Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. American Journal of Medicine 113(Suppl 1A), 5S-13S. [Abstract]

FSRH (2011) Drug interactions with hormonal contraception. Faculty of Sexual and Reproductive Healthcare. www.fsrh.org [Free Full-text]

Grabe, M., Bjerklund-Johansen, T.E., Botto, H. et al. (2013) Guidelines on urological infections. European Association of Urology. www.uroweb.org [Free Full-text]

Gratacós, E., Torres, P.J., Vila, J. et al. (1994) Screening and treatment of asymptomatic bacteriuria in pregnancy prevent pyelonephritis. Journal of Infectious Diseases 169(6), 1390-1392. [Abstract]

Hailey, F.J. and Glascock, H.W. (1967) Gastrointestinal tolerance to a new macrocrystalline form of nitrofurantoin: a collaborative study. Current Therapeutic Research, Clinical and Experimental 9(12), 600-605.

Harris, N., Teo, R., Mayne, C. and Tincello, D. (2008) Recurrent urinary tract infection in gynaecological practice. Obstetrician & Gynaecologist 10(1), 17-21. [Free Full-text]

Hay, A.D., Thomas, M., Montgomery, A. et al. (2005) The relationship between primary care antibiotic prescribing and bacterial resistance in adults in the community: a controlled observational study using individual patient data. Journal of Antimicrobial Chemotherapy 56(1), 146-153. [Abstract] [Free Full-text]

Hernández-Díaz, S., Werler, M.M., Walker, A.M. and Mitchell, A.A. (2001) Neural tube defects in relation to use of folic acid antagonists during pregnancy. American Journal of Epidemiology 153(10), 961-968. [Abstract] [Free Full-text]

Hillier, S.L., Magee, J.T., Howard, A.J. and Palmer, S.R. (2002) How strong is the evidence that antibiotic use is a risk factor for antibiotic-resistant, community-acquired urinary tract infection? Journal of Antimicrobial Chemotherapy 50(2), 241-247. [Abstract] [Free Full-text]

Hillier, S., Roberts, Z., Dunstan, F. et al. (2007) Prior antibiotics and risk of antibiotic-resistant community-acquired urinary tract infection: a case-control study. Journal of Antimicrobial Chemotherapy 60(1), 92-99. [Abstract] [Free Full-text]

Hooton, T.M., Bradley, S.F., Cardenas, D.D. et al. (2010) Diagnosis, prevention, and treatment of catheter-associated urinary tract infection in adults: 2009 International Clinical Practice Guidelines from the Infectious Diseases Society of America. Clinical Infectious Diseases 50(5), 625-663. [Abstract] [Free Full-text]

Howard, A.J., Magee, J.T., Fitzgerald, K.A. et al. (2001) Factors associated with antibiotic resistance in coliform organisms from community urinary tract infection in Wales. Journal of Antimicrobial Chemotherapy 47(3), 305-313. [Abstract] [Free Full-text]

HPA (2011) Diagnosis of UTI: quick reference guide for primary care for consultation and local adaptation. Health Protection Agency. www.hpa.org.uk [Free Full-text]

HPA and Association of Medical Microbiologists (2008) Management of infection guidance for primary care for consultation and local adaptation. Health Protection Agency. www.hpa.org.uk [Free Full-text]

HPA and Association of Medical Microbiologists (2010) Management of infection guidance for primary care for consultation and local adaptation. Health Protection Agency. www.hpa.org.uk [Free Full-text]

HPA and British Infection Association (2013) Management of infection guidance for primary care for consultation and local adaptation. Health Protection Agency. www.hpa.org.uk [Free Full-text]

Hummers-Pradier, E. and Kochen, M.M. (2002) Urinary tract infections in adult general practice patients. British Journal of General Practice 52(482), 752-761. [Abstract] [Free Full-text]

Iravani, A., Richard, G.A. and Baer, H. (1982) Trimethoprim once daily vs nitrofurantoin in treatment of acute urinary tract infections in young women, with special reference to periurethral, vaginal, and fecal flora. Reviews of Infectious Diseases 4(2), 378-387. [Abstract]

Jepson, R.G., Williams, G. and Craig, J.C. (2012) Cranberries for preventing urinary tract infections (Cochrane Review). The Cochrane Library. Issue 10. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Kahlmeter, G. (2003) An international survey of the antimicrobial susceptibility of pathogens from uncomplicated urinary tract infections: the ECO.SENS Project. Journal of Antimicrobial Chemotherapy 51(1), 69-76. [Abstract] [Free Full-text]

Kalowski, S., Radford, N. and Kincaid-Smith, P. (1974) Crystalline and macrocrystalline nitrofurantoin in the treatment of urinary-tract infection. New England Journal of Medicine 290(7), 385-387.

Lawrenson, R.A. and Logie, J.W. (2001) Antibiotic failure in the treatment of urinary tract infections in young women. Journal of Antimicrobial Chemotherapy 48(6), 895-901. [Abstract] [Free Full-text]

Lee, B.S.B., Bhuta, T., Simpson, J.M. and Craig, J.C. (2012) Methenamine hippurate for preventing urinary tract infections (Cochrane Review). The Cochrane Library. Issue 10. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Le, J., Briggs, G.G., McKeown, A. and Bustillo, G. (2004) Urinary tract infections during pregnancy. Annals of Pharmacotherapy 38(10), 1692-1701. [Abstract]

Lightstone, B.L., Greaves, W.E. and Humphries, J.M. (1988) Comparison of Mictral with amoxycillin, trimethoprim and nitrofurantoin in the treatment of acute cystitis. British Journal of Clinical Practice 42(7), 283-288.

Little, P., Turner, S., Rumsby, K. et al. (2009) Dipsticks and diagnostic algorithms in urinary tract infection: development and validation, randomised trial, economic analysis, observational cohort and qualitative study. Health Technology Assessment 13(19), 1-96. [Abstract] [Free Full-text]

Loeb, M., Brazil, K., Lohfeld, L. et al. (2005) Effect of a multifaceted intervention on number of antimicrobial prescriptions for suspected urinary tract infections in residents of nursing homes: cluster randomised controlled trial. BMJ 331(7518), 669. [Abstract] [Free Full-text]

Lumbiganon, P., Villar, J., Laopaiboon, M. et al. (2009) One-day compared with 7-day nitrofurantoin for asymptomatic bacteriuria in pregnancy: a randomized controlled trial. Obstetrics and Gynecology 113(2 Pt 1), 339-345. [Abstract]

Lutters, M. and Vogt-Ferrier, N. (2008) Antibiotic duration for treating uncomplicated, symptomatic lower urinary tract infections in elderly women (Cochrane Review). The Cochrane Library. Issue 3. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Magee, J.T., Pritchard, E.L., Fitzgerald, K.A. et al. (1999) Antibiotic prescribing and antibiotic resistance in community practice: retrospective study 1996-98. British Medical Journal 319(7219), 1239-1240. [Free Full-text]

McNulty, C.A., Richards, J., Livermore, D.M. et al. (2006) Clinical relevance of laboratory-reported antibiotic resistance in acute uncomplicated urinary tract infection in primary care. Journal of Antimicrobial Chemotherapy 58(5), 1000-1008. [Abstract] [Free Full-text]

Melekos, M.D., Asbach, H.W., Gerharz, E. et al. (1997) Post-intercourse versus daily ciprofloxacin prophylaxis for recurrent urinary tract infections in premenopausal women. Journal of Urology 157(3), 935-939. [Abstract]

MeReC (1995) Urinary tract infection. MeReC Bulletin 6(8), 1-7.

MHRA (2013) Nitrofurantoin: reminder on precautions for use, especially renal impairment in (elderly) patients. Drug Safety Update 7(1), A3. [Free Full-text]

Mignini, L., Carroli, G., Abalos, E. et al. (2009) Accuracy of diagnostic tests to detect asymptomatic bacteriuria during pregnancy. Obstetrics and Gynecology 113(2 Pt 1), 346-352. [Abstract]

Milo, G., Katchman, E.A., Paul, M. et al. (2005) Duration of antibacterial treatment for uncomplicated urinary tract infection in women (Cochrane Review). The Cochrane Library. Issue 2. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

National Clinical Guideline Centre (2012) Infection: prevention and control of healthcare-associated infections in primary and community care (full NICE guideline). Clinical guidance 139National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

National Collaborating Centre for Primary Care (2005) Referral guidelines for suspected cancer in adults and children. Part one: chapters 1-12 (full NICE guideline). . Clinical guideline 27. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

Newell, A., Bunting, P., Anson, K. et al. (2005) Multicentre audit of the treatment of uncomplicated urinary tract infection in South Thames. International Journal of STD & AIDS 16(1), 74-77. [Abstract]

NICE (2005) Referral for suspected cancer (NICE guideline). . Clinical guideline 27. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2008a) Chronic kidney disease: early identification and management of chronic kidney disease in adults in primary and secondary care (NICE guideline). Clinical guidance 73National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2008b) Antenatal care: routine care for the healthy pregnant woman (NICE guideline). . Clinical guideline 62. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2008c) Prophylaxis against infective endocarditis: antimicrobial prophylaxis against infective endocarditis in adults and children undergoing interventional procedures (NICE guideline). . Clinical guideline 64. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2008d) Osteoarthritis. The care and management of osteoarthritis in adults (NICE guideline) [Replaced by CG177]. . Clinical guideline 59. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2009a) Rheumatoid arthritis: the management of rheumatoid arthritis (NICE guideline). . Clinical guideline 79. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2009b) Low back pain: early management of persistent non-specific low back pain (NICE guideline). . Clinical guideline 88. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

NICE (2013) Key therapeutic topics - medicines management options for local implementation. National Institute for Health and Care Excellence. www.nice.org.uk [Free Full-text]

Nickel, J.C., Lee, J.C., Grantmyre, J.E. and Polygenis, D. (2005) Natural history of urinary tract infection in a primary care environment in Canada. Canadian Journal of Urology 12(4), 2728-2737. [Abstract]

NPC (2011) Key therapeutic topics 2010/11 - Medicines management options for local implementation. National Prescribing Centre. www.npc.nhs.uk [Free Full-text]

NPC (2012) Key therapeutic topics - medicines management options for local implementation. National Prescribing Centre. www.npc.nhs.uk [Free Full-text]

Parsons, M. and Toozs-Hobson, P. (2005) The investigation and management of interstitial cystitis. Journal of the British Menopause Society 11(4), 132-139. [Abstract]

Perrotta, C., Aznar, M., Mejia, R. et al. (2008) Oestrogens for preventing recurrent urinary tract infection in postmenopausal women (Cochrane Review). The Cochrane Library. Issue 2. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Pohl, A. (2007) Modes of administration of antibiotics for symptomatic severe urinary tract infections (Cochrane Review). The Cochrane Library. Issue 4. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Priest, P., Yudkin, P., McNulty, C. and Mant, D. (2001) Antibacterial prescribing and antibacterial resistance in English general practice: cross sectional study. British Medical Journal 323(7320), 1037-1041. [Abstract] [Free Full-text]

Ramakrishnan, K. and Scheid, D.C. (2005) Diagnosis and management of acute pyelonephritis in adults. American Family Physician 71(5), 933-942. [Abstract] [Free Full-text]

Raz, R., Schiller, D. and Nicolle, L.E. (2000) Chronic indwelling catheter replacement before antimicrobial therapy for symptomatic urinary tract infection. Journal of Urology 164(4), 1254-1258. [Abstract]

RCOG (2012) The prevention of early-onset neonatal group B streptococcal disease. . 2nd edn. Royal College of Obstetricians and Gynaecologists. www.rcog.org.uk [Free Full-text]

Renal Association and British Association of Urological Surgeons (2008) Joint consensus statement on the initial assessment of haematuria. Renal Association and British Association of Urological Surgeons. www.baus.org.uk [Free Full-text]

Richards, D., Toop, L., Chambers, S. and Fletcher, L. (2005) Response to antibiotics of women with symptoms of urinary tract infection but negative dipstick urine test results: double blind randomised controlled trial. British Medical Journal 331(7509), 143. [Abstract] [Free Full-text]

Rouphael, N.G., O'Donnell, J.A., Bhatnagar, J. et al. (2008) Clostridium difficile-associated diarrhea: an emerging threat to pregnant women. American Journal of Obstetrics and Gynecology 198(6), 635.e1-635.e6. [Abstract]

Ruxton, C.H.S. and Derbyshire, E. (2010) Women's diet quality in the UK. Nutrition Bulletin 35(2), 126-137.

Sandegren, L., Lindqvist, A., Kahlmeter, G. and Andersson, D.I. (2008) Nitrofurantoin resistance mechanism and fitness cost in Eschericia coli. Journal of Antimicrobial Resistance 62(3), 495-503. [Abstract] [Free Full-text]

Sander, J., Aandahl, E., Fellner, H. and Kalstad, S. (1981) The treatment of urinary tract infections in out-patients: a double-blind comparison between trimethoprim and nitrofurantoin. Journal of International Medical Research 9(3), 181-185. [Abstract]

Sanderson, P.J. (1998) Laboratory methods. In: Brumfitt, W., Hamilton-Miller, J. and Bailey, R.R. (Eds.) Urinary tract infections. London: Chapman & Hall Medical.

Schaefer, C., Peters, P. and Miller, R.K. (Eds.) (2007) Drugs during pregnancy and lactation: treatment options and risk assessment. 2nd edn. Oxford: Academic Press.

SIGN (2012) Management of suspected bacterial urinary tract infection in adults: a national clinical guideline. Scottish Intercollegiate Guidelines Network. www.sign.ac.uk [Free Full-text]

Simerville, J.A., Maxted, W.C. and Pahira, J.J. (2005) Urinalysis: a comprehensive review. American Family Physician 71(6), 1153-1162. [Abstract] [Free Full-text]

Smaill, F.M. and Vazquez, J.C. (2007) Antibiotics for asymptomatic bacteriuria in pregnancy (Cochrane Review). The Cochrane Library. Issue 2. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Sobel, J.D. and Kaye, D. (2010) Urinary tract infections. In: Mandell, G.L., Bennett, J.E. and Dolin, R. (Eds.) Mandell, Douglas, and Bennett's principles and practice of infectious diseases. 7th edn. Philadelphia: Churchill Livingstone. 957-985.

SOGC (2010) SOGC clinical practice guideline. Recurrent urinary tract infection. Society of Obstetricians and Gynaecologists of Canada. www.sogc.org [Free Full-text]

Spencer, R.C., Moseley, D.J. and Greensmith, M.J. (1994) Nitrofurantoin modified release versus trimethoprim or co-trimoxazole in the treatment of uncomplicated urinary tract infection in general practice. Journal of Antimicrobial Chemotherapy 33(Suppl A), 121-129. [Abstract]

Stamatiou, C., Petrakos, G., Bovis, C. et al. (2005) Efficacy of prophylaxis in women with sex induced cystitis. Clinical and Experimental Obstetrics & Gynecology 32(3), 193-195. [Abstract]

Stapleton, A., Latham, R.H., Johnson, C. and Stamm, W.E. (1990) Postcoital antimicrobial prophylaxis for recurrent urinary tract infection. A randomized, double-blind, placebo-controlled trial. Journal of the American Medical Association 264(6), 703-706. [Abstract]

Steinke, D.T., Seaton, R.A., Phillips, G. et al. (1999) Factors associated with trimethoprim-resistant bacteria isolated from urine samples. Journal of Antimicrobial Chemotherapy 43(6), 841-843. [Abstract] [Free Full-text]

Steinke, D.T., Seaton, R.A., Phillips, G. et al. (2001) Prior trimethoprim use and trimethoprim-resistant urinary tract infection: a nested case-control study with multivariate analysis for other risk factors. Journal of Antimicrobial Chemotherapy 47(6), 781-787. [Abstract] [Free Full-text]

Tenke, P., Kovacs, B., Bjerklund Johansen, T.E. et al. (2008) European and Asian guidelines on management and prevention of catheter-associated urinary tract infections. International Journal of Antimicrobial Agents 31(Suppl 1), S68-S78. [Abstract]

Tomson, C. and Armitage, A. (2010) Urinary tract infection. In: Warrell, D.A., Cox, T.M. and Firth, J.D. (Eds.) Oxford textbook of medicine. 5th edn. Oxford: Oxford University Press. 4103-4122.

UKTIS (2012a) Use of cephalosporins in pregnancy. TOXBASEUK Teratology Information Service. www.toxbase.org

UKTIS (2012b) Use of nitrofurantoin in pregnancy. TOXBASEUK Teratology Information Service. www.toxbase.org

UKTIS (2012c) Use of amoxicillin in pregnancy. TOXBASEUK Teratology Information Service. www.toxbase.org

UKTIS (2013) Use of trimethoprim in pregnancy. TOXBASEUK Teratology Information Service. www.toxbase.org

Vahid, B. and Wildemore, B.M.M. (2006) Nitrofurantoin pulmonary toxicity: a brief review. Internet Journal of Pulmonary Medicine 6(2), . [Free Full-text]

Vazquez, J.C. and Abalos, E. (2011) Treatments for symptomatic urinary tract infections during pregnancy (Cochrane Review). The Cochrane Library. Issue 1. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]

Vogel, T., Verreault, R., Gourdeau, M. et al. (2004) Optimal duration of antibiotic therapy for uncomplicated urinary tract infection in older women: a double blind randomized controlled trial. Canadian Medical Association Journal 170(4), 469-473. [Abstract] [Free Full-text]

Wazait, H.D., Patel, H.R., Veer, V. et al. (2003) Catheter-associated urinary tract infections: prevalence of uropathogens and pattern of antimicrobial resistance in a UK hospital (1996-2001). BJU international 91(9), 806-809. [Abstract]

Widmer, M., Gülmezoglu, A.M., Mignini, L. and Roganti, A. (2011) Duration of treatment for asymptomatic bacteriuria during pregnancy (Cochrane Review). The Cochrane Library. Issue 12. John Wiley & Sons, Ltd. www.thecochranelibrary.com [Free Full-text]