|Home | About | Journals | Submit | Contact Us | Français|
We would like to acknowledge the previous contributors of this review, including Lisa Chew, Bruce Cooper, Stephan Finn, Ruth Jepson, Adriana Wechsler, and Bazian Ltd.
Cystitis is a bacterial infection of the lower urinary tract which causes pain when passing urine, and causes urgency, haematuria, and suprapubic pain not associated with passing urine. Recurrent cystitis is usually defined as three episodes of urinary tract infection in the previous 12 months, or two episodes in the previous 6 months.
We conducted a systematic review and aimed to answer the following clinical question: Which interventions prevent further recurrence of cystitis in women experiencing at least two infections per year? We searched: Medline, Embase, The Cochrane Library, and other important databases up to April 2007 (BMJ Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).
We found 14 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.
In this systematic review we present information relating to the effectiveness and safety of the following interventions: continuous antibiotic prophylaxis (trimethoprim, co-trimoxazole, nitrofurantoin, cefaclor, or a quinolone or cephalexin); continuous prophylaxis with methenamine hippurate; cranberry juice and cranberry products; oestrogen (topical) in postmenopausal women; passing urine after intercourse; postcoital antibiotic prophylaxis; single-dose self-administered antibiotic.
Cystitis is a bacterial infection of the lower urinary tract which causes pain when passing urine, and causes frequency, urgency, haematuria, and suprapubic pain not associated with passing urine.
Continuous antibiotic prophylaxis lasting 6–12 months reduces the rate of recurrence, although there is no consensus about when to start the treatment, or about how long it should last.
Postcoital antibiotics (taken within 2 hours of intercourse) reduce the rate of clinical recurrence of cystitis as effectively as continuous treatment.
We don't know whether single-dose self-administered trimethoprim–sulfamethoxazole or continuous prophylaxis with methenamine hippurate are effective in preventing recurrence of cystitis, as the studies were too small to show any clinically relevant differences.
Cranberry products (either juice or capsules) seem to significantly reduce the recurrence of symptomatic cystitis.
There is no evidence examining whether passing urine after intercourse is effective at preventing UTI.
We found insufficient evidence on the effects of topical oestrogen in postmenopausal women in the prophylaxis of recurrent cystitis.
In most cases, cystitis is a bacterial infection of the lower urinary tract which causes pain when passing urine, and causes frequency, urgency, haematuria, and suprapubic pain not associated with passing urine. White blood cells and bacteria are almost always present in the urine. A recurrent UTI is a symptomatic UTI that follows clinical resolution of an earlier infection generally, but not necessarily, after treatment. Recurrent cystitis is usually defined in the literature as three episodes of UTI in the previous 12 months or two episodes in the previous 6 months. Recurrent UTIs cause serious discomfort to women, and have a high impact on ambulatory healthcare costs, through outpatient visits, diagnostic tests, and prescriptions.
Recurrent cystitis is common among young, healthy women, even though they generally have anatomically and physiologically normal urinary tracts. One study found that nearly half of the women whose uncomplicated UTIs resolved spontaneously developed a recurrent UTI within a year. In a study of college women with their first UTI, 27% experienced at least one culture-confirmed recurrence within 6 months of the initial infection, and 2.7% had a second recurrence during this period. In a Finnish study of women aged 17–82 years who had Escherichia Coli cystitis, 44% had a recurrence within 1 year (53% in women older than 55 years, 36% in younger women). No large population-based studies have been done to determine proportionately how many women with UTI develop a pattern of high-frequency recurrence. Occasionally, recurrences are due to a persistent focus of infection, but the vast majority is thought to represent reinfection. A recurrence is defined clinically as a relapse if it is caused by the same species as caused the original UTI, and if it occurs within 2 weeks after treatment. It is considered reinfection if it occurs more than 2 weeks after treatment of the original infection. Most women are able to diagnose their own episodes of recurrent cystitis from symptoms (positive predictive value in one RCT 92%).
Cystitis is caused by uropathogenic bacteria in the faecal flora, that colonise the vaginal and periurethral openings and ascend the urethra into the bladder. Sexual intercourse, diaphragm–spermicide use, and a history of recurrent UTI have been shown to be strong and independent risk factors for cystitis. Use of spermicide-coated condoms may also increase the risk of UTI. Antimicrobial use has been shown to adversely affect the vaginal flora in animals and humans, and recent use of antibiotics is strongly associated with risk of cystitis. However, risk factors specific to women with recurrent cystitis have received little study. In a large, case controlled study of women with and without a history of recurrent UTI, comprising 229 cases and 253 controls, the strongest risk factor for recurrence in a multivariate analysis was the frequency of sexual intercourse. Other risk factors included spermicide use in the past year, new sex partner during the past year, having a first UTI at or before 15 years of age, and having a mother with history of UTI. Urine-voiding disorders, such as those associated with prolapse, multiple sclerosis, bladder cancer, or bladder stones, are also associated with increased risk. An association has been found with pre- and postcoital voiding, frequency of urination, delayed voiding habits, douching, and BMI. A possible association between smoking (which is strongly associated with bladder cancer) and recurrent cystitis has not been assessed. These behavioural patterns have never been evaluated in prospective, randomised trials. Data suggest that pelvic anatomical differences may have a role in predisposing some young women to recurrent UTI, especially those without other risk factors. In postmenopausal women, reduced oestrogen levels seem to contribute to recurrent cystitis in healthy women. The vagina, bladder, and urethra respond to oestrogen, and when the hormonal level in the body is reduced, the tissues of these organs become thinner, weaker, and dry. The changes in the tissues of the bladder and urethra, and the associated loss of protection against infection-causing germs, may increase the risk of UTI in postmenopausal women. Cystitis is also more common during pregnancy because of changes in the urinary tract. As the uterus grows, its increased weight can block the drainage of urine from the bladder, causing an infection. Women are at increased risk for recurrent cystitis from weeks 6–24 of pregnancy.
We found little evidence on the long-term effects of untreated cystitis. One study found that progression to pyelonephritis was infrequent, and that most cases of cystitis regressed spontaneously, although symptoms sometimes persisted for several months. However, bacteriuria in pregnant women carries a much greater risk of progressing to pyelonephritis than in non-pregnant women (28% v 1%), and is associated with serious risks.
To prevent recurrent cystitis in women with three episodes of UTI in the past 12 months or two episodes in the past 6 months, with minimal adverse effects of treatment. A recurrence is defined clinically as a relapse if it is caused by the same species which caused the original UTI, and if it occurs within 2 weeks after treatment. It is considered reinfection if it occurs more than 2 weeks after treatment of the original infection. We aim to consider both.
Rates of recurrent infection, based on symptoms and urine culture, quality of life, adverse effects of treatment.
BMJ Clinical Evidence search and appraisal April 2007. The following databases were used to identify studies for this systematic review: Medline 1966 to April 2007, Embase 1980 to April 2007, and The Cochrane Library (all databases) 2007, Issue 1. Additional searches used these websites: NHS Centre for Reviews and Dissemination (CRD) — for all databases, Turning Research into Practice (TRIP) and NICE. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the author for additional assessment, using pre-determined criteria to identify relevant studies. Study design criteria for evaluation in this review were: published systematic reviews and RCTs in any language, at least single blinded, and containing at least 20 individuals of whom more than 80% were followed up. There was no minimum length of follow-up required to include studies. We excluded all studies described as “open”, “open label”, or not blinded unless blinding was impossible. The author also did his own supplementary internet search on the terms "recurrent cystitis" and "meta-analysis" in the Google search engine. In addition, we use a regular surveillance protocol to capture harms alerts from organisations such as the FDA and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. We reviewed all systematic reviews and RCTs comparing different forms of prophylaxis, or comparing prophylaxis versus placebo in non-pregnant women with a history of recurrent cystitis. We excluded studies in populations consisting mainly of men or pregnant women. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ).
The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.
Pyelonephritis (acute) in non-pregnant women
RECURRENCE OF INFECTIONS Compared with placebo: Continuous antibiotic prophylaxis for 6–12 months may be more effective at reducing the proportion of women with clinical and microbiological recurrence during prophylaxis ( low-quality evidence ). Different continuous antibiotics for prophylaxis compared with each other: We don't know whether any one continuous antibiotic prophylaxis regimen is more effective than any other ( very low-quality evidence ). Compared with continuous methenamine hippurate: We don’t know whether antibiotics are more effective at reducing the proportion of women with at least one clinical or microbiological recurrence at 1 year (very low-quality evidence). Compared with postcoital prophylaxis: We don’t know whether taking ciprofloxacin daily is more effective than postcoital ciprofloxacin at reducing the number of microbiological recurrences (low-quality evidence). Compared with oestrogen (topical) in postmenopausal women: We don’t know whether antibiotics are more effective at reducing the proportion of women free of bacteriuria (symptomatic or asymptomatic) at 3–9 months (very low-quality evidence). NOTE There is no clear consensus about when to start continuous antibiotic prophylaxis, and about how long it should continue.
We found one systematic review (search date 2005, 10 RCTs, 430 non-pregnant women aged over 14 years with at least 2 episodes of uncomplicated cystitis in the past year) assessing continuous antibiotic prophylaxis for 2 months (2 RCTs) or 6 months (8 RCTs). The antibiotics given were: cefalexin 125 mg daily, trimethoprim–sulfamethoxazole (TMP-SMX, co-trimoxazole) 40–200 mg daily, nitrofurantoin 50–100 mg daily, a quinolone (norfloxacin 200 mg daily, cinoxacin 250–500 mg daily, or ciprofloxacin 500 mg daily), and postcoital TMP-SMX 40–200 mg daily (see comment below). The review found that, compared with placebo, continuous antibiotic prophylaxis significantly reduced the proportion of women with clinical recurrence during prophylaxis over 2–6 months (7 RCTs, 10/136 [13%] with antibiotics v 62/121 [51%] with placebo; RR of at least 1 recurrence 0.15, 95% CI 0.08 to 0.28; NNT 3, 95% CI 2 to 3). There were fewer clinical recurrences per person per year with continuous antibiotic prophylaxis (4 RCTs, 136 women: 0–0.27 infections/person-year with antibiotics v 1.12–3.6 infections/person-year with placebo). The review also found that, compared with placebo, continuous antibiotic prophylaxis significantly reduced the proportion of women with microbiological recurrence during prophylaxis (10 RCTs, 24/195 [12%] with antibiotics v 116/177 [65%] with placebo; RR of at least 1 recurrence 0.21, 95% CI 0.13 to 0.33; NNT 2, 95% CI 2 to 3). There were fewer microbiological recurrences per person-year with continuous antibiotic prophylaxis than with placebo (8 RCTs: 0–0.9 recurrences per person-year with antibiotics v 0.8–3.6 infections per person-year with placebo). The review found no significant difference in microbiological recurrence at 6 months between continuous antibiotic prophylaxis (cinoxacin, TMP-SMX, or nitrofurantoin) and placebo (23/44 [52%] with antibiotics v 15/26 [58%] with placebo; RR 0.82, 95% CI 0.44 to 1.53). Neither RCT assessed clinical recurrence after prophylaxis.
In a subgroup analysis performed in one of the RCTs identified by the review, women were stratified into three groups according to their intercourse frequency (twice a week or less, 2–3 times a week, and 3 times a week or more) and infection rates for each group were calculated. The analysis found that an increase in intercourse frequency was associated with a slight increase in microbiological recurrence rate; in women receiving placebo, the increase was significant; P = 0.004).
The review identified six poor-quality RCTs (458 women), three of which were open label. None of the trials assessed the same comparison. Five RCTs compared nitrofurantoin versus another antibiotic: cefaclor 250 mg daily, norfloxacin 200 mg or 400 mg daily, trimethoprim 40 mg or 100 mg daily, or sulphamethoxazole 200 mg daily. The sixth compared trimethoprim versus cinoxacin. There was significant heterogeneity among the trials and differences in the way clinical outcomes were assessed, so data could not be pooled. The individual results of these RCTs did not show a clear benefit of one antibiotic over another.
The review identified two RCTs (177 women). The first RCT (67 women) compared three interventions: trimethoprim 100 mg daily, methenamine hippurate 1 g twice daily, and povidone iodine solution, and found no significant difference between trimethoprim and methenamine hippurate in the proportion of women who had at least one clinical or microbiological recurrence over 1 year (clinical recurrence: 11/20 [55%] with trimethoprim v 18/25 [72%] with methenamine hippurate; RR 0.76, 95% CI 0.48 to 1.22; microbiological recurrence: 8/20 [40%] with trimethoprim v 10/25 [40%] with methenamine hippurate; RR 1.00, 95% CI 0.49 to 2.05). The second RCT found that nitrofurantoin 50 mg twice daily significantly reduced clinical and microbiological recurrence over 1 year compared with methenamine hippurate 1 g twice daily (clinical recurrence: 18/43 [42%] with nitrofurantoin v 41/56 [73%] with methenamine hippurate; RR 0.57, 95% CI 0.39 to 0.84; microbiological recurrence: 4/43 [9%] with nitrofurantoin v 19/56 [34%] with methenamine hippurate; RR 0.27, 95% CI 0.10 to 0.75).
See benefits of postcoital prophylaxis.
See benefits of oestrogen (topical) in postmenopausal women.
The review found that proportionately more women taking antibiotics than taking placebo stopped treatment owing to adverse effects, although the difference was not significant (9/225 [4%] with antibiotics v 4/195 [2%] with placebo; RR 1.58, 95% CI 0.47 to 5.28). The proportion of women who stopped treatment was small. The most common adverse effects causing women to stop treatment were skin rash and severe nausea. The review also found that, compared with placebo, antibiotics significantly increased the proportion of women with less-severe adverse effects, such as vaginal itching and mild nausea (34/225 [15%] v 15/195 [8%]; RR 2.36, 95% CI 1.22 to 4.54).
The first RCT found that four women taking methenamine hippurate stopped treatment owing to adverse effects, including nausea and vomiting, indigestion, and vulval rash; no-one taking trimethoprim withdrew from treatment. The second RCT found that 12 women taking nitrofurantoin and eight women taking methenamine hippurate had adverse effects, including nausea, vomiting, headache, indigestion, dysuria, urge frequency, and dizziness.
See harms of postcoital prophylaxis.
See harms of oestrogen (topical) in postmenopausal women.
In all RCTs identified by the review, prophylaxis was stopped in cases of recurrence. The review included one RCT of postcoital prophylaxis with TMP-SMX in its meta-analysis, which we also cover separately in the option on postcoital prophylaxis.
RCTs have found that, during treatment, continuous antibiotic prophylaxis reduced the number of clinical and microbiological recurrences compared with placebo. The results of the trials were consistent both in the direction and magnitude of the effect, and the quality of the trials was acceptable. However, once antibiotics were suspended, cystitis recurred and, in the two RCTs that assessed this outcome, rates equalled those of the placebo arm. These data confirm clinical suspicions that the effect on UTIs is short acting. The review did not assess whether the number of previous recurrences of cystitis correlated with the effectiveness of antibiotics. The reduction in rates of recurrent cystitis must be balanced against the increase in adverse effects in women taking antibiotics. Balancing benefits and harms is made more difficult because trials reported adverse effects in different ways, making it hard to collate data; and we found no RCTs assessing antibiotic resistance in women with recurrent cystitis. The other decision for clinicians is: for how long should continuous prophylaxis be given? Unfortunately, the RCTs identified by the review did not assess prophylaxis for longer than 12 months. The threshold for starting prophylactic treatment is also controversial; some experts advocate beginning prophylaxis after as few as two UTIs within a year, and others only after as many as six episodes within a year.
Continuous antibiotic prophylaxis (trimethoprim, trimethoprim–sulfamethoxazole, nitrofurantoin, cefaclor, or a quinolone or cephalexin) One already included systematic review updated. No new data or studies added to this review. One new comparison of continuous antibiotic prophylaxis versus topical oestrogen in postmenopausal women also added, which includes one new systematic review with two included RCTs.Categorisation unchanged (Beneficial).
RECURRENCE OF INFECTIONS Compared with placebo/no treatment: Postcoital antibiotic prophylaxis within 2 hours of sexual intercourse may be more effective at reducing the proportion of women with clinical recurrence ( low-quality evidence ). Compared with continuous antibiotic treatment: We don’t know whether postcoital antibiotic prophylaxis with ciprofloxacin is more effective than daily ciprofloxacin at reducing the number of microbiological recurrences (low-quality evidence).
We found one systematic review (search date 2004, 2 RCTs, 162 women with at least 2 episodes of UTI a year) comparing postcoital (within 2 hours of sexual intercourse) antibiotic regimens versus placebo or versus continuous antibiotic prophylaxis.
The first RCT (27 women) identified found that, compared with placebo, postcoital prophylaxis significantly reduced the proportion of women with clinical recurrence (2/16 [12%] with postcoital antibiotic prophylaxis v 9/11 [82%] with placebo; RR 0.15, 95% CI 0.04 to 0.58).
The second RCT (135 women) identified by the review compared postcoital oral ciprofloxacin 125 mg versus daily oral ciprofloxacin 125 mg. It found no significant difference in number of microbiological recurrences per person-year between postcoital and daily ciprofloxacin (0.46 recurrences per person-year with postcoital v 0.42 per person-year with daily; P = 0.80).
The first RCT identified by the review found that four women taking trimethoprim–sulfamethoxazole had nausea, vaginal candidiasis, and vaginal symptoms (unspecified); no-one taking placebo had adverse effects.
The second RCT identified by the review found no significant difference in adverse effects, including gastrointestinal distress, headaches, rash, and vaginal candidiasis, between continuous and postcoital ciprofloxacin (4/70 [6%] with postcoital v 9/65 [14%] with continuous; RR 0.40, 95% CI 0.13 to 1.24).
No new evidence
RECURRENCE OF INFECTIONS Compared with continuous co-trimoxazole: Single-dose self-administered trimethoprim–sulfamethoxazole taken at the onset of cystitis symptoms may be less effective at reducing the number of episodes of cystitis in women with 2 or more culture-documented UTIs in the previous 12 months ( very low-quality evidence ). NOTE The higher rate of cystitis in women using single-dose prophylaxis is to be expected, because treatment is only administered after the onset of symptoms. We found no direct information about other single-dose antibiotics in the treatment of recurrent cystitis in non-pregnant women.
We found no systematic review, but found one crossover RCT (38 non-pregnant women with 2 or more culture-documented UTIs in the previous 12 months). The RCT compared single-dose, self-administered trimethoprim–sulfamethoxazole (TMP-SMX, co-trimoxazole) (40 mg or 200 mg) to be taken at the onset of cystitis symptoms versus continuous oral TMP-SMX prophylaxis (40 mg or 200 mg). It found that single-dose TMP-SMX was significantly less effective in reducing the number of episodes of cystitis compared with continuous TMP-SMX (2.2 infections per person-year with treatment at onset of symptoms v 0.22 infections per person-year with continuous prophylaxis; P less than 0.001). We found no RCTs of other single-dose antibiotics.
The RCT reported that 10/38 (26%) women did not complete the full study protocol, and it is not clear whether analysis of results was by intention to treat. It found that women were almost always able to diagnose their own episodes of cystitis from symptoms (positive predictive value 92%). The higher rate of cystitis in women using single-dose prophylaxis is to be expected because treatment was only administered after the onset of symptoms.
No new evidence
RECURRENCE OF INFECTIONS Compared with placebo: Cranberry products (juice or capsules) may be more effective at reducing the recurrence of symptomatic cystitis at 12 months ( low-quality evidence ).
We found one systematic review (2 RCTs, 511 women) comparing cranberry juice or other cranberry products versus placebo. The review found that cranberry products (juice or capsules) significantly reduced the recurrence of symptomatic cystitis over 12 months compared with placebo (2 RCTs, 300 women: 31/146 [21%] with cranberry v 35/95 [37%] with placebo; RR 0.61, 95% CI 0.40 to 0.91). Neither of the RCTs identified by the review assessed microbiological recurrence.
In one RCT identified by the review, two women taking cranberry juice withdrew owing to symptoms of reflux. Women taking cranberry capsules had mild nausea and an increased frequency of bowel movements. However, women taking placebo also complained of headache and mild nausea.
The two RCTs identified by the review gave different doses of cranberry; the first RCT included in the meta-analysis gave 7.5 g cranberry concentrate daily (in 50 ml), and the second RCT gave 1:30 concentrate either in 250 ml juice or in tablet form. Neither of the trials explained the dose given to participants and there was no standardisation of the description of the dosage (i.e. concentration) given.
There is some evidence, from the results of two well-conducted RCTs, to recommend cranberry juice for the prevention of recurrent cystitis. There is no clear evidence about the amount and concentration of cranberry juice that needs to be consumed, or about the length of time needed for the treatment to be most effective.
No new evidence
RECURRENCE OF INFECTIONS Compared with placebo: Methenamine hippurate may be more effective at reducing recurrence of cystitis at 1 year ( very low-quality evidence ). Compared with continuous antibiotics: We don’t know whether continuous prophylaxis with methenamine hippurate is more effective at reducing the proportion of women with at least one clinical or microbiological recurrence at 1 year (very low-quality evidence).
We found one systematic review (search date 2003, 3 RCTs, 372 women) comparing methenamine hippurate versus placebo for 6–12 months. All RCTs had important problems with their methods — primarily that each participant could contribute more than once to assessment of recurrence rate. This may have overestimated the effects of treatment, and therefore the review reported few data. All RCTs found that methenamine hippurate reduced recurrence compared with placebo. The smaller RCTs (30 and 52 women with recurrent lower UTI) both found that methenamine hippurate reduced recurrence compared with placebo (monthly recurrence rate 0.03–0.08 episodes per month with methenamine hippurate v 0.25–0.34 episodes per month with placebo; CI not reported). The largest of the RCTs (290 people [92% women] with recurrent UTI) also included women with chronic pyelonephritis. It compared four interventions: methenamine hippurate 1 g, nitrofurantoin 75 mg, trimethoprim 100 mg, and placebo. It also found that methenamine hippurate reduced the recurrence of cystitis after 1 year compared with placebo (recurrence was observed on 34% of tests in women taking methenamine hippurate v 63% with placebo; CI not reported).
See benefits of continuous antibiotic prophylaxis.
In one of the smaller RCTs (52 women), one woman taking methenamine hippurate had severe bladder pain. In the other small RCT (30 women), two women in each group withdrew owing to adverse effects, including rash, nausea, and abdominal pain. The largest RCT found that one woman taking methenamine hippurate had nausea and rash, and one woman taking placebo had constipation.
See harms of continuous antibiotic prophylaxis.
No new evidence
We found no direct information about the effects of passing urine after intercourse in women with recurrent cystitis.
We found no systematic review or RCTs.
We found no RCTs.
The concept of passing urine after intercourse to prevent UTIs is based on anecdotal evidence and biological plausibility, but we found no RCTs of this intervention.
No new evidence
RECURRENCE OF INFECTIONS Topical (intravaginal) oestrogen compared with placebo or no treatment: Intravaginal oestrogen may be more effective at reducing the proportion of women with bacteriuria at 5–8 months ( very low-quality evidence ). Topical (intravaginal) oestrogen compared with antibiotics: We don’t know whether intravaginal oestrogen is more effective at reducing the proportion of women with bacteriuria (symptomatic or asymptomatic) at 3–9 months (very low-quality evidence).
We found one systematic review (search date not reported; 5 RCTs, 402 women; see comment below) assessing intravaginal oestrogen for the prevention of recurrent cystitis in postmenopausal women which pooled data. We found no subsequent RCTs.
The review included three RCTs comparing intravaginal oestrogen versus placebo or no treatment (see comment below). Of these three RCTs, one used oestriol cream, one used an oestradiol tablet, and the remaining RCT used an oestradiol-releasing vaginal ring. The follow-up in the RCTs was between 5–8 months. The review's primary outcome was the absence of symptomatic or asymptomatic episodes of bacteriuria. It did not report results for symptomatic episodes only. The review found that intravaginal oestrogen significantly reduced the proportion of women with bacteriuria compared with control at 5–8 months (3 RCTs: 51/99 [52%] with intravaginal oestrogen v 63/90 [70%] with control, RR 0.72, 95% CI 0.57 to 0.91). However, the largest RCT (108 women) included in the meta-analysis was an open study with a control group that received no treatment (compared with an oestradiol-releasing vaginal ring). Hence, the result should be interpreted with caution. The review did not combine data for the remaining two blinded placebo-controlled RCTs alone.
The reviewincluded 2 RCTs comparing intravaginal oestrogen versus antibiotics (see comment below). The first included RCT (171 women) compared an oestriol pessary plus oral placebo versus oral nitrofurantoin plus a placebo vaginal pessary for 9 months. The RCT found that oral nitrofurantoin significantly increased the proportion of women free of bacteriuria (symptomatic or asymptomatic) compared with oestriol pessary (no episodes of bacteriuria: 41/85 [48%] with oral nitrofurantoin v 28/86 [33%] with oestriol pessary; Kaplan–Meier analysis, P = 0.0004 by log-rank test).Unlike the other RCTs included in the review, this RCT used a pessary. The second included RCT (45 women) compared a conjugated oestrogen (premarin) cream versus oral antibiotics (not further defined in the review) for 3 months. The review found that intravaginal oestrogen significantly reduced the proportion of women with bacteriuria compared with antibiotics (2/27 [7%] with intravaginal oestrogen v 12/15 [80%] with oral antibiotics, RR 0.09, 95% CI 0.02 to 0.36). However, interpretation is difficult owing to the small size and duration of the study, and the non-classification of the antibiotic and regimen used. Hence, the results should be interpreted with caution.
The review reported that full text articles of two included RCTs were unavailable ("no available English translation" in one case, and "full text cannot be accessed" in the other), and data obtained from the abstracts were included in the review. The review also presented an analysis of intravaginal oestrogens versus all control groups combined (including placebo, no treatment, and antibiotics), which we have not presented here, as it is difficult to draw conclusions when comparing topical oestrogens with data combined from clinically heterogeneous arms (including placebo and antibiotics). Some included RCTs lacked data on time of onset of cystitis (pre- or post-menopausal) and sufficient information on co-morbidities or aetiology of the cystitis, or both. It may be possible that women who did not benefit from topical oestrogen had other anatomico-pathophysiological factors exacerbating recurrent cystitis. The review reported that participants had variable length of treatment, with different types of oestrogen and dose regimens, making them heterogenous as a group. We found one earlier systematic review (search date 1998), with a meta-analysis that included data from postmenopausal women treated with either oral or topical oestrogen, which found a significant benefit in reduction of UTIs with oestrogen compared with placebo (OR 2.51, CI 1.48 to 4.25).However, this analysis also included studies using oral oestrogens, which we do not report in this option. The earlier review did not include any additional studies of topical oestrogens which were not identified by the later systematic review reported in the benefits section.