Skip to main content
NCBI home page
The Cochrane Database of Systematic Reviews logoLink to The Cochrane Database of Systematic Reviews
. 2005 Apr 20;2005(2):CD004682. doi: 10.1002/14651858.CD004682.pub2

Duration of antibacterial treatment for uncomplicated urinary tract infection in women

Gai Milo 1,✉, Eugene Katchman 1, Mical Paul 2, Thierry Christiaens 3, Anders Baerheim 4, Leonard Leibovici 5
Editor: Cochrane Kidney and Transplant Group
PMCID: PMC12700497  PMID: 15846726

Abstract

Background

Uncomplicated urinary tract infection (UTI) is a common disease, occurring frequently in young sexually active women. In the past, seven day antibiotic therapy was recommended while the current practice is to treat uncomplicated UTI for three days.

Objectives

TO compare the efficacy and safety of three‐day antibiotic therapy to multi‐day therapy (five days or longer) on relief of symptoms and bacteriuria at short‐term and long‐term follow‐up.

Search methods

The Cochrane Library (Issue 1, 2004), the Cochrane Renal Group's Register of trials (July 2003), EMBASE (January 1980 to August 2003), and MEDLINE (January 1966 to August 2003) were searched. We scanned references of all included studies and contacted the first or corresponding author of included trials and the pharmaceutical companies.

Selection criteria

Randomised controlled trials comparing three‐days oral antibiotic therapy with multi‐day therapy (five days and longer) for uncomplicated cystitis in 18 to 65 years old non‐pregnant women without signs of upper UTI.

Data collection and analysis

Data concerning bacteriological and symptomatic failure rates, occurrence of pyelonephritis and adverse effects were extracted independently by two reviewers. Risk ratio (RR) and their 95% confidence intervals (CI) were estimated. Outcomes were also extracted by intention‐to‐treat analysis whenever possible.

Main results

Thirty‐two trials (9605 patients) were included. For symptomatic failure rates, no difference between three‐day and 5‐10 day antibiotic regimen was seen short‐term (RR 1.06, 95% CI 0.88 to 1.28) and long‐term follow‐up (RR 1.09, 95% CI 0.94 to 1.27). Comparison of the bacteriological failure rates showed that three‐day therapy was less effective than 5‐10 day therapy for the short‐term follow‐up, however this difference was observed 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). This difference was more significant at long‐term follow‐up (RR 1.43, 95% CI 1.19 to 1.73, P = 0.0002). Adverse effects were significantly more common in the 5‐10 day treatment group (RR 0.83, 95% CI 0.74 to 0.93, P = 0.0010). Results were consistent for subgroup and sensitivity analyses.

Authors' conclusions

Three days of antibiotic therapy is similar to 5‐10 days in achieving symptomatic cure during uncomplicated UTI treatment, while the longer treatment is more effective in obtaining bacteriological cure. In spite of the higher rate of adverse effects, treatment for 5‐10 days could be considered for treatment of women in whom eradication of bacteriuria is important.

Keywords: Female; Humans; Anti-Infective Agents, Urinary; Anti-Infective Agents, Urinary/therapeutic use; Randomized Controlled Trials as Topic; Urinary Tract Infections; Urinary Tract Infections/drug therapy

Plain language summary

Duration of antibacterial treatment for uncomplicated urinary tract infection in women

Uncomplicated urinary tract infection (UTI) is a common disease occurring frequently in young women. It is caused by bacteria multiplying in urine, and the patient usually complains of urgency and burning pain while urinating. The present practice is to treat the patient with antibiotics for three days. In this review we included all studies that compared three‐day therapy with longer treatment (five days or more). Three days of treatment were adequate to achieve symptomatic relief for most patients, but it appears that longer therapy is better in terms of bacteria elimination from the urine, no matter what antibiotic is used. Longer therapy for UTI is related to higher rate of adverse reactions to the antibiotics used. Pending further research, it could be considered for women in whom eradication of bacteria in the urine is important.

Background

Uncomplicated urinary tract infection (UTI) is a common disease, occurring frequently in young sexually active women. In one cohort study the incidence of the disease was estimated to be 0.5‐0.7/person‐year (Hooton 1996). All over the world the most common pathogens of uncomplicated UTI are similar: 80‐90% Escherichia coli, 5‐10% Staphylococcus saprophyticus, the remaining infections being caused by Proteus spp., and other Gram‐negative rods. Most are bacteria from the gut that colonize the perineum and then ascend through the urethra to infect the bladder mucosa. The infection causes specific symptoms, mainly the triad of dysuria (painful urination), urgency (the urgent need to void) and frequency (very frequent urination). In randomised controlled trials (RCTs) the diagnosis is based on positive urine cultures in symptomatic subjects. In the past, the threshold for diagnosis of UTI was >105 colony forming units (CFU)/mL of voided midstream urine (Stamm 1982). However two decades ago studies have shown that in young symptomatic women with leucocyturia even 100 CFU/mL voided midstream urine can establish the diagnosis (Stamm 1980; Stamm 1982; Kunin 1993).

A large range of antimicrobials with different rates of cure and side effects are used in the treatment of UTI. It is thought that a short‐course therapy consisting of a three‐day antibacterial regimen is sufficient for uncomplicated urinary tract infection, as it is probably as effective as 7‐10 days therapy, and may be associated with less side effects and lower costs (Hooton 1997). Single dose therapy has been advocated for years but about a decade ago reviews have raised doubts as to its use because of a higher frequency of bacteriological recurrence (Leibovici 1991; Norrby 1990), and it is no longer common clinical practice. On the other hand, single‐dose treatment probably achieves symptomatic relief more rapidly than seven days of treatment (Arav‐Boger 1994).

In most clinical trials assessing effectiveness of therapy, cure was defined as bacteriological cure, rather than symptomatic relief. Uncomplicated UTI is not considered a serious disease. It is not clear whether untreated UTI can progress to pyelonephritis, and if so how often. Progression to pyelonephritis probably occurs at a very low rate, while asymptomatic bacteriuria in young, healthy and non‐pregnant women is not associated with renal damage (Stamm 1991).

Thus since our last systematic review on the length of treatment of uncomplicated UTI in young women (Leibovici 1991), the following questions arose:

  1. What is the relative effectiveness of three days treatment compared with multi‐day treatment?

  2. Is any difference modified by the antibiotic used (old versus new) or CFU/mL count?

  3. Do persistent positive cultures lead to persistent symptoms?

  4. What is the relative effectiveness of single dose and three‐day treatment, compared with seven days treatment, when the outcome of interest is symptomatic cure rather than bacteriological one?

  5. Does the duration of treatment influence the development of resistant strains during treatment?

Objectives

The main objective of this review was to assess the evidence, as found in RCTs for the relative effectiveness of different regimens of antibacterial treatments in acute, uncomplicated lower urinary tract infection in otherwise healthy 16 to 65 years old females.

Specific objectives were:

  1. To assess the evidence for the relative effectiveness as assessed in RCT's comparing three‐day versus multi‐day therapy on:

    1. Relief of symptoms within two weeks after start of treatment ( mostly within seven days)

    2. Resolution of bacteriuria within two weeks after start of treatment (bacteriological cure)

    3. Recurrence of symptoms or bacteriuria between cure and up‐to eight weeks after start of treatment

    4. To assess the frequency of adverse events in the different regimens

  2. To assess the evidence for the relative effectiveness of the different antibacterial drugs used in the RCTs.

  3. To assess the evidence for development of resistance for different durations of treatment with different drugs (comparing resistance of grown bacteria before and after therapy).

Methods

Criteria for considering studies for this review

Types of studies

We attempted to identify all RCTs comparing the relative effectiveness of three day versus five days or longer oral antibacterial therapy for uncomplicated UTI in women.

Types of participants

We included studies on ambulatory, otherwise healthy women, aged 16‐65 years, with uncomplicated UTI defined by the presence of urinary complaints (and by the absence of upper UTI signs); whenever possible, analysis for the review was limited to women with positive urine cultures of more than 100 CFU/mL of voided midstream urine or obtained via urinary catheter.

Uncomplicated UTI was defined as the absence of all the following:

  1. Costovertebral pain or tenderness

  2. Fever (more than 37.8 C)

  3. Positive blood cultures.

In addition, trials of the following groups of people were excluded from the review:

  1. Multiple vomiting

  2. Sepsis

  3. Children up to the age of 16 years

  4. Hospital acquired infection

  5. Pregnancy

  6. Indwelling urinary catheter

  7. Recent urinary tract instrumentation

  8. Known pathological, functional or anatomic abnormality of the urinary tract

  9. Diabetes mellitus

  10. Immunocompromised patients

Types of interventions

Three days oral antibacterial treatment versus antibacterial treatment for five days or more (antibacterial therapy given in both arms did not have to be identical).

Types of outcome measures

  1. Short‐term symptomatic failure, defined as persistence or recurrence of symptoms up to two weeks after starting treatment.

  2. Long‐term symptomatic failure, defined as persistence or recurrence of urinary symptoms up to eight weeks after start of treatment.

  3. Short‐term bacteriological failure, defined as a positive urine culture at the first follow‐up within two weeks after start of treatment.

  4. Long‐term bacteriological failure, defined as a positive urine culture up to eight weeks after start of treatment.

  5. Occurrence of pyelonephritis during follow‐up.

  6. Adverse events:

    1. Any serious adverse events that are fatal, life‐threatening, or requiring hospitalisation;

    2. Any adverse events that result in significant disability or incapacity;

    3. Any important medical events that may not be immediately life‐threatening, or result in death or hospitalisation, but may jeopardize the patient or may require intervention to prevent one of the above outcomes;

    4. Any adverse events that require discontinuation of medication.

    5. Adverse events by the involved organs: skin, gastro‐intestinal, vaginal discharge, central nervous system, others.

  7. The percentage of pathogens resistant to the study drug two to eight weeks after start of treatment.

Search methods for identification of studies

A). The Cochrane Library (Issue 3, 2003), the Cochrane Renal Group's Register of trials (July 2003), EMBASE (January 1980 to August 2003 ), and MEDLINE (January 1966 to August 2003) were searched with the phrase: 
 [(urinary near infection*) or cystitis or uti] and [(treatment near duration) or (single near dos*) or (3 near day*) or (three near day*)]

We included all languages. By leaving single dose in the search strategy we found articles that include single and three‐day doses versus multi‐day.

B). An additional search was performed in January 2004 with the assistance of the Trials Search Coordinator (see additional Table 1 ‐ Electronic databases searched)

1. Electronic search strategies.

Database searched Terms used
CENTRAL #1) URINARY TRACT INFECTIONS 
 #2) (urinary next tract next infection*) 
 #3) uti and utis 
 #4) bacteriuria* 
 #5) pyuria* 
 #6) (#1 or #2 or #3 or #4 or #5) 
 #7) ANTI‐INFECTIVE AGENTS 
 #8) anti‐infective* 
 #9) antiinfective* 
 #10) antibiotic* 
 #11) quinoline* 
 #12) cinoxacin 
 #13) (nalidixic next acid) 
 #14) (oxolinic next acid) 
 #15) fluoroquinolone* 
 #16) ciprofloxacin 
 #17) enoxacin 
 #18) fleroxacin 
 #19) norfloxacin 
 #20) ofloxacin 
 #21) perfloxacin 
 #22) (#8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21) 
 #23) (#6 and #22)
MEDLINE 1. exp Urinary Tract Infections/ 
 2. urinary tract infection$.tw. 
 3. uti.tw. 
 4. utis.tw. 
 5. pyuria.tw. 
 6. bacteriuria.tw. 
 7. or/1‐6 
 8. exp Anti‐Infective Agents/ 
 9. anti‐infective$.tw. 
 10. antiinfective$.tw. 
 11. antibiotic$.tw. 
 12. antibacterial$.tw. 
 13. quinolone$.tw. 
 14. cinoxacin.tw. 
 15. nalidixic acid.tw. 
 16. oxolinic acid.tw. 
 17. fluoroquinolone.tw. 
 18. ciprofloxacin.tw. 
 19. enoxacin.tw. 
 20. fleroxacin.tw. 
 21. norfloxacin.tw. 
 22. ofloxacin.tw. 
 23. pefloxacin.tw. 
 24. or/8‐23 
 25. 7 and 24 
 26. randomized controlled trial.pt. 
 27. controlled clinical trial.pt. 
 28. randomized controlled trials/ 
 29. random allocation/ 
 30. double blind method/ 
 31. single blind method/ 
 32. or/26‐31 
 33. animal/ not (animal/ and human/) 
 34. 32 not 33 
 35. clinical trial.pt. 
 36. exp clinical trials/ 
 37. (clinic$ adj25 trial$).ti,ab. 
 38. cross‐over studies/ 
 39. (crossover or cross‐over or cross over).tw. 
 40. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. 
 41. placebos/ 
 42. placebo$.ti,ab. 
 43. random$.ti,ab. 
 44. research design/ 
 45. or/35‐44 
 46. 45 not 33 
 47. 34 or 46 
 48. 25 and 47
Open in a new tab

C). Reference searching and personal contact: The references of all identified studies were inspected for more studies. Additionally, the first or corresponding author of each included study was contacted for complementary information on his own trial as needed.

Data collection and analysis

Two reviewers independently inspected each reference identified by the search and applied the inclusion criteria. For possible relevant articles, or in cases of disagreement between the two reviewers, the full article was obtained and inspected independently by a third reviewer.

Quality assessment

Trials fulfilling the review inclusion criteria were assessed for methodological quality by two reviewers. This was done using the criteria described in the Cochrane Handbook (Clarke 1999), based on the evidence of a strong association between poor allocation concealment and overestimation of effect (Schulz 1995) and defined as below:

Allocation concealment

A. Low risk of bias (adequate allocation concealment) 
 B. Moderate risk of bias (some doubt about the allocation concealment) 
 C. High risk of bias (inadequate allocation concealment)

For the purpose of the analyses in this review, trials were included if they meet the criteria A or B in the Handbook (Clarke 1999; Kunz 1998).

Intention‐to‐treat (ITT) analysis

ITT analysis was performed regarding all dropouts in study as failures to achieve symptomatic or bacteriological cure. Whenever possible, we regarded only the patients with positive urine cultures (significant bacteriuria) as the reference total patient number in the two study arms. When the numbers of randomised women with positive cultures in the study groups was unavailable, the total number of randomised patients was taken for performing the ITT analysis for symptomatic short‐term and long‐term failures, but not for the bacteriologic outcomes.

Data collection

Two reviewers independently extracted the data of included trials. Trials were identified by the name of the first author and year in which the trial was first published and ordered chronologically. The following data will be extracted, checked and recorded:

(Characteristics of trials

  • Date, location, period of data collection, year of publication;

  • Publication status;

  • Case definitions (symptomatic, bacteriological, both)

  • Bacteriologic definition (105 or 10² CFU/mL)

  • Sponsor of trial (commercial, academic, pharmaceutical, or unknown)

  • Blinding

  • Allocation concealment (yes, no and method)

  • Definitions of cure (symptomatic, bacteriological or both)

Characteristics of participants

  • Number of participants in each group;

  • Age (as described in the article: mean, median or range);

Characteristics of interventions

  • Type, dose and duration of antibacterial therapy;

Characteristics of outcome measures

  • No. of patients with bacteriological cure (as defined above) in each group;

  • No. of patients with symptomatic recurrence (as defined above) in each group, divided into local and systemic recurrences;

  • No. of patients with bacteriological recurrence (as defined above) in each group;

  • No. of patients with adverse reactions, per type and total;

  • No. of patients with resistant microorganisms, as defined above;

  • Lost to each follow‐up (dropouts) before end of study.

Data synthesis

Dichotomous data was analysed by calculating the risk ratio (RR) for each trial with the uncertainty in each result being expressed using 95% confidence intervals (CI). Whenever comparisons made between the mean duration of symptoms in the two groups were normally distributed, these continuous data were analysed by using the mean and standard deviation of each trial and calculating the effect size (average mean difference) and the 95% CI.

Heterogeneity and publication bias

Heterogeneity in the results of the trials was initially assessed by inspection of graphical presentations and by calculating a test of heterogeneity (Chi² and I² ‐ Higgins 2003). We anticipated between‐trial variation in estimation of morbidity for those patients who were treated with different antibiotics. Subgroup analyses were performed in order to assess the impact of this possible source of heterogeneity in the main results. The following factors were checked: allocation generation and concealment methods, different antibiotics groups (quinolones, beta‐lactams etc), per cent of dropouts in the studies.

A funnel plot estimating the precision of trials (plots of RR for efficacy against the sample size) was examined in order to estimate potential asymmetry. A fixed effect model was used throughout the review, except in the event of significant heterogeneity between the trials (P < 0.10), when the random effect model was chosen.

Results

Description of studies

The computerised search strategy identified a large number of publications comparing different regimens of antibiotic therapy for UTI, not all relevant for the present review. These were screened for RCTs, uncomplicated UTIs, antibiotics treatment duration and presence of exclusion criteria. Of 56 trials obtained this way 24 were excluded for different reasons (see Characteristics of excluded studies) while 32 RCTs were considered eligible for this review. 
 
 Two reports were identified as duplicate publications and are considered under their primary reference (Sandberg 1985). Five publications were found to be case‐control or non randomised studies (Bargelloni 1972; Furusawa 1994; Hoigne 1977; Liudvig 1996; Loran 1997), while five others were reviews of different trials of which several were included in the analysis (Blomer 1986; Hooton 1989; Iravani 1991; Iravani 1995; Vogel 1984). Eight RCTs compared two different antibiotic regimens of at least five days duration (Bailey 1983; Fancourt 1984; Hill 1985; Little 1979; Martin 1983; McCarthy 1972; Pelta 1985; Zorbas 1995), two additional trials were excluded for they compared a single‐dose antibiotic to ten‐day (Schultz 1984) or three‐day (Gellerman 1988) regimen. Another trial reported only clinical improvement but not cure (Ishihara 1998). One further study was excluded because it included only elderly postmenopausal women (mean age 66 ± 20) (Raz 1996). Two trials were excluded as they appeared to be quasi‐randomised or criterion C in the Handbook (Charlton 1976; Fair 1980).

Thirty‐two trials were included in the review (see Characteristics of included studies). One trial compared two different antibiotics with subgroups of three‐day and ten‐day treatment regimens in each, and the results for these two drugs were regarded separately as two different trials (Gordin 1987a; Gordin 1987b).

The contact authors of these 32 included and two excluded as quasi‐RCTs (Charlton 1976; Fair 1980) were contacted (by mail and if possible by e‐mail) of whom 10 replied. Unpublished data were obtained for seven studies.

Patient characteristics

The included studies were performed between the years 1980‐2002 and included 9605 randomised patients. The median number of patients/trial was 300.

In six trials (1356 patients) men were included (Basista 1991; Cox 1992; Hansen 1981; Menday 2000; Rapoport 1981; Stein 1987). Their number was less than 10% in each study group and it was impossible to separate the results for men and women for any of these trials. One additional trial (Bitsch 1985) included men, but analysis of men and women was separated and only data regarding women was used for this review.

Fourteen studies included women above 65 years of age (Basista 1991; Bitsch 1985; Cox 1992; Guibert 1997; Hansen 1981; Internordic 1988; Iravani 1999; Menday 2000; Piipo 1990; Rapoport 1981; Sandberg 1985; Stein 1987; Stein 1992; Tsugawa 1999). In all these 14 trials patients above 65 years made up the minority of the study groups and in 7 of these trials the mean age reported (33 to 45 years) was well below the upper limit we defined for this review (Basista 1991; Bitsch 1985; Guibert 1997; Hansen 1981; Rapoport 1981; Sandberg 1985; Stein 1992). Unfortunately, it was impossible to analyse data for patients below and above the age of 65 separately.

Nearly all trials defined bacteriuria as more than 105 CFU/mL for any bacteria or the same concentration for Gram‐negative bacteria and 104 CFU/mL for Staphylococcus. Several studies included patients with lower urine bacteria concentration of 104 CFU/mL (Hovelius 1985; Neringer 1992; Stein 1992; Tsugawa 1999), 10³ CFU/mL (Iravani 1999) and 10² CFU/mL (Hooton 1991) for any bacteria. In one trial, positive urine culture was not necessary for patient inclusion and the case definition was based on the clinical signs and pathologic urinalysis (Guibert 1997).

In two trials several women with asymptomatic bacteriuria were treated and taken into account for the bacterial cure results (Gordin 1987a; Gordin 1987b; Hooton 1991).

Antibiotic regimens

The same antibiotics in the three‐day and 5‐10 day groups were used in 19 trials, of these quinolones were used in six trials (Garcia 2002; Internordic 1988; Neringer 1992; Piipo 1990; Trienekens 1993; Tsugawa 1999), beta‐lactams in eight (Gordin 1987b; Greenberg 1986; Hansen 1981; Hovelius 1985; Marsh 1980; Pitkajarvi 1990; Richards 1984; Sandberg 1985) and different combinations of sulfonamides and trimethoprim in five trials (Gordin 1987a; Gossius 1984; Gossius 1985; Iravani 1983; Trienekens 1989). In one of these studies different doses of the same antibiotic drug (pivmecillinam) were used in the two study groups (Hansen 1981).

Fourteen trials compared different antibiotics given in the three‐day and in the 5‐10 day groups. The drug in the three‐day group was a quinolone in nearly all of these studies, and was compared to 5‐10 day regimens of beta‐lactam (Winwick 1981), different combinations of sulfonamides and trimethoprim (Basista 1991; Bitsch 1985; Butler 1983; Cox 1992; Hooton 1991; Stein 1987), another quinolone (Henry 1999; Guibert 1997; Stein 1992) or a combination of nitrofurantoin with trimethoprim‐sulfamethoxazole (Iravani 1999). One additional trial compared three‐day treatment with trimethoprim‐sulfamethoxazole to seven‐day treatment with any of a long list of antibiotics (Rapoport 1981). In two trials three‐day therapy with beta‐lactam was compared to seven‐day treatment with another drug of the beta‐lactam group (Menday 2000) or trimethoprim‐sulfamethoxazole (Figueroa 1999).

Risk of bias in included studies

Randomisation and allocation concealment

Adequate allocation concealment, using sealed envelopes or central randomisation, was described in 12 trials (Basista 1991; Bitsch 1985; Gordin 1987a; Gordin 1987b; Henry 1999; Hooton 1991; Hovelius 1985; Iravani 1999; Neringer 1992; Piipo 1990; Richards 1984; Sandberg 1985; Trienekens 1993). Allocation generation was adequate in all 12 and in additional six (Butler 1983; Gossius 1985; Marsh 1980; Pitkajarvi 1990; Stein 1987; Stein 1992). These studies used computer‐generated lists or predetermined randomised codes. Randomisation methods were not described in all other trials.

Blinding

Ten trials were double‐blinded (Henry 1999; Internordic 1988; Iravani 1999; Menday 2000; Neringer 1992; Piipo 1990; Stein 1992; Trienekens 1989; Trienekens 1993; Tsugawa 1999), one single‐blinded (Richards 1984) and the remaining open RCTs.

ITT analysis

ITT analysis was presented in only two of the 32 trials included for treatment failure (Henry 1999; Iravani 1999). Dropouts and numbers of patients with positive urine cultures were reported by their allocation group in 21 of 32 trials presenting per protocol analysis for treatment failure, permitting a second ITT analysis assuming dropouts as failures. The number of patients excluded from the analysis at the first follow‐up ranged between 0% to 20% for bacteriological cure outcome and 0% to 26% for clinical (symptomatic) cure; at the second follow‐up these numbers were 0% to 29% and 6% to 45%, respectively.

The first follow‐up was performed between two to 15 days from the end of the treatment (short‐term), and the second follow‐up was performed four to 10 weeks from the treatment (long‐term).

Effects of interventions

Trials were divided into two major subgroups: those with the same antibiotics in the two allocation groups and those with different drugs.

Effectiveness

Symptomatic failure
Short‐term

Assessment of short‐term symptomatic failure rate was possible in 24 trials (8752 patients). Data for efficacy analysis was available in 5165 patients. No significant difference between three‐day and 5‐10 day antibiotic treatment was observed (Analysis 1.1: RR 1.06, 95% CI 0.0.88 to 1.28, P = 0.52), with no significant heterogeneity observed for this comparison (Chi² = 27.14, df = 23, P = 0.25, I² = 15.3%)

Separate analysis of trials with same or different antibiotic in the two treatment arms showed no significant difference. In 14 trials comparing the same antibiotic the RR was 1.15 (95% CI 0.95 to 1.39, Analysis 1.1.1) in 10 trials with different antibiotics the RR was 0.90 (95% CI 0.62 to 1.29, Analysis 1.1.2). No differences were shown after performing subgroup analyses for the factors: antibiotic classes (quinolones, beta‐lactams, sulfonylamides with or without trimethoprim); allocation generation and concealment; or per cent of dropouts.

1.1. Analysis.

1.1

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 1 Short‐term symptomatic failure (2‐15 days from end of treatment).

Long‐term

Assessment of long‐term symptomatic failure rate was available from eight trials (3141 patients). No difference was found between the two arms (Analysis 1.3: RR 1.09, 95% CI 0.94 to 1.27). After performing subgroup analysis as for the first follow‐up results no differences were shown.

1.3. Analysis.

1.3

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 3 Long‐term symptomatic failure (4‐10 weeks from end of treatment).

A secondary ITT analysis counting dropouts as failures of treatment showed similar results (Analysis1.4.1; Analysis1.4.2).

Bacteriologic failure
Short‐term

Assessment of short‐term bacteriological failure rate was possible in 31 trials (8874 patients). For efficacy analysis 5368 patients were included, the majority of the excluded persons having negative urine cultures after being allocated to one of the study regimens. Five to 10‐day antibiotic regimen appeared to be superior to the three‐day regimen although the result was not significant using the random effects model (Analysis 1.5: RR 1.19, 95% CI 0.98 to 1.44, P = 0.08), but just significant with the fixed effect model (RR 1.20, 95% CI 1.00 to 1.44, P = 0.05). No significant heterogeneity was observed for this comparison (Chi² = 24.54, df = 29, P=0.70, I² = 0%). This advantage was observed in trials comparing the same antibiotic (Analysis 1.5.1: RR 1.37, 95% CI 1.07 to 1.74; P = 0.01), and absent in the subgroup analysis of trials comparing different drugs (RR 0.96, 95% CI 00.68 to 1.35, P = 0.80). The trials using same antibiotic drug in the two treatment arms was further divided for subgroup analysis based on the different antibiotic classes (Analysis 1.6) and showed that the results were not significantly influenced by the drug choice. The results remain unchanged after performing the other subgroup analyses (for allocation generation and concealment class, trial size or per cent of dropouts).

1.5. Analysis.

1.5

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 5 Short‐term bacteriologic failure (2‐15 days from end of treatment).

1.6. Analysis.

1.6

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 6 Short‐term bacteriological failure by antibiotic class (same drug) (2‐15 days from end of treatment).

A secondary ITT analysis for the short‐term results was only possible in 21/31 trials. Its results showed actually no difference between the two treatment regimens, (Analysis 1.7: RR 0.92, 95% CI 0.80 to 1.06). No difference was observed in any of the subgroups analyses.

1.7. Analysis.

1.7

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 7 Short‐term bacteriological failure ‐ ITT (2‐15 days from end of treatment).

Long‐term

Assessment of the long‐term bacteriological failure rate was possible in 18 trials (3715 patients) (13 trials and 2502 patients in the same antibiotic subgroup; five trials and 1213 patients in the different regimens subgroup). The 5‐10‐day antibiotic regimen was superior to the three‐day regimen (Analysis 1.8: RR 1.31, 95% CI 1.08 to 1.60, P = 0.006) and no significant heterogeneity was observed (Chi² = 24.40, df = 17, P = 0.11, I² = 30.3%). A significant difference was shown in the subgroup of trials with the same drug in both allocation arms (Analysis 1.8.1: RR 1.43, 95% CI 1.19 to 1.73, P = 0.0002), while no difference was observed between 5‐10 day and three day regimens when different drugs were used. These results also remain unchanged after performing the additional subgroup analyses for antibiotic class (Analysis 1.9), allocation generation, trial size and concealment class or per cent of dropouts.

1.8. Analysis.

1.8

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 8 Long‐term bacteriological failure (4‐10 weeks from end of treatment).

1.9. Analysis.

1.9

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 9 Long‐term bacteriological failure by antibiotic class (same drug) (4‐10 weeks from end of treatment).

A secondary ITT analysis for the second follow‐up results showed the same results as the efficacy analysis, confirming the observed significant advantage of 5‐10 day antibiotic regimen over the three‐day regimen for all trials (Analysis 1.10: RR 1.19, 95% CI 1.06 to 1.35, P = 0.004), and for the subgroup of the same drug regimen (Analysis 1.10.1: RR 1.26, 95% CI 1.08 to 1.47, P = 0.003). The results of the subgroup analysis for the class of antibiotic drug are shown in Analysis 1.11.

1.10. Analysis.

1.10

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 10 Long‐term bacteriological failure ‐ ITT (4‐10 weeks from end of treatment).

1.11. Analysis.

1.11

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 11 Long‐term bacteriological failure ‐ ITT by antibiotic class (same drug) (4‐10 weeks from end of treatment).

Pyelonephritis

Only five of the included trials reported the incidence of pyelonephritis (Cox 1992; Gossius 1984; Gossius 1985; Hovelius 1985; Winwick 1981). Only two cases of pyelonephritis were reported, both in the three‐days therapy groups (Gossius 1984; Gossius 1985). As this outcome was extremely uncommon in the population of young women with uncomplicated lower UTI, no difference could be observed between the two treatment regimens (Analysis 1.13).

1.13. Analysis.

1.13

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 13 Patients developed pyelonephritis.

Adverse effects

All side effects were observed more frequently in the 5‐10 day regimen than in the three‐day group. The risk for the development of any side effect during therapy was 17% lower in the three‐day group (Analysis 1.12: RR 0.83, 95% CI 0.74 to 0.93, P = 0.0010). This difference was more prominent in trials comparing the same antibiotic (Analysis 1.12.1: RR 0.76, 95% CI 0.63 to 0.92) and especially when the drug was sulfonylamide/trimethoprim (Analysis 1.20: RR 0.40, 95% CI 0.19 to 0.88).

1.12. Analysis.

1.12

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 12 Patients with any adverse effects during treatment.

1.20. Analysis.

1.20

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 20 Patients with any adverse effects during treatment by antibiotic class (same drug).

A substantially lower percentage of patients had to discontinue therapy in the three‐day group, (Analysis 1.14: RR 0.51, 95% CI 0.328 to 0.91, P = 0.02), particularly when the same drug was given in the two groups (Analysis 1.14.1: RR 0.35, 95% CI 0.12 to 0.98, P = 0.04).

1.14. Analysis.

1.14

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 14 Adverse effects requiring therapy discontinuation.

Gastrointestinal side effects appeared less frequently during three‐day treatment (Analysis 1.15: RR 0.81, 95% CI 0.67 to 0.94, P = 0.02). The difference in the frequency of development a skin rash was significant in the trials comparing the same antibiotic (Analysis 1.16.1: RR 0.51,95% CI 0.33 to 0.77, P = 0.002), while no such difference was observed in the trials with different drugs (Analysis 1.16.2: RR 0.69, 95% CI 0.21 to 2.28). The rate of side effects related to central nervous system was also slightly more frequent in the 5‐10 day group, but this difference was not significant overall (Analysis 1.17: RR 0.83, 95% CI 0.65 to 1.06, P = 0.13).

1.15. Analysis.

1.15

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 15 Gastrointestinal adverse effects.

1.16. Analysis.

1.16

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 16 Skin adverse effects.

1.17. Analysis.

1.17

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 17 CNS adverse effects.

As for anaphylactic reactions, only two trials described one case, both in the 5‐10 day group (Butler 1983; Gossius 1984), and no difference could be observed between the two treatment regimens.

Resistant organisms

Only a minority of the included trials described the antibiotic resistance profile of the bacteria cultured from patients urine before and after treatment. In two studies using quinolones in both treatment arms, no persistent or recurrent pathogen developed resistance to the study drugs during treatment or during the follow‐up period (Internordic 1988; Neringer 1992). In one trial studying thee‐day versus seven‐day pivmecillinam regimens (Richards 1984) the number of resistant bacteria isolates after therapy did not change, and an additional trial using the same drug (Hansen 1981) showed only total rate of resistance development after therapy without specification to different study groups. Two studies using sulfonamide (Iravani 1983) and co‐trimoxazole (Trienekens 1989) mentioned the prevalence of the drug‐resistant E. coli in the failure cases, but it was unclear whether these were primary resistant strains or the resistance developed during the treatment. One study (Basista 1991) showed significant difference in the development of urine bacteria resistance between the three‐day (no cases) and the seven‐day (three cases) protocols but the drugs used in the two treatment arms were different (quinolone versus trimethoprim/sulfamethoxazole), so this data is of only limited value.

Dropouts and selection bias

Funnel plots for symptomatic (Figure 1 ‐ Funnel plot symptomatic failure) and bacteriological failure (Figure 2 ‐ Funnel plot bacteriologic failure) showed that several smaller studies favouring the three‐day regimen may be missing from this review. It is important to mention that all the studies included in this meta‐analysis were planned to check the hypothesis that the three‐day antibiotic therapy is as effective as a longer one.

1.

1

Open in a new tab

Funnel plot ‐ symptomatic failure

2.

2

Open in a new tab

Funnel plot ‐ bacteriologic failure

The number of patients excluded from each study arm was nearly equal, both for symptomatic and bacteriological outcomes assessment.

Discussion

Thirty‐two RCTs, including 9605 patients, comparing three‐day antibiotic treatment to 5‐10 day treatment for the empirical therapy of uncomplicated UTIs in the young and middle‐aged women were analysed. Two outcomes were chosen for comparison: symptomatic failure and bacteriological failure as defined by positive post‐treatment urine cultures. Primary treatment failures and recurrences or re‐infections were considered together as therapy failures, for in the majority of the studies no distinction could be made between them.

Symptomatic failure rates did not differ significantly both in the short‐term (RR 1.06, 95% CI 0.88 to 1.28) or long‐term (RR 1.07, 95% CI 0.99 to 1.16) after treatment with three‐day or 5‐10 day regimens. No information about the timing of the symptomatic cure could be found in the included studies.

Five to 10 day antibiotic regimen was more effective than three day therapy, keeping the patients' urine sterile two to 15 days after the end of treatment (same drug therapy RR 1.37, 95% CI 1.07 to 1.74, P = 0.01). This means that 41 women would have to be treated for seven days to prevent one case of recurrence or persistence of bacteriuria for a short period. The ITT analysis showed no difference between short and long treatment regimens. Data considering the numbers of randomised patients with positive urine cultures were unattainable from the published articles or any additional source in six major studies in this subgroup (Garcia 2002; Gossius 1984; Gossius 1985; Marsh 1980; Richards 1984; Trienekens 1993), so it was impossible to include these trials into the ITT analysis. This fact, together with the high rate of dropouts, could explain why we failed to show a significant effect of therapy duration on the short‐term bacteriologic failure rates in the ITT analysis.

A larger advantage of 5‐10 day over three‐day antibiotic therapy in preventing bacteriological failure was observed after 4 to 10 weeks (RR 1.43, 95% CI 1.19 to 1.73, P = 0.0002) when treatment with the same drug was compared (number needed to treat (NNT) = 4). This difference remained significant also with an ITT analysis (RR 1.26, 95% CI 1.08 to 1.47, P = 0.003). It is important to mention that the advantage of the longer therapy in terms of bacteriological success appeared to be independent of the antibiotic class chosen for UTI treatment including quinolones.

One reason for the advantage of longer therapy might be the survival of bacteria in subepithelial loci of the lower urinary tract after a shorter course of antibiotic treatment. Recently the ability of E. coli to invade epithelial cells and create biofilms with pod‐like bulges on the bladder surface was discovered (Anderson 2003). These pods contain bacteria encased in a polysaccharide‐rich matrix surrounded by a protective shell of uroplakin, and allow bladder infections to persist in the face of robust host defences and short‐term antibiotic treatment. Another recently published study showed that asymptomatic bacteriuria is associated with an increased risk of symptomatic UTI in young women (Hooton 2000). Thus, bacteriological failure might also carry a clinical significance for the patients.

The probable cause for the absence of such difference in the trials comparing different drugs in the two study groups was the fact that all but three of these trials compared three‐day quinolone therapy with 5 to 10 day regimen of beta‐lactams or sulfonylamides/trimethoprim. Both the higher urine concentration and the lower rate of bacteria drug resistance favoured the newer quinolones. When trying to answer the question concerning the optimal treatment duration for UTI one should probably consider trials comparing the same drug in the two therapy groups.

We found a discrepancy between symptomatic cure, which was not influenced by treatment duration, and bacteriological cure. Fewer included trials showed results of symptomatic cure as compared to bacteriologic results (21 versus 31 studies at the first follow‐up and 10 versus 18 at the second follow‐up, respectively). This could be one of the reasons explaining the discrepancy between the efficacy results for these two outcomes.

Not surprisingly, the cost of the higher bacteriological cure rates after longer antibiotic therapy is a significantly higher rate of adverse events, including those leading to treatment discontinuation. Again, the difference was observed in the trials dealing with two regimens of the same drug. The per cent of patients who stopped the treatment because of adverse effects in the three‐day group was 1.5% compared to 3.2% in the 5‐10 day group (RR 0.35, 95% CI 0.12 to 0.98, P = 0.04), number needed to harm = 79. However all adverse effects were minor.

We performed sensitivity analyses that did not detect sources of bias originating in studies`design, methodology or class of antibiotic drug used. However, allocation concealment was known to be adequate in only 12 of included 31 trials, and only 11 were blinded. All but two of the included studies did not adhere to the principle of ITT analysis. Larger numbers of patients excluded from the efficacy analysis was due to negative urine cultures after admission, which should be considered as exclusions rather than dropouts, but the high rate of dropouts during the follow‐up was a major problem in many included studies.

Authors' conclusions

Implications for practice.

The present practice of treating uncomplicated UTIs in young women for only three days to achieve symptomatic relief is probably sufficient for the majority of patients. However it leaves a significant risk of recurrent or persistent bacteriuria independent of the class of the drug.

Pending further research, antibiotic treatment for 5‐10 days could be considered for women in whom bacteriological eradication might be of importance: e.g. women suffering from recurrent and painful lower UTIs, planning pregnancy or with underlying disorders. Ultimately the decision regarding therapy duration should be taken with the patient, balancing the higher bacteriological cure rate versus the similar symptomatic outcome and increased risk for adverse events.

The risk of pyelonephritis development as a function of therapy duration is probably irrelevant as it is an extremely rare event in patients with lower UTI.

Implications for research.

We propose that future research in this area should address the question of the link between the bacteriuria and symptomatic UTIs. Future trials should use the same antibiotic drug in the different treatment duration groups to exclude the effect of antibiotic efficacy. It is important to perform antibiotic susceptibility tests during the follow‐up to assess whether duration of the antibiotic therapy influences the rate of resistance development. 
 
 The duration of treatment in special groups of women (i.e. those suffering from recurrent and painful lower UTIs, planning pregnancy, or with underlying disorders) should be addressed in further studies.

What's new

Date Event Description
29 September 2008 Amended Converted to new review format.
Open in a new tab

Acknowledgements

We would like to thank Dr. Alexey G. Dolinin of the Lund University Hospital, Sweden for his help in translating several articles for this review. We would also like to thank Dr Karla Soares‐Weiser who assisted the authors with the drafting of the protocol.

Data and analyses

Comparison 1. Three days versus 5‐10 day antibiotic therapy.

Outcome or subgroup title No. of studies No. of participants Statistical method Effect size
1 Short‐term symptomatic failure (2‐15 days from end of treatment) 24 5165 Risk Ratio (M‐H, Random, 95% CI) 1.06 [0.88, 1.28]
1.1 Same antibiotic therapy in each group 14 2678 Risk Ratio (M‐H, Random, 95% CI) 1.15 [0.95, 1.39]
1.2 Different antibiotic therapy in each group 10 2487 Risk Ratio (M‐H, Random, 95% CI) 0.90 [0.62, 1.29]
2 Short‐term symptomatic failure ‐ ITT (2‐15 days from end of treatment) 17 5029 Risk Ratio (M‐H, Random, 95% CI) 0.98 [0.88, 1.10]
2.1 Same antibiotic therapy in each group 10 2469 Risk Ratio (M‐H, Random, 95% CI) 1.02 [0.89, 1.18]
2.2 Different antibiotic therapy in each group 7 2560 Risk Ratio (M‐H, Random, 95% CI) 0.91 [0.74, 1.12]
3 Long‐term symptomatic failure (4‐10 weeks from end of treatment) 10 3141 Risk Ratio (M‐H, Random, 95% CI) 1.09 [0.94, 1.27]
3.1 Same antibiotic therapy in each group 8 2121 Risk Ratio (M‐H, Random, 95% CI) 1.16 [0.95, 1.42]
3.2 Different antibiotic therapy in each group 2 1020 Risk Ratio (M‐H, Random, 95% CI) 0.97 [0.77, 1.21]
4 Long‐term symptomatic failure ‐ ITT (4‐10 weeks from end of treatment) 10 3910 Risk Ratio (M‐H, Random, 95% CI) 1.07 [0.99, 1.16]
4.1 Same antibiotic therapy in each group 8 2417 Risk Ratio (M‐H, Random, 95% CI) 1.07 [0.95, 1.20]
4.2 Different antibiotic therapy in each group 2 1493 Risk Ratio (M‐H, Random, 95% CI) 1.07 [0.96, 1.19]
5 Short‐term bacteriologic failure (2‐15 days from end of treatment) 31 5368 Risk Ratio (M‐H, Random, 95% CI) 1.19 [0.98, 1.44]
5.1 Same antibiotic therapy in each group 18 3146 Risk Ratio (M‐H, Random, 95% CI) 1.37 [1.07, 1.74]
5.2 Different antibiotic therapy in each group 13 2222 Risk Ratio (M‐H, Random, 95% CI) 0.96 [0.68, 1.35]
6 Short‐term bacteriological failure by antibiotic class (same drug) (2‐15 days from end of treatment) 18 3146 Risk Ratio (M‐H, Random, 95% CI) 1.37 [1.07, 1.74]
6.1 Quinolones 6 1614 Risk Ratio (M‐H, Random, 95% CI) 1.47 [1.01, 2.16]
6.2 Beta‐lactams 7 798 Risk Ratio (M‐H, Random, 95% CI) 1.11 [0.76, 1.63]
6.3 TMP/sulfonamides 5 734 Risk Ratio (M‐H, Random, 95% CI) 1.86 [1.04, 3.34]
7 Short‐term bacteriological failure ‐ ITT (2‐15 days from end of treatment) 20 4163 Risk Ratio (M‐H, Random, 95% CI) 0.92 [0.80, 1.06]
7.1 Same antibiotic therapy in each group 12 2473 Risk Ratio (M‐H, Random, 95% CI) 1.06 [0.87, 1.29]
7.2 Different antibiotic therapy in each group 8 1690 Risk Ratio (M‐H, Random, 95% CI) 0.83 [0.70, 0.98]
8 Long‐term bacteriological failure (4‐10 weeks from end of treatment) 18 3715 Risk Ratio (M‐H, Random, 95% CI) 1.31 [1.08, 1.60]
8.1 Same antibiotic therapy in each group 13 2502 Risk Ratio (M‐H, Random, 95% CI) 1.43 [1.19, 1.73]
8.2 Different antibiotic therapy in each group 5 1213 Risk Ratio (M‐H, Random, 95% CI) 1.13 [0.73, 1.77]
9 Long‐term bacteriological failure by antibiotic class (same drug) (4‐10 weeks from end of treatment) 13 2502 Risk Ratio (M‐H, Random, 95% CI) 1.43 [1.19, 1.73]
9.1 Quinolones 5 1385 Risk Ratio (M‐H, Random, 95% CI) 1.43 [1.04, 1.97]
9.2 Beta‐lactams 3 367 Risk Ratio (M‐H, Random, 95% CI) 1.41 [0.97, 2.06]
9.3 TMP/sulfonamides 5 750 Risk Ratio (M‐H, Random, 95% CI) 1.51 [1.05, 2.18]
10 Long‐term bacteriological failure ‐ ITT (4‐10 weeks from end of treatment) 13 2943 Risk Ratio (M‐H, Random, 95% CI) 1.19 [1.06, 1.35]
10.1 Same antibiotic therapy in each group 10 2127 Risk Ratio (M‐H, Random, 95% CI) 1.26 [1.08, 1.47]
10.2 Different antibiotic therapy in each group 3 816 Risk Ratio (M‐H, Random, 95% CI) 1.18 [0.87, 1.60]
11 Long‐term bacteriological failure ‐ ITT by antibiotic class (same drug) (4‐10 weeks from end of treatment) 10 2127 Risk Ratio (M‐H, Random, 95% CI) 1.26 [1.08, 1.47]
11.1 Quinolones 4 1153 Risk Ratio (M‐H, Random, 95% CI) 1.22 [0.95, 1.56]
11.2 Beta‐lactams 3 421 Risk Ratio (M‐H, Random, 95% CI) 1.26 [0.96, 1.65]
11.3 TMP/sulfonamides 3 553 Risk Ratio (M‐H, Random, 95% CI) 1.32 [0.98, 1.76]
12 Patients with any adverse effects during treatment 29 7617 Risk Ratio (M‐H, Random, 95% CI) 0.83 [0.74, 0.93]
12.1 Same antibiotic therapy in each group 17 3852 Risk Ratio (M‐H, Random, 95% CI) 0.76 [0.63, 0.92]
12.2 Different antibiotic therapy in each group 12 3765 Risk Ratio (M‐H, Random, 95% CI) 0.90 [0.80, 1.03]
13 Patients developed pyelonephritis 5 582 Risk Ratio (M‐H, Random, 95% CI) 3.04 [0.32, 28.93]
13.1 Same antibiotic therapy in each group 3 381 Risk Ratio (M‐H, Random, 95% CI) 3.04 [0.32, 28.93]
13.2 Different antibiotic therapy in each group 2 201 Risk Ratio (M‐H, Random, 95% CI) 0.0 [0.0, 0.0]
14 Adverse effects requiring therapy discontinuation 24 6177 Risk Ratio (M‐H, Random, 95% CI) 0.51 [0.28, 0.91]
14.1 Same antibiotic therapy in each group 13 2817 Risk Ratio (M‐H, Random, 95% CI) 0.35 [0.12, 0.98]
14.2 Different antibiotic therapy in each group 11 3360 Risk Ratio (M‐H, Random, 95% CI) 0.78 [0.45, 1.34]
15 Gastrointestinal adverse effects 24 6973 Risk Ratio (M‐H, Random, 95% CI) 0.81 [0.67, 0.97]
15.1 Same antibiotic therapy in each group 15 3400 Risk Ratio (M‐H, Random, 95% CI) 0.77 [0.61, 0.97]
15.2 Different antibiotic therapy in each group 9 3573 Risk Ratio (M‐H, Random, 95% CI) 0.88 [0.60, 1.29]
16 Skin adverse effects 21 6582 Risk Ratio (M‐H, Random, 95% CI) 0.62 [0.36, 1.06]
16.1 Same antibiotic therapy in each group 14 3305 Risk Ratio (M‐H, Random, 95% CI) 0.51 [0.34, 0.77]
16.2 Different antibiotic therapy in each group 7 3277 Risk Ratio (M‐H, Random, 95% CI) 0.69 [0.21, 2.28]
17 CNS adverse effects 21 5748 Risk Ratio (M‐H, Random, 95% CI) 0.83 [0.65, 1.06]
17.1 Same antibiotic therapy in each group 14 3198 Risk Ratio (M‐H, Random, 95% CI) 0.67 [0.43, 1.04]
17.2 Different antibiotic therapy in each group 7 2550 Risk Ratio (M‐H, Random, 95% CI) 0.91 [0.68, 1.23]
18 Vaginal discharge as an adverse effect of therapy 18 5127 Risk Ratio (M‐H, Random, 95% CI) 0.73 [0.49, 1.10]
18.1 Same antibiotic therapy in each group 11 2304 Risk Ratio (M‐H, Random, 95% CI) 0.75 [0.35, 1.59]
18.2 Different antibiotic therapy in each group 7 2823 Risk Ratio (M‐H, Random, 95% CI) 0.73 [0.45, 1.17]
19 Other adverse effects 19 5250 Risk Ratio (M‐H, Random, 95% CI) 0.98 [0.72, 1.32]
19.1 Same antibiotic therapy in each group 15 3400 Risk Ratio (M‐H, Random, 95% CI) 1.05 [0.71, 1.55]
19.2 Different antibiotic therapy in each group 4 1850 Risk Ratio (M‐H, Random, 95% CI) 0.87 [0.53, 1.42]
20 Patients with any adverse effects during treatment by antibiotic class (same drug) 17 3852 Risk Ratio (M‐H, Random, 95% CI) 0.76 [0.63, 0.92]
20.1 Quinolones 5 1823 Risk Ratio (M‐H, Random, 95% CI) 0.86 [0.72, 1.04]
20.2 Beta‐lactams 8 1258 Risk Ratio (M‐H, Random, 95% CI) 0.82 [0.61, 1.11]
20.3 TMP/sulfonamides 4 771 Risk Ratio (M‐H, Random, 95% CI) 0.40 [0.19, 0.88]
Open in a new tab

1.2. Analysis.

1.2

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 2 Short‐term symptomatic failure ‐ ITT (2‐15 days from end of treatment).

1.4. Analysis.

1.4

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 4 Long‐term symptomatic failure ‐ ITT (4‐10 weeks from end of treatment).

1.18. Analysis.

1.18

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 18 Vaginal discharge as an adverse effect of therapy.

1.19. Analysis.

1.19

Open in a new tab

Comparison 1 Three days versus 5‐10 day antibiotic therapy, Outcome 19 Other adverse effects.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Basista 1991.

Methods Randomisation: computer‐generated 
 Blinding: none 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 40/97 patients (19+21) ‐ 25 of them (14+11) due to negative urine cultures 
 Excluded for safety analysis: 3/97 (2+1) 
 Follow‐up: 4 to 10 days after treatment
Participants USA (8 centers) 
 97 patients (over 90% ‐ female and white) 
 Age: 18‐84 (mean = 33) 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Ofloxacin 200 mg x 1 for 3 days 
 vs 
 TMP‐SMX 160/800 mg x 2 for 7 days
Outcomes Clinical cure (but results not shown) 
 Bacteriological cure 
 Adverse effects
Notes 90% ‐ female and white (the exact number of males not mentioned) 
 Age: 18‐84 
 The trial was terminated early by the sponsor's medical monitor after 97 patients (instead of 150) involved 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Bitsch 1985.

Methods Randomisation: sealed envelopes method 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 84/394 (30 ‐ no urine cultures was taken; 41 ‐ no significant bacteriuria; 13 ‐ dropouts) 
 Follow‐up : 2 days and 10 weeks after end of treatment
Participants Denmark 
 394 patients (92% ‐ non‐pregnant women) 
 Age: 16‐70 (mean = 38) 
 Data collection: 5/81 ‐ 5/82 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam 400 mg x 3 for 3 days 
 vs 
 Sulfametizol 1 g x 2 for 6 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes ˜8% (25 of 310 included in efficacy analysis) were males but results for women with uncomplicated lower UTI only can be separated 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Butler 1983.

Methods Randomisation: randomised list 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for clinical efficacy: 16/141 (12 ‐ lost to follow‐up; 3 ‐ stopped treatment due to side effects; 1 ‐ admitted to hospital due to gastritis) 
 Excluded for bacteriological efficacy analysis: 75/141 (no significant bacteriuria) 
 Follow‐up: 2‐3 days after end of treatment and 4 week after it
Participants UK 
 110 non‐pregnant women 
 Age: 18‐32 (median = 20) 
 Data collection: no information Bacteriuria > 105 CFU/mL
Interventions Nalidixic acid 660 mg + sodium citrate 3.75 g x 3 for 3 days 
 vs 
 TMP/SMX 160/800 mg x 2 for 5 days
Outcomes Clinical cure 
 Bacteriological cure
Notes Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Cox 1992.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 65/202 (39 ‐ diagnosis not confirmed; 9 ‐ resistance or intermediate sensitivity in TMP/SMX group; 7 ‐ no compliance to treatment; 6 ‐ lost to follow‐up; 4 ‐ reasons not reported) 
 Excluded for safety analysis: 2/202 patients 
 Follow‐up: 5‐9 days after treatment
Participants USA 
 202 patients 
 Males: 3 of 137 finally analysed 
 Age: 18‐80 (female) 
 37‐46 (male) 
 Data collection: 2/88 ‐ 10/88 
 Bacteriuria > 105 CFU/mL
Interventions Ofloxacin 200 mg x 1 for 3 days 
 vs 
 TMP/SMX 160/800 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Males not excluded 
 Age of females: 18‐80 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Figueroa 1999.

Methods Randomisation: no information 
 Blinding: No 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Follow‐up: 7‐10 days and 21‐28 days after treatment
Participants Mexico 
 60 non‐pregnant women 
 Age: 18 ‐ 50 
 Bacteriuria > 105 CFU/mL 
 Data collection: no information
Interventions Ceftibuten 400 mg single dose 
 vs 
 Ceftibuten 400 mg x 1 for 3 days 
 vs 
 TPM/SMX 160/800 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Different antibiotics were compared 
 Additional group of patients was studied ‐ a single‐dose of ceftibuten 
 Only short‐term results shown
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Garcia 2002.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 33/151 (5 ‐ bacteria resistant to norfloxacin; 12 ‐ negative cultures; 16 ‐ lost for follow‐up) 
 Follow‐up: 3 days and 30 days after treatment
Participants Spain 
 151 non‐pregnant women 
 Age: above 18 
 Data collection: 1998 ‐ 1999 
 Bacteriuria > 105 CFU/mL
Interventions Norfloxacin 400 mg x 2 for 3 days 
 vs 
 Norfloxacin 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure
Notes Upper age limit not mentioned 
 Only short‐term results shown
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Gordin 1987a.

Methods Randomisation: Latin square method 
 Blinding: No 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded to efficacy analysis: 27/159 (20‐ negative urine cultures; 4‐ lost to follow‐up; 3‐ discontinued treatment due to side effects) 
 Follow‐up : 3‐5 days and 4 weeks after treatment
Participants Finland 
 159 women 
 Age: 17‐63 (mean = 32) 
 Data collection : 9/82 ‐ 10/84 
 Bacteriuria > 105 CFU/mL
Interventions TMP‐sulfadiazine(160 mg + 500 mg) x 2 for 3 days 
 vs 
 TMP‐sulfadiazine(160 mg + 500 mg) x 2 for 10 days
Outcomes Bacteriological cure 
 Adverse effects
Notes 7 of 159 ‐ patients with asymptomatic bacteriuria included 
 A trial with 4 groups was analysed as two separate subtrials
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Gordin 1987b.

Methods Randomisation: Latin square method 
 Blinding: No 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded to efficacy analysis: 27/159 (negative urine cultures (20); lost to follow‐up (4); discontinued treatment due to side effects (3)) 
 Follow‐up: 3‐5 days and 4 weeks after treatment
Participants Finland 
 159 women 
 Age: 17‐63 (mean = 32) 
 Data collection : 9/82 ‐ 10/84 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam 200 mg x 3 for 3 days 
 vs 
 Pivmecillinam 200 mg x 3 for 10 days
Outcomes Bacteriological cure 
 Adverse effects
Notes 7 of 159 ‐ patients with asymptomatic bacteriuria included 
 A trial with 4 groups was analysed as two separate subtrials
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Gossius 1984.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded ‐ 185/464 (143 ‐ negative cultures; 7 ‐ resistant organisms; 11 ‐ lost to follow up; 24 ‐ adverse reactions necessitated cessation of treatment) 
 (Side effects assessed in 408 patients) 
 Follow‐up period: 2 weeks and 6 weeks after treatment
Participants Norway 
 464 women 
 Age: 16‐60 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions TMP‐SMX (160 mg + 800 mg) x 2 for 3 days 
 vs 
 TMP‐SMX( 160 mg + 800 mg) x 2 for 10 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Additional group of patients was studied ‐ a single‐dose TMP‐SMX
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Gossius 1985.

Methods Randomisation: boxes with code numbers and tablets wrapped in plain aluminium foil 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 63/135 (44 ‐ nonsignificant pre‐therapy bacteriuria; 6 ‐ lost to follow up; 2 ‐ initially resistant organisms; 1 ‐ developed pyelonephritis(in 3‐day group); 7 ‐ side effects leading to therapy cessation) 
 Follow‐up: 2 and 6 weeks after treatment
Participants Norway 
 135 women 
 Age: 16 to 60 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions TMP 200 mg x 2 for 3 days 
 vs 
 TMP 200 mg x 2 for 10 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects (for 114 patients who completed treatment)
Notes Clinical response for patients without significant bacteriuria mentioned for total number (not divided for the treatment groups)
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Greenberg 1986.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 15/126 at 3‐days follow‐up visit ; 49/126 at 2‐weeks follow‐up visit 
 Follow‐up: 3 days, 2 weeks and 4 weeks post‐therapy
Participants USA 
 126 non‐pregnant women 
 Age: > 12 
 Data collection: 4/83 ‐ 11/84 
 Bacteriuria > 105 CFU/mL
Interventions Cefadroxil 1 g single dose 
 vs 
 Cefadroxil 500 mg x 2 for 3 days 
 vs 
 Cefadroxil 500 mg x 2 for 7 days 
 vs 
 TMP/SMX 320/1600 mg single dose 
 vs 
 TMP/SMX 160/800 mg x 2 for 3 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Different antibiotics were compared 
 Two additional groups of single dose treatment 
 Only two groups (cefadroxil 500 mg x 2 for 3 vs 7 days) will be analysed here
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Guibert 1997.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: yes 
 Interim analysis: no information 
 Follow‐up: 14 days after end of treatment 
 Excluded to clinical efficacy analysis: 81/421 (non‐compliance to study protocol)
Participants France 
 421 non‐pregnant women 
 Data collection: 12/94 ‐ 6/95 
 Bacteriuria: not defined (case definition by clinical signs and symptoms)
Interventions Lomefloxacin 400 mg x 1 for 3 days 
 vs 
 Norfloxacin 400 mg x 2 for 10 days
Outcomes Clinical cure 
 Adverse effects
Notes Different antibiotics were compared 
 Bacteriuria: not defined
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Hansen 1981.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Follow‐up: 2 days and 8‐10 weeks after end of treatment
Participants Denmark 
 221 patients 
 Women ‐ 92% (non‐pregnant) 
 Age: 16‐80 (mean = 39) 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam 400 mg x 3 for 3 days 
 vs 
 Pivmecillinam 200 mg x 3 for 7 days
Outcomes Bacteriological cure 
 Adverse effects
Notes 8% ‐ males 
 Different antibiotic doses were compared 
 Multicenter trial
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Henry 1999.

Methods Randomisation: allocation numbers generated by the pharmaceutic company; cards with the listing of the medication distributed to the centers 
 Blinding: double‐blinded, double‐dummy 
 Intention‐to‐treat: Yes 
 Interim analysis : no information 
 Excluded to clinical efficacy analysis: 221/1175 
 Excluded to bacteriological efficacy analysis: 685/1175 
 Follow‐up: 13 to 15 days after beginning of the treatment and 4 to 6 weeks after therapy
Participants USA 
 1175 non‐pregnant women 
 Age: 18‐64 (mean = 34) 
 Data collection: 1/94 ‐ 2/95 
 Bacteriuria > 105 CFU/mL
Interventions Sparfloxacin 400 mg single dose 
 vs 
 Sparfloxacin 400 mg on the first day followed by 200 mg x 1 (3 days total) 
 vs 
 Ciprofloxacin 250 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Multicenter trial 
 Additional group of single‐dose drug 
 Higher percentage of patients with previous urinary tract surgery in the 7‐day group 
 More drop‐out in the 7‐day group than in 2 other groups 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Hooton 1991.

Methods Randomisation: computer‐generated randomization list provided by pharmaceutical company, patients allocated sequentially 
 Blinding: none 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded to efficacy analysis: 6/150 (5 ‐ no significant bacteriuria; 1 ‐ no follow‐up) 
 Follow‐up: 5‐9 days and 4‐6 weeks after treatment
Participants USA 
 150 non‐pregnant women 
 Age: > 18 (mean = 24‐25) 
 Data collection : no information 
 Bacteriuria > 10² CFU/mL with symptoms or Bacteriuria > 105 CFU/mL asymptomatic
Interventions Ofloxacin 400 mg single dose 
 vs 
 Ofloxacin 200 mg x 1 for 3 days 
 vs 
 TMP/SMX 160/800 mg x 2 for 7 days
Outcomes Bacteriological cure 
 Adverse effects
Notes Significant bacteriuria defined as > 10² CFU/mL + symptoms or pyuria Asymptomatic bacteriuria treated 
 Different antibiotics were compared 
 Additional single dose group
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Hovelius 1985.

Methods Randomisation: sealed envelopes with treatment protocol 
 Blinding: No 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 38/160 ‐ No significant bacteriuria 
 Follow‐up: 1 week and 4 weeks after treatment
Participants Sweden 
 160 women 
 Age: 15‐45 
 Data collection: no information 
 Bacteriuria > 104 CFU/mL
Interventions Pivmecillinam 400 mg x 3 for 3 days 
 vs 
 Pivmecillinam 200 mg x 3 for 7 days 
 vs 
 Nalidixic acid 1 g x 3 for 3 vs 7 days
Outcomes Bacteriological cure 
 Adverse effects
Notes 1) Only pivmecillinam groups can be analysed due to treatment regimen change in patients of nalidixic acid groups 
 2) Different doses of pivmecillinam were used 
 3) 2 patients with S.saprophyticus <104 CFU included separately 
 4) Age: 15‐45
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Internordic 1988.

Methods Randomisation: no information 
 Blinding: double‐blinded 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for safety analysis: 6/485 patients 
 Excluded for efficacy analysis: 112/485 (84 ‐ no significant bacteriuria; 3 ‐ lost to follow‐up; 8 ‐ treatment less than 13 doses; 17 ‐ others) 
 Follow‐up: "short‐term" ‐ 3 to 13 days after treatment and "accumulated efficacy" ‐ worst result 3 until 45 days after treatment
Participants Norway, Sweden 
 485 non‐pregnant women 
 Age: 18‐80 
 Data collection: 11/85 ‐ 6/87 
 Bacteriuria > 105 CFU/mL for Gram‐negative and 104 for Staphylococcus saprophyticus
Interventions Norfloxacin 400 mg x 2 for 3 days 
 vs 
 Norfloxacin 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Age: 18‐80 
 Multicenter trial
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Iravani 1983.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 12/158 (reasons not mentioned) 
 Follow‐up: 1, 2 and 4 weeks after treatment
Participants USA 
 158 women college students 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Sulfisoxazole 2 g as first dose followed by
Sulfisoxazole 1 g x 4 for 3 days 
 vs 
 Sulfisoxazole 1 g x 4 for 7 days 
 vs 
 Sulfisoxazole 1 g x 4 for 14 days 
 vs 
 Sulfisoxazole 1 g x 4 for 21 days
Outcomes Clinical cure 
 Bacteriological cure
Notes 30 patients had costovertebral tenderness on examination 
 Age not mentioned ("college coeds") 
 Groups of 7, 14 and 21 days will be analysed together ("multi‐days")
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Iravani 1999.

Methods Randomisation: opaque gelatin capsules with medication or placebo 
 Blinding: double blinded 
 Intention‐to‐treat: yes 
 Interim analysis: no information 
 Excluded: 192/713 (128 ‐ negative cultures; 28 ‐ cultures not obtained; 14 ‐ entry criteria violations; 12 ‐ inadequate duration of treatment; 3 ‐ insufficient pretreatment colony counts; 3 ‐ administration of concomitant antibiotics; 2 ‐ noncompliance; 1 ‐ no follow‐up; 1 ‐ resistant organism) 
 Follow‐up: 4‐10 days and 4‐6 weeks after treatment
Participants USA 
 713 women 
 Age: 18‐85 
 Data collection : no information 
 Bacteriuria > 103 CFU/mL
Interventions Ciprofloxacin 100 mg x 2 for 3 days 
 vs 
 TMP‐SMX 160/800 mg x 2 or 
 Nitrofurantoin 100 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Different antibiotics were compared 
 Multicenter trial 
 Age: 18‐85 
 Bacteriuria > 103 CFU/mL 
 Two groups of 7‐days treatment will be analysed together
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Marsh 1980.

Methods Randomisation: randomised list 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for clinical efficacy: 16/141 (12 ‐ lost to follow‐up; 3 ‐ stopped treatment due to side effects; 1 ‐ admitted to hospital due to gastritis) 
 Excluded for bacteriological efficacy analysis: 75/141 (no significant bacteriuria) 
 Follow‐up: 2‐3 days after end of treatment and 4 week after it
Participants UK 
 141 non‐pregnant women 
 Age: 15‐55 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam (dose not mentioned) for 3 days 
 vs 
 Pivmecillinam (dose not mentioned) for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Results of clinical cure are presented in the form of symptom score (mean and range) ‐ cannot be analysed here 
 Doses of antibiotics not mentioned
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Menday 2000.

Methods Randomisation: no information 
 Blinding: double‐blind (double‐dummy technique) 
 Excluded for efficacy analysis: 224/440 (129 ‐ <105 CFU/mL of bacterial pathogen; 37 ‐ inadequate urinary cultures; 54 ‐ bacteria in vitro susceptibility not confirmed; 3 ‐ non‐compliance or concominant antibiotic use; 2 ‐ violated protocol inclusion criteria) 
 Follow‐up: day 10 (+/‐2) and day 14 (+/‐2) from the beginning of treatment
Participants UK 
 440 patients 
 Women: 212 of 216 patients included in efficacy analysis 
 Age: 18‐87 years 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam 200 mg x 3 for 3 days 
 vs 
 Cephalexin 250 mg x 4 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Different antibiotics doses were compared 
 Age: 18‐87 years 
 4 of 216 patients included in efficacy analysis were men 
 Results of clinical cure and improvement are presented together (it`s impossible to separate between them)
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Neringer 1992.

Methods Randomisation: computer‐generated randomisation schedule 
 Blinding: double‐dummy method 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 116/703 (no significant bacteriuria) 
 Follow‐up : 5‐9 days posttreatment and "accumulated results" at 3‐4 weeks posttreatment
Participants Sweden 
 703 non‐pregnant women 
 Age : 18‐65 
 Data collection : 8/88 ‐ 1/90 
 Bacteriuria > 104 CFU/mL
Interventions Lomefloxacin 400 mg x 1 for 3 days 
 vs 
 Lomefloxacin 400 mg x 1 for 7 days 
 vs 
 Norfloxacin 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes One additional group of another antibiotic was included as a 7‐day treatment (norfloxacin) 
 Only two groups (lomefloxacin 400 mg x 1 for 3 vs 7 days) will be analysed here
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Piipo 1990.

Methods Randomisation: no information 
 Blinding: double‐blind 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 73/400 (60 ‐ no significant bacteriuria; 4 ‐ no posttreatment cultures; 4 ‐ change to other antibiotics; 4 ‐ patients did not take drugs as prescribed; 1 ‐ lost for follow‐up) 
 Follow‐up: 3 to 13 days posttreatment and accumulated efficacy (worst result 3 days posttreatment to day 45 after treatment start)
Participants Finland 
 400 non‐pregnant women 
 Age: 18‐80 
 Data collection : no information 
 Bacteriuria > 10^5 CFU/mL (104 for Staphylococcus saprophyticus)
Interventions Norfloxacin 400 mg x 2 for 3 days 
 vs 
 Norfloxacin 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Age: 18‐80 (results for accumulated long‐term efficacy showed for women 18 to 65 years old separately)
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Pitkajarvi 1990.

Methods Randomisation: envelope method 
 Blinding: no information 
 Intention‐to‐treat: yes 
 Interim analysis: none 
 Excluded for clinical and bacteriological effect: 46/345 (no growth in th urine cultures) ‐ 23 in both groups 
 Follow‐up: 5 days and 4‐5 weeks after treatment
Participants Finland 
 345 women 
 Age: 16‐65 (mean=35) 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL (104 for Staphylococcus saprophyticus)
Interventions Pivmecillinam 400 mg x 3 for 3 days 
 vs 
 Pivmecillinam 200 mg x 3 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Different antibiotics doses were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Rapoport 1981.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: none 
 Excluded: all the cases without significant bacteriuria; 16 of 91 with bacteriuria (lost at follow‐up) 
 Follow‐up: 10 to 14 days after treatment
Participants UK 
 187 patients 
 Women: 69 of 75 included in analysis 
 Mean age: 45(14‐78) 
 Data collection: 3/79 ‐ 10/79 
 Bacteriuria > 105 CFU/mL
Interventions TMP‐SMX 2 tabs x 1 for 3 days 
 vs 
 different drugs* for 7 days 
 (* TMP‐SMX ‐ 17, sulfamethizole ‐ 4, sulfadimidine ‐ 4, amoxicillin ‐ 6, mecillinam ‐ 2, nalidixic acid ‐ 2, nitrofurantoin ‐2)
Outcomes Clinical cure 
 Bacteriological cure
Notes Age: 14‐78 years 
 Different antibiotics were compared 
 No outcomes in subgroups of antibiotics in the 7‐days group 
 6 of 75 included in the analysis are males
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Richards 1984.

Methods Randomisation: numbered sealed envelopes (opaque not mentioned) 
 Blinding: Single blinded (investigator) 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded ‐ 8 of 183 (3 ‐ not completed the course due to side effects; 3 ‐ lost to follow‐up; 1 ‐ age > 55; 1 ‐ change in treatment due to worsening symptoms) 
 Follow‐up: 1 week after treatment
Participants UK 
 183 non‐pregnant women 
 Age: 17‐55 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Pivmecillinam 400 mg x 2 for 3 days 
 vs 
 Pivmecillinam 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Multicentre study
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Sandberg 1985.

Methods Randomisation: randomisation tables, sealed opaque envelopes containing the allocation number 
 Blinding: none (open) 
 Intention‐to‐treat: no information 
 Interim analysis: yes (Henning 1982) 
 Excluded : 80/310 (39 ‐ non‐significant bacteriuria; 11 ‐ unknown urine test results; 10 ‐ resistant bacteria; 13 ‐ sensitivity testing for antibiotic not performed; 4 ‐ lost to follow‐up; 1 ‐ male; 2 ‐ known anomalies of urinary tract) 
 Follow‐up: 1 week and 5 weeks after the end of treatment
Participants Sweden 
 310 non‐pregnant women 
 Age: 16‐76 (mean = 35.7) 
 Data collection : 9/81 ‐ 12/82 
 Bacteriuria > 105 CFU/mL (104 for Staphylococcus saprophyticus)
Interventions Cefadroxil 1 g x 1 for 3 days 
 vs 
 Cefadroxil 1 g x 1 for 7 days 
 vs 
 Amoxycillin 375 mg x 3 for 7 days
Outcomes Both clinical and bacteriological cure 
 Adverse effects
Notes Different antibiotics were compared 
 Two groups of 7‐days treatment with different antibiotics were compared with one 3‐days group 
 Only two groups (cefadroxil 1 g x 1 for 3 vs 7 days) will be analysed here
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Stein 1987.

Methods Randomisation: a pre assigned random ‐ number code 
 Blinding: none 
 Intention‐to‐treat: yes 
 Interim analysis: no information 
 Excluded for efficacy analysis: No significant bacteriuria; Not available for follow‐up 
 Follow‐up: 5 to 9 days , 4 to 6 weeks
Participants USA 
 209 patients 
 (192 of 209 ‐ women) 
 Age: 17‐85 
 Data collection: no information 
 Bacteriuria > 105 CFU/mL
Interventions Norfloxacin 400 mg x 2 for 3 days 
 vs 
 TMP‐SMX 160/800 mg x 2 for 10 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Age: 17‐85 
 Different antibiotics were compared 
 17 of 209 patients ‐ males
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Stein 1992.

Methods Randomisation: a reassigned random ‐ number code 
 Blinding: double‐blind 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy evaluation: 184/404 (most common reason ‐ lack of pretherapy urinary pathogen) 
 No drop‐outs to safety analysis 
 Follow‐up: 5 to 9 days after completion of therapy
Participants USA 
 404 non‐pregnant women 
 Age : > 18 mean = 44 ; 81/404 ‐ age 65 or more 
 Data collection: no information 
 Bacteriuria > 104 CFU/mL
Interventions Temafloxacin 400 mg x 1 for 3 days 
 vs 
 Ciprofloxacin 250 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes 81/404 ‐ age 65 or more 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Trienekens 1989.

Methods Randomisation: no information 
 Blinding: double dummy technique, placebo tablets identical to active drug 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Follow‐up: 1, 2 and 6 weeks after entry
Participants The Netherlands 
 327 non‐pregnant women 
 Age: 12‐65 
 Data collection : 1/88 ‐ 4/89 
 Bacteriuria > 105 CFU/mL
Interventions TMP‐SMX 960 mg x 2 for 3 days 
 vs 
 TMP‐SMX 960 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Age: 12‐65
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Trienekens 1993.

Methods Randomisation: the code was supplied by pharmaceutic company and was not known to the investigators, it was kept in the sealed envelopes that was broken 6 weeks after the last patient was included 
 Blinding: double dummy technique, placebo tablets identical to active drug 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 11/395 (not returned for follow‐up) 
 Follow‐up: 1 week and 6 weeks (only for bacteriological cure)
Participants The Netherlands 
 395 non‐pregnant women 
 Age: 18‐65 
 Data collection: 4/89 ‐ 10/90 
 Bacteriuria > 10^5 CFU/mL
Interventions Norfloxacin 400 mg x 2 for 3 days 
 vs 
 Norfloxacin 400 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes  
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Low risk A ‐ Adequate
Open in a new tab

Tsugawa 1999.

Methods Randomisation: no information 
 Blinding: double‐blind 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded for efficacy analysis: 28/99 (14 ‐ no significant bacteriuria; 1 ‐ withdrawal of informed consent; 2 ‐ urinary tract infection within 4 weeks before treatment; 2 ‐ lost for follow‐up; 1 ‐ fungi in urine before therapy; 3 ‐ shortage of dosage; 1 ‐ prohibited medication; 1 ‐ anamnesis of epilepsy; 1 ‐ 71 years or older; 1 ‐ out of target disease) 
 Follow‐up: days 7, 14 and 35 from the treatment start
Participants Japan 
 99 women 
 Age: 20‐70 
 Bacteriuria > 104 CFU/mL
Interventions Gatifloxacin 100 mg x 2 for 3 days 
 vs 
 Gatifloxacin 100 mg x 2 for 7 days
Outcomes Clinical cure 
 Bacteriological cure 
 Adverse effects
Notes Age: 20‐70
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

Winwick 1981.

Methods Randomisation: no information 
 Blinding: no information 
 Intention‐to‐treat: no information 
 Interim analysis: no information 
 Excluded: 23/81 ("not fulfilled the stipulated criteria for entry") 
 Follow‐up: 14 days after treatment start
Participants UK 
 81 non‐pregnant women 
 Age : 18‐65 (mean = 34) 
 Data collection: no information 
 Bacteriuria: no information
Interventions Nalidixic acid + sodium citrate x 3 for 3 days 
 vs 
 Ampicillin 500 mg x 3 for 7 days
Outcomes Bacteriological cure 
 Adverse effects
Notes Exact dosage of antibiotic in the 3‐day group not mentioned (probably ‐ 660 mg + 3.75 g) 
 Different antibiotics were compared
Risk of bias
Bias Authors' judgement Support for judgement
Allocation concealment? Unclear risk B ‐ Unclear
Open in a new tab

TMP = trimethoprim 
 SMX = sulfamethoxazole 
 CFU = colony‐forming units

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion
Aliaev 2005 Not RCT
Bailey 1983 New Zealand 
 RCT 
 Compares 2 treatment regimens 5 days duration both
Bargelloni 1972 Italy 
 Not RCT 
 Phase III trial
Blomer 1986 West Germany 
 Not RCT 
 Review ‐ not systematic
Bonfiglio 2005 Not RCT
Buck 2005 Not duration study
Carmignani 2005 Not duration study
Charlton 1976 UK 
 Quasi‐RCT (alternate months)
Cui 2004 Not duration study
Ejrnaes 2006 Not duration study
Fair 1980 USA 
 Quasi‐RCT (alternate patients)
Fancourt 1984 UK 
 RCT 
 Inpatients only 
 Compares 2 treatment regimens both of 7 days duration (does not include a 3‐day regimen)
Ferry 2004 Not duration study
Fourcroy 2005 Not duration study
Furusawa 1994 Japan 
 Not RCT 
 Case reports
Gellerman 1988 Germany 
 Randomised controlled study 
 Compares single dose and three‐days regimens of ciprofloxacin
Hill 1985 UK 
 RCT 
 Compares 2 treatment regimens 10 days duration both (does not include a 3‐day regimen)
Hoigne 1977 Switzerland 
 Clinical controlled study 
 Compares treatment for 2 weeks and for 4 weeks
Hooton 1989 USA 
 Review of two RCT: 
 1) comparing 2 treatment regimens of 3 days both 
 2) comparing 2 treatment regimens of 7 days both
Iravani 1991 USA 
 Review of 3 different studies (only one of them ‐ RCT and included separately)
Iravani 1995 USA 
 Review of 3 separate studies
Ishihara 1998 Japan 
 RCT 
 Results ‐ only clinical improvement and not cure
Little 1979 New Zealand 
 RCT 
 Compares several treatment regimens all of which were 5 to 7 days long (does not include a 3‐day regimen)
Liu 2004 Not duration study
Liudvig 1996 Germany 
 Not randomised controlled study
Loran 1997 Russia 
 Not RCT (case‐control study)
Martin 1983 UK 
 RCT 
 Compares 2 treatment regimens both of 7 days duration (does not include a 3‐day regimen)
McCarthy 1972 USA 
 RCT 
 Compares 2 treatment regimens10 days duration both
Naber 2004a Not short versus long duration study
Naber 2004b Not short versus long duration study
Naber 2004c Not short versus long duration study
Noorbakhsh 2004 Study in children
Pelta 1985 UK 
 RCT 
 Compares 2 treatment regimens 7 days duration both (does not include a 3‐day regimen)
Petrou 2004 Not RCT
Raz 1996 Israel 
 RCT 
 Only postmenopausal women (mean age = 66 +\‐ 20 years)
Richards 2005 Not duration study
Schultz 1984 USA 
 RCT 
 Compares single‐dose with 10‐days antibiotic regimens (does not include a 3‐day regimen)
Talan 2004 Not duration study
Vogel 1984 UK 
 A summary of few studies comparing different regimens of 3‐days therapy and 7‐days therapy separately (neither comparing 3‐days treatment to 7‐days)
Vogel 2004 Study in older women (> 65 years)
Wells 2004 Complicated UTI in men and women
Zorbas 1995 Greece 
 RCT 
 Duration of all treatment regimens ‐ 12 weeks (does not include a 3‐day regimen) 
 Only elderly patients (mean age = 82.8 years)
Open in a new tab

Contributions of authors

  • Gai Milo: Literature search, obtaining articles, Study selection, quality appraisal, data extraction, data analysis, writing review, updating review.

  • Mical Paul: Study selection, quality appraisal, data extraction, writing review

  • Thierry Christiaens: Data analysis, writing protocol and review.

  • Eugene Katchman: Data analysis, writing protocol and review.

  • Andres Barheim: Data analysis, writing protocol and review.

  • Leonard Leibovici: Data analysis, writing protocol and review.

Declarations of interest

None known

Edited (no change to conclusions)

References

References to studies included in this review

Basista 1991 {published data only}

  1. Basista MP. Randomized study to evaluate efficacy and safety of ofloxacin vs. trimethoprim and sulfamethoxazole in treatment of uncomplicated urinary tract infection. Urology 1991;37(3 Suppl):21‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Bitsch 1985 {published data only}

  1. Bitsch M, Hansen PH, Pagh J. Treatment of acute urinary infections. Comparison between pivmecillinam for 3 days and sulfamethizole therapy for 6 days. Ugeskrift for Laeger 1985;147(17):1392‐5. [MEDLINE: ] [PubMed] [Google Scholar]

Butler 1983 {published data only}

  1. Butler AV, Cullen MJ, Parry MO, Sylvester DG, Speller DC. Acute cystitis in young women. Treatment with citrated nalidixic acid compared with co‐trimoxazole. Practitioner 1983;227(1379):833‐5. [MEDLINE: ] [PubMed] [Google Scholar]

Cox 1992 {published data only}

  1. Cox CE, Serfer HS, Mena HR, Briefer C, Childs SJ, Gordon SF, et al. Ofloxacin versus trimethoprim/sulfamethoxazole in the treatment of uncomplicated urinary tract infection. Clinical Therapeutics 1992;14(3):446‐57. [MEDLINE: ] [PubMed] [Google Scholar]

Figueroa 1999 {published data only}

  1. Figueroa‐Damian R, Arredondo‐Garcia JL. Comparison of the clinical and microbiologic efficacy of single‐dose ceftibuten, 3‐dose ceftibuten, and 7‐day trimethoprim/sulfamethoxazole in the treatment of uncomplicated cystitis. Current Therapeutic Research, Clinical & Experimental 1999;60(7):371‐8. [EMBASE: 1999261432] [Google Scholar]

Garcia 2002 {published data only}

  1. Garcia Bernal G, Fava Aixendri E, Rubio Carque V, Luna Jarque J. Urinary infections without complications: comparison of a treatment with norfloxacin for 7 days versus norfloxacin for 3 days [Infecciones urinarias no complicadas: comparacion de una pauta con norfloxacino durante 7 dias frente a norfloxacino durante 3 dias]. Atencion Primaria 2002;29(1):62‐3. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Gordin 1987a {published data only}

  1. Gordin A, Kalima S, Makela P, Antikainen R. Comparison of three‐ and ten‐day regimens with a sulfadiazine‐ trimethoprim combination and pivmecillinam in acute lower urinary tract infections. Scandinavian Journal of Infectious Diseases 1987;19(1):97‐102. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Gordin 1987b {published data only}

  1. Gordin A, Kalima S, Makela P, Antikainen R. Comparison of three‐ and ten‐day regimens with a sulfadiazine‐ trimethoprim combination and pivmecillinam in acute lower urinary tract infections. Scandinavian Journal of Infectious Diseases 1987;19(1):97‐102. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Gossius 1984 {published data only}

  1. Gossius G, Vorland L. A randomised comparison of single‐dose vs. three‐day and ten‐day therapy with trimethoprim‐sulfamethoxazole for acute cystitis in women. Scandinavian Journal of Infectious Diseases 1984;16(4):373‐9. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]
  2. Gossius G, Vorland L. Treatment of acute cystitis in women. Single‐dose versus a 3‐day and 10‐day therapeutic regimen with trimethoprim‐sulfamethoxazole. Tidsskrift for Den Norske Laegeforening 1986;106(16):1395‐8. [MEDLINE: ] [PubMed] [Google Scholar]

Gossius 1985 {published data only}

  1. Gossius G, Vorland L. The treatment of acute dysuria‐frequency syndrome in adult women: Double‐blind, randomized comparison of three‐day vs ten‐day trimethoprim therapy. Current Therapeutic Research, Clinical & Experimental 1985;37(1):34‐42. [EMBASE: 1985072556] [Google Scholar]

Greenberg 1986 {published data only}

  1. Greenberg RN, Reilly PM, Luppen KL, Weinandt WJ, Ellington LL, Bollinger MR. Randomized study of single‐dose, three‐day, and seven‐day treatment of cystitis in women. Journal of Infectious Diseases 1986;153(2):277‐82. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Guibert 1997 {published data only}

  1. Guibert J, Herman H, Capron MH. Treatment of uncomplicated recurrent cystitis in women: lomefloxacin versus norfloxacin. Fertilite Contraception Sexualite 1997;25(1):79‐84. [MEDLINE: ] [PubMed] [Google Scholar]

Hansen 1981 {published data only}

  1. Hansen PH, Kristensen KH, Lenler‐Eriksen HA, Pagh J, Ostergard JE. Pivmecillinam (Selexid) in acute cystitis. A comparative study of 3‐ and 7‐day treatments. Ugeskrift for Laeger 1981;143(11):670‐3. [MEDLINE: ] [PubMed] [Google Scholar]

Henry 1999 {published data only}

  1. Henry DC, Nenad RC, Iravani A, Tice AD, Mansfield DL, Magner DJ, et al. Comparison of sparfloxacin and ciprofloxacin in the treatment of community‐acquired acute uncomplicated urinary tract infection in women. Sparfloxacin Multicenter Uncomplicated Urinary Tract Infection Study Group. Clinical Therapeutics 1999;21(6):966‐81. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Hooton 1991 {published data only}

  1. Hooton TM, Johnson C, Winter C, Kuwamura L, Rogers ME, Roberts PL, et al. Single‐dose and three‐day regimens of ofloxacin versus trimethoprim‐ sulfamethoxazole for acute cystitis in women. Antimicrobial Agents & Chemotherapy 1991;35(7):1479‐83. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Hovelius 1985 {published data only}

  1. Hovelius B, Mardh PA, Nygaard‐Pedersen L, Wathne B. Nalidixic acid and pivmecillinam for treatment of acute lower urinary tract infections. Scandinavian Journal of Primary Health Care 1985;3(4):227‐32. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Internordic 1988 {published data only}

  1. Anonymous. Double‐blind comparison of 3‐day versus 7‐day treatment with norfloxacin in symptomatic urinary tract infections. The Inter‐Nordic Urinary Tract Infection Study Group. Scandinavian Journal of Infectious Diseases 1988;20(6):619‐24. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Iravani 1983 {published data only}

  1. Iravani A, Pryor ND, Richard GA. Treatment of urinary tract infections with varying regimens of sulfisoxazole. Journal of Urology 1983;130(3):484‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Iravani 1999 {published data only}

  1. Iravani A, Klimberg I, Briefer C, Munera C, Kowalsky SF, Echols RM. A trial comparing low‐dose, short‐course ciprofloxacin and standard 7 day therapy with co‐trimoxazole or nitrofurantoin in the treatment of uncomplicated urinary tract infection. Journal of Antimicrobial Chemotherapy 1999;43 Suppl A:67‐75. [MEDLINE: ] [PubMed] [Google Scholar]

Marsh 1980 {published data only}

  1. Marsh BT, Menday AP. Comparative efficacy of 3‐day and 7‐day chemotherapy with pivmecillinam in urinary tract infections in general practice. Journal of International Medical Research 1980;8(2):105‐11. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Menday 2000 {published data only}

  1. Menday AP. Comparison of pivmecillinam and cephalexin in acute uncomplicated urinary tract infection. International Journal of Antimicrobial Agents 2000;13(3):183‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Neringer 1992 {published data only}

  1. Neringer R, Forsgren A, Hansson C, Ode B. Lomefloxacin versus norfloxacin in the treatment of uncomplicated urinary tract infections: three‐day versus seven‐day treatment. The South Swedish Lolex Study Group. Scandinavian Journal of Infectious Diseases 1992;24(6):773‐80. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Piipo 1990 {published data only}

  1. Piipo T, Pitkajarvi T, Salo SA. Three‐day versus seven‐day treatment with norfloxacin in acute cystitis. Current Therapeutic Research, Clinical & Experimental 1990;47(4):644‐53. [1990140005] [Google Scholar]

Pitkajarvi 1990 {published data only}

  1. Pitkajarvi T, Pyykonen ML, Kannisto K, Piippo T, Viita P. Pivmecillinam treatment in acute cystitis. Three versus seven days study. Arzneimittel‐Forschung 1990;40(10):1156‐8. [MEDLINE: ] [PubMed] [Google Scholar]

Rapoport 1981 {published data only}

  1. Rapoport J, Rees GA, Willmott NJ, Slack RC, O'Grady FW. Treatment of acute urinary tract infection with three doses of co‐trimoxazole. British Medical Journal Clinical Research Ed 1981;283(6302):1302‐3. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Richards 1984 {published data only}

  1. Richards HH. Comparative efficacy of 3‐day and 7‐day chemotherapy with twice‐daily pivmecillinam in urinary tract infections seen in general practice. Current Medical Research & Opinion 1984;9(3):197‐203. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Sandberg 1985 {published data only}

  1. Henning C, Iwarson S, Paulsen O, Sandberg T. Cefadroxil single‐dose long and short therapy versus amoxicillin in female urinary tract infections. Journal of Antimicrobial Chemotherapy 1982;10 Suppl B:73‐6. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]
  2. Sandberg T, Henning C, Iwarson S, Paulsen O. Cefadroxil once daily for three or seven days versus amoxycillin for seven days in uncomplicated urinary tract infections in women. Scandinavian Journal of Infectious Diseases 1985;17(1):83‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Stein 1987 {published data only}

  1. Stein GE, Mummaw N, Goldstein EJ, Boyko EJ, Reller LB, Kurtz TO, et al. A multicenter comparative trial of three‐day norfloxacin vs ten‐day sulfamethoxazole and trimethoprim for the treatment of uncomplicated urinary tract infections. Archives of Internal Medicine 1987;147(10):1760‐2. [MEDLINE: ] [PubMed] [Google Scholar]

Stein 1992 {published data only}

  1. Stein GE, Philip E. Comparison of three‐day temafloxacin with seven‐day ciprofloxacin treatment of urinary tract infections in women. Journal of Family Practice 1992;34(2):180‐4. [MEDLINE: ] [PubMed] [Google Scholar]

Trienekens 1989 {published data only}

  1. Trienekens TA, Stobberingh EE, Winkens RA, Houben AW. Different lengths of treatment with co‐trimoxazole for acute uncomplicated urinary tract infections in women. BMJ 1989;299(6711):1319‐22. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Trienekens 1993 {published data only}

  1. Trienekens TA, London NH, Houben AW, Jong RA, Stobberingh EE. Treating acute urinary tract infections. An RCT of 3‐day versus 7‐day norfloxacin. Canadian Family Physician 1993;39:514‐8. [MEDLINE: ] [PMC free article] [PubMed] [Google Scholar]

Tsugawa 1999 {published data only}

  1. Tsugawa M, Nasu Y, Kumon H, Ohmori H, Nanba K, Kondo K, et al. Comparative study on 3‐day and 7‐day treatment with gatifloxacin in acute uncomplicated cystitis. Japanese Journal of Chemotherapy 1999;47(11):772‐85. [EMBASE: 2000001262] [Google Scholar]

Winwick 1981 {published data only}

  1. Winwick JG, Savage SJ. A comparison of a 3‐day course of Mictral with a 7‐day course of ampicillin in the treatment of urinary tract infection. Journal of International Medical Research 1981;9(1):58‐61. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

References to studies excluded from this review

Aliaev 2005 {published data only}

  1. Aliaev I, Amosov AV, Grigorian VA, Sultanova EA, Krupinov GE, Akopian GN. [A phytogenic drug kanefron H in patients with chronic cystitis and urolithiasis]. [Russian]. Urologiia (Moscow, Russia) 2005, (4):29‐33. [PubMed] [Google Scholar]

Bailey 1983 {published data only}

  1. Bailey RR, Bishop V, Peddie B, Chambers PF, Davies PR, Crofts HG. Comparison of augmentin with co‐trimoxazole for treatment of uncomplicated urinary tract infections. New Zealand Medical Journal 1983;96(744):970‐2. [MEDLINE: ] [PubMed] [Google Scholar]

Bargelloni 1972 {published data only}

  1. Bargelloni U. New treatment of acute urinary tract infections. Minerva Urologica 1972;24(4):140‐4. [MEDLINE: ] [PubMed] [Google Scholar]

Blomer 1986 {published data only}

  1. Blomer R, Bruch K, Zahlten RN. Summarized results of clinical phase II and III studies with ofloxacin (HOE 280) in Europe. Infection 1986;14 Suppl 1:102‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Bonfiglio 2005 {published data only}

  1. Bonfiglio G, Mattina R, Lanzafame A, Cammarata E, Tempera G, Italian Medici Medicina Generale (MMG) Group. Fosfomycin tromethamine in uncomplicated urinary tract infections: a clinical study. Chemotherapy 2005;51(2‐3):162‐166. [DOI] [PubMed] [Google Scholar]

Buck 2005 {published data only}

  1. Buck C, Bertram N, Ackermann T, Sauerbruch T, Derendorf H, Paar WD. Pharmacokinetics of piperacillin‐tazobactam: intermittent dosing versus continuous infusion. International journal of antimicrobial agents 2005;25(1):62‐67. [DOI] [PubMed] [Google Scholar]

Carmignani 2005 {published data only}

  1. Carmignani G, Rose AF, Olivieri L, Salvatori E, Rosignoli MT, Dionisio P. Prulifloxacin versus ciprofloxacin in the treatment of adults with complicated urinary tract infections. Urologia Internationalis 2005;74(4):326‐31. [DOI] [PubMed] [Google Scholar]

Charlton 1976 {published data only}

  1. Charlton CA, Crowther A, Davies JG, Dynes J, Haward MW, Mann PG, et al. Three‐day and ten‐day chemotherapy for urinary tract infections in general practice. British Medical Journal 1976;1(6002):124‐6. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Cui 2004 {published data only}

  1. Cui H, Hou F, Xue F, Li J‐T, Gu J‐M, Wang H‐L, Huo L, Mu A‐P, Xue Y‐W, Yue S‐H, Cai J‐L, Sun Q. [A multicenter single blind randomized controlled clinical trial mezlocillin/sulbactam and pipracillin/tazobactam in the treatment of bacterial infections]. Chinese Journal of Antibiotics 2004;29(2):103‐110. [Google Scholar]

Ejrnaes 2006 {published data only}

  1. Ejrnaes K, Sandvang D, Lundgren B, Ferry S, Holm S, Monsen T, et al. Pulsed‐field gel electrophoresis typing of Escherichia coli strains from samples collected before and after pivmecillinam or placebo treatment of uncomplicated community‐acquired urinary tract infection in women. Journal of Clinical Microbiology 2006;44(5):1776‐81. [DOI] [PMC free article] [PubMed] [Google Scholar]

Fair 1980 {published data only}

  1. Fair WR, Crane DB, Peterson LJ, Dahmer C, Tague B, Amos W. Three‐day treatment of urinary tract infections. Journal of Urology 1980;123(5):717‐21. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Fancourt 1984 {published data only}

  1. Fancourt GJ, Matts SG, Mitchell CJ. Augmentin (amoxycillin‐clavulanic acid) compared with co‐trimoxazole in urinary tract infections. British Medical Journal Clinical Research Ed 1984;289(6437):82‐3. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Ferry 2004 {published data only}

  1. Ferry SA, Holm SE, Stenlund H, Lundholm R, Monsen TJ. The natural course of uncomplicated lower urinary tract infection in women illustrated by a randomized placebo controlled study.[see comment]. Scandinavian journal of infectious diseases 2004;36(4):296‐301. [DOI] [PubMed] [Google Scholar]

Fourcroy 2005 {published data only}

  1. Fourcroy JL, Berner B, Chiang YK, Cramer M, Rowe L, Shore N. Efficacy and safety of a novel once‐daily extended‐release ciprofloxacin tablet formulation for treatment of uncomplicated urinary tract infection in women. Antimicrobial Agents & Chemotherapy 2005;49(10):4137‐4143. [DOI] [PMC free article] [PubMed] [Google Scholar]

Furusawa 1994 {published data only}

  1. Furusawa T, Hiratake Y, Mishina T, Ooe H, Maegawa M, Furudama H, et al. Evaluation of clinical efficacy and safety of cefpodoxime proxetil (CPDX‐PR) in acute uncomplicated cystitis. Hinyokika Kiyo ‐ Acta Urologica Japonica 1994;40(9):853‐60. [MEDLINE: ] [PubMed] [Google Scholar]

Gellerman 1988 {published data only}

  1. Gellermann HJ, Grote J, Peters‐Haertel W, Verbeek H. Short‐term therapy with ciprofloxacin of uncomplicated infections of the urinary tract in female patients [Kurzzeit‐Therapie von unkomplizierten Harnwegsinfektionen der Frau mit Ciprofloxacin]. Medizinische Welt 1988;39(51‐52):1586‐91. [1989012389] [Google Scholar]

Hill 1985 {published data only}

  1. Hill S, Yeates M, Pathy J, Morgan JR. A controlled trial of norfloxacin and amoxycillin in the treatment of uncomplicated urinary tract infection in the elderly. Journal of Antimicrobial Chemotherapy 1985;15(4):505‐6. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Hoigne 1977 {published data only}

  1. Hoigne R, Sturm H, Fahrer H, Spiess J, Patrizzi R. Choice of the therapeutic schedule of cotrimoxazole in urinary tract infections; comparison of the effect with this of trimethoprim alone (author's transl). Schweizerische Rundschau fur Medizin Praxis 1977;66(4):111‐6. [MEDLINE: ] [PubMed] [Google Scholar]

Hooton 1989 {published data only}

  1. Hooton TM, Latham RH, Wong ES, Johnson C, Roberts PL, Stamm WE. Ofloxacin versus trimethoprim‐sulfamethoxazole for treatment of acute cystitis. Antimicrobial Agents & Chemotherapy 1989;33(8):1308‐12. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Iravani 1991 {published data only}

  1. Iravani A. Treatment of uncomplicated urinary tract infections with temafloxacin. American Journal of Medicine 1991;91(6A):124‐8. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Iravani 1995 {published data only}

  1. Iravani A, Tice AD, McCarty J, Sikes DH, Nolen T, Gallis HA, et al. Short‐course ciprofloxacin treatment of acute uncomplicated urinary tract infection in women. The minimum effective dose. The Urinary Tract Infection Study Group [corrected] [see comment][erratum appears in Arch Intern Med 1995 Apr 24;155(8):871]. Archives of Internal Medicine 1995;155(5):485‐94. [MEDLINE: ] [PubMed] [Google Scholar]

Ishihara 1998 {published data only}

  1. Ishihara S, Ban Y, Kawada Y, Ito S, Ito Y, Doi T, et al. Fleroxacin treatment for acute uncomplicated cystitis in women: comparison of 3‐day and 7‐day therapy. Hinyokika Kiyo ‐ Acta Urologica Japonica 1998;44(6):431‐6. [MEDLINE: ] [PubMed] [Google Scholar]

Little 1979 {published data only}

  1. Little PJ, Peddie BA, Sincock A. The treatment of symptomatic urinary tract infection. Australian Family Physician 1979;8(8):895‐7. [MEDLINE: ] [PubMed] [Google Scholar]

Liu 2004 {published data only}

  1. Liu Y‐N, Wang R, Yu B‐X, Cui J‐C, Duan Y‐Y, Gao H, Chen P, Wang X‐D, Zhang J‐P. [A randomized controlled multicentre clinical trial on efficacy and safety of gatifloxacin methanesulfonate tablet in the treatment of acute bacterial infections]. Chinese Journal of Antibiotics 2004;29(9):564‐569. [Google Scholar]

Liudvig 1996 {published data only}

  1. Liudvig G. Clinical experience with the use of ofloxacin in infections of the upper and lower urinary tracts: demonstrations of the results of clinical trials. Antibiotiki i Khimioterapiia 1996;41(9):84‐5. [MEDLINE: ] [PubMed] [Google Scholar]

Loran 1997 {published data only}

  1. Loran OB, Pushkar DU, Tevlin KP. Experience with the use of ciprofloxacin in patients with acute uncomplicated cystitis. Antibiotiki i Khimioterapiia 1997;42(6):42‐4. [MEDLINE: ] [PubMed] [Google Scholar]

Martin 1983 {published data only}

  1. Martin AJ, Lacey RW. A blind comparison of the efficacy and incidence of unwanted effects of trimethoprim and co‐trimoxazole in the treatment of acute infection of the urinary tract in general practice. British Journal of Clinical Practice 1983;37(3):105‐11, inside back cover. [MEDLINE: ] [PubMed] [Google Scholar]

McCarthy 1972 {published data only}

  1. McCarthy CG. Clinical study new short acting sulfanilamide (sulfacytine). Protocol 636‐48. Rocky Mountain Medical Journal 1972;69(5):45‐8. [MEDLINE: ] [PubMed] [Google Scholar]

Naber 2004a {published data only}

  1. Naber KG, Allin DM, Clarysse L, Haworth DA, James IG, Raini C, et al. Gatifloxacin 400 mg as a single shot or 200 mg once daily for 3 days is as effective as ciprofloxacin 250 mg twice daily for the treatment of patients with uncomplicated urinary tract infections. International Journal of Antimicrobial Agents 2004;23(6):596‐605. [DOI] [PubMed] [Google Scholar]

Naber 2004b {published data only}

  1. Naber KG, Bartnicki A, Bischoff W, Hanus M, Milutinovic S, Belle F et a, Gatifloxacin 200 mg or 400 mg once daily is as effective as ciprofloxacin 500 mg twice daily for the treatment of patients with acute pyelonephritis or complicated urinary tract infections. International Journal of Antimicrobial Agents 2004;23 Suppl 1:S41‐S53. [DOI] [PubMed] [Google Scholar]

Naber 2004c {published data only}

  1. Naber KG, Eisenstein BI, Tally FP. Daptomycin versus ciprofloxacin in the treatment of complicated urinary tract infection due to gram‐positive bacteria. Infectious Diseases in Clinical Practice 2004;12(6):322‐7. [Google Scholar]

Noorbakhsh 2004 {published data only}

  1. Noorbakhsh S, Lari AR, Masjedian F, Mostafavi H, Alaghehbandan R. Comparison of intravenous aminoglycoside therapy with switch therapy to cefixime in urinary tract infections. Saudi Medical Journal 2004;25(10):1513‐5. [PubMed] [Google Scholar]

Pelta 1985 {published data only}

  1. Pelta DE, Bowring AR. Management of the urethral syndrome in general practice. Practitioner 1985;229(1399):47‐9. [MEDLINE: ] [PubMed] [Google Scholar]

Petrou 2004 {published data only}

  1. Petrou SP. Urinary urgency and frequency, and chronic urethral and/or pelvic pain in females. Can doxycycline help?. International Braz J Urol 2004;30(4):354‐5. [PubMed] [Google Scholar]

Raz 1996 {published data only}

  1. Raz R, Rozenfeld S. 3‐day course of ofloxacin versus cefalexin in the treatment of urinary tract infections in postmenopausal women. Antimicrobial Agents & Chemotherapy 1996;40(9):2200‐1. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Richards 2005 {published data only}

  1. Richards D, Toop L, Chambers S, Fletcher L. Response to antibiotics of women with symptoms of urinary tract infection but negative dipstick urine test results: double blind randomised controlled trial. BMJ 2005;331(7509):143. [DOI] [PMC free article] [PubMed] [Google Scholar]

Schultz 1984 {published data only}

  1. Schultz HJ, McCaffrey LA, Keys TF, Nobrega FT. Acute cystitis: a prospective study of laboratory tests and duration of therapy. Mayo Clinic Proceedings 1984;59(6):391‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Talan 2004 {published data only}

  1. Talan DA, Klimberg IW, Nicolle LE, Song J, Kowalsky SF, Church DA. Once daily, extended release ciprofloxacin for complicated urinary tract infections and acute uncomplicated pyelonephritis. Journal of Urology 2004;171(2 Pt 1):734‐9. [DOI] [PubMed] [Google Scholar]

Vogel 1984 {published data only}

  1. Vogel R, Deaney NB, Round EM, VandenBurg MJ, Currie WJ. Norfloxacin, amoxycillin, cotrimoxazole and nalidixic acid. A summary of 3‐day and 7‐day therapy studies in the treatment of urinary tract infections. Journal of Antimicrobial Chemotherapy 1984;13 Suppl B:113‐20. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Vogel 2004 {published data only}

  1. Vogel T, Verreault R, Gourdeau M, Morin M, Grenier‐Gosselin L, Rochette L. Optimal duration of antibiotic therapy for uncomplicated urinary tract infection in older women: a double‐blind randomized controlled trial.[see comment]. CMAJ Canadian Medical Association Journal 2004;170(4):469‐73. [PMC free article] [PubMed] [Google Scholar]

Wells 2004 {published data only}

  1. Wells WG, Woods GL, Jiang Q, Gesser RM. Treatment of complicated urinary tract infection in adults: combined analysis of two randomized, double‐blind, multicentre trials comparing ertapenem and ceftriaxone followed by appropriate oral therapy. Journal of Antimicrobial Chemotherapy 2004;53 Suppl 2:ii67‐ii74. [DOI] [PubMed] [Google Scholar]

Zorbas 1995 {published data only}

  1. Zorbas P, Giamarellou H, Staszewska Pistoni M, Petrikkos G, Grammatikou M, et al. Comparison of 2 oral ofloxacin regimens for the treatment of bacteriuria in elderly subjects. Drugs 1995;49 Suppl 2:384‐6. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Additional references

Anderson 2003

  1. Anderson GG, Palermo JJ, Schilling JD, Roth R, Heuser J, Hultgren SJ. Intracellular bacterial biofilm‐like pods in urinary tract infections. Science 2003;301(5629):105‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Arav‐Boger 1994

  1. Arav‐Boger R, Leibovici L, Danon YL. Urinary tract infections with low and high colony counts in young women. Spontaneous remission and single‐dose vs multiple‐day treatment. Archives of Internal Medicine 1994;154(3):300‐4. [MEDLINE: ] [PubMed] [Google Scholar]

Clarke 1999

  1. Clarke M, Oxman AD, editors. The Cochrane Reviewers' Handbook. 4.0. The Cochrane Collaboration, 1999. [Google Scholar]

Higgins 2003

  1. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta‐analyses. BMJ 2003;327(7414):557‐60. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Hooton 1996

  1. Hooton TM. A prospective study of risk factors for symptomatic urinary tract infection in young women. New England Journal of Medicine 1996;335(7):468‐74. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Hooton 1997

  1. Hooton TM, Stamm WE. Diagnosis and treatment of uncomplicated urinary tract infection. Infectious Disease Clinics of North America 1997;11(3):551‐81. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Hooton 2000

  1. Hooton TM, Scholes D, Stapleton AE, Roberts PL, Winter C, Gupta K, et al. A prospective study of asymptomatic bacteriuria in sexually active young women. New England Journal of Medicine 2000;343(14):992‐7. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Kunin 1993

  1. Kunin CM, White LV, Hua TH. A reassessment of the importance of "low‐count" bacteriuria in young women with acute urinary symptoms. Annals of Internal Medicine 1993;119(6):454‐60. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Kunz 1998

  1. Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non‐randomised clinical trials. BMJ 1998;317(7167):1185‐90. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Leibovici 1991

  1. Leibovici L, Wysenbeek AJ. Single‐dose antibiotic treatment for symptomatic urinary tract infections in women: a meta‐analysis of randomized trials. Quarterly Journal of Medicine 1991;78(285):43‐57. [MEDLINE: ] [PubMed] [Google Scholar]

Norrby 1990

  1. Norrby SR. Short‐term treatment of uncomplicated lower urinary tract infections in women. Reviews of Infectious Diseases 1990;12(3):458‐67. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Schulz 1995

  1. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273(5):408‐12. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Stamm 1980

  1. Stamm WE, Wagner KF, Amsel R, Alexander ER, Turck M, Counts GW, et al. Causes of the acute urethral syndrome in women. New England Journal of Medicine 1980;303(8):409‐15. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Stamm 1982

  1. Stamm WE, Counts GW, Running KR, Fihn S, Turck M, Holmes KK. Diagnosis of coliform infection in acutely dysuric women. New England Journal of Medicine 1982;307(8):463‐8. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Stamm 1991

  1. Stamm WE, McKevitt M, Roberts PL, White NJ. Natural history of recurrent urinary tract infections in women. Reviews of Infectious Diseases 1991;13(1):77‐84. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

References to other published versions of this review

Katchman 2005

  1. Katchman EA, Milo G, Paul M, Christiaens T, Baerheim A, Leibovici L. Three‐day vs longer duration of antibiotic treatment for cystitis in women: systematic review and meta‐analysis. American Journal of Medicine 2005;118(11):1196‐207. [MEDLINE: ] [DOI] [PubMed] [Google Scholar]

Milo 2001

  1. Milo G, Katchman E, Christiaens T, Baerheim A, Soares‐Weiser K, Leibovici L. Duration of antibacterial treatment for uncomplicated urinary tract infection in women. Cochrane Database of Systematic Reviews 2001, Issue 3. [DOI: 10.1002/14651858.CD004682] [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Cochrane Database of Systematic Reviews are provided here courtesy of Wiley

RESOURCES