Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Antibiotics for Cystitis: How Long?

Systematic review
Symptomatic failure
Bacteriologic failure
Adverse events

Cystitis is relatively common in younger women, affecting perhaps 1 in 200 every year. It causes dysuria, urgency, and frequency. While shorter courses of antibiotic therapy tend to be the norm, because of the small size of many of the trials, benefits of longer duration of antibiotic therapy might have been missed. A systematic review [1] of shorter versus longer duration suggests that, while there are some differences, they are relatively minor.

Systematic review

The review looked for randomised trials examining three-day oral antibiotic therapy with five-day or longer regimens, to mid-2003. Women had to be otherwise healthy with uncomplicated urinary tract infections defined by the triad of symptoms and the absence of signs of upper urinary tract infection. Studies on pregnant or immunocompromised women, or those with diabetes, hospital acquired infection, or indwelling urinary catheter were excluded.

Outcomes were symptomatic failure, defined by the presence of cystitis symptoms, within eight weeks (longer outcomes) or two weeks (shorter outcomes), or bacteriological failure (positive urine culture within eight or two weeks). Additional outcomes were women with any adverse event, or who discontinued treatment. Analysis examined both studies with the same antibiotic in shorter and longer treatment, and different antibiotics.


Randomised trials included in the review were performed between 1980 and 2002, and 9,605 women were included in 32 trials. Nineteen compared the same antibiotic, and 14 different antibiotics.

Symptomatic failure

Symptomatic failure rates averaged 10% over two weeks, and 20% over eight weeks, but there was no difference between three-day therapy with antibiotics and longer regiments (Table 1; Figures 1 and 2).

Table 1: Outcomes of symptomatic and bacteriological failure of 3-day compared with longer therapy, as well as any woman experiencing an adverse event, and discontinuation

Number /total
Percent with
Relative risk
(95% CI)
(95% CI)
Symptomatic failure over 8 weeks
1.1 (0.99 to 1.3)
not calculated
Symptomatic failure over 2 weeks
1.1 (0.9 to 1.2)
not calculated
Bacteriological failure over 8 weeks
1.3 (1.1 to 1.5)
35 (19 to 201)
Bacteriological failure over 2 weeks
1.2 (1.0 to 1.4)
not calculated
Patients with any adverse event
0.83 (79 to 0.91
23 (17 to 39)
0.5 (0.4 to 0.7)
10 (8-15)
NNH in bold, shaded box

Figure 1: Symptomatic failure over two weeks with 3-day compared with longer antibiotic therapy

Figure 2: Symptomatic failure over eight weeks with 3-day compared with longer antibiotic therapy

Analyses were not affected by comparisons of the same or different antibiotics. Analyses for particular classes of antibiotics involved smaller numbers in subgroups, but none showed any difference for duration of therapy.

Older women whose mean age was above 42 years (the median for included trials) were represented in three trials and 765 women. In this subgroup here was a significantly greater failure rate with three-day antibiotic regimens.

Bacteriologic failure

Bacteriologic failure rates averaged 7% over two weeks, and 16% over eight weeks. Shorter duration therapy led to significantly higher bacteriological failure over eight weeks (Table 1). For every 35 women treated with three days of antibiotics compared with a longer duration, one more would have had a positive culture over the eight weeks following.

Adverse events

With the shorter three-day course of antibiotics, fewer women had any adverse event, and for every ten women treated with a three-day rather than a longer course, there was one fewer discontinuation.


What we have here is a degree of comfort that three-day oral antibiotic regimens will do as good a job as longer duration therapies most of the time in uncomplicated cystitis in mainly younger women. In older women, or where more effective bacteriological success is needed, longer duration therapy with antibiotic does better, but at a cost of some more adverse events.


  1. EA Katchman et al. Three-day vs longer duration of antibiotic treatment for cystitis in women: systematic review and meta-analysis. American Journal of Medicine 2005 118: 1196-1207.

previous or next story