Journal of Antimicrobial Chemotherapy (1999) 43, Suppl.

A, 6775

JAC

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
Abdollah Iravania*, Ira Klimbergb, Caesar Brieferc, Catherine Munerad, Steven F. Kowalskye,
Roger M. Echolse and the Urinary Tract Infection Group
a

Central Florida Medical Research Center, Orlando, FL; bThe Urology Center of Florida, Ocala, FL;
c
University of Michigan Health Service, Ann Arbor, MI; dSynergex Inc., Greenwich, CT;
e
Bayer Corporation, Pharmaceutical Division, West Haven, CT, USA
The study was undertaken to compare the safety and efficacy of twice-daily ciprofloxacin for 3
days with standard 7 day therapy with either co-trimoxazole or nitrofurantoin in the treatment
of women with acute, uncomplicated urinary tract infections (UTI). This multicentre, prospective, randomized, double-blind trial compared oral ciprofloxacin (100 mg bd) for 3 days with
co-trimoxazole (160/800 mg bd) or nitrofurantoin (100 mg bd) for 7 days. Bacteriological and
clinical evaluations were performed at study entry, during therapy and 410 days and 46
weeks after the completion of therapy. The primary efficacy parameter was eradication of the
causative organism 410 days following treatment. Of 713 women enrolled and evaluable for
safety, 521 were evaluable for efficacy (168 ciprofloxacin, 174 co-trimoxazole, 179 nitrofurantoin). Escherichia coli (83%) was the most frequently isolated pathogen in all treatment groups.
Bacteriological eradication was reported in 88% of ciprofloxacin patients, 93% of co-trimoxazole patients and 86% of nitrofurantoin patients. At the 46 week follow-up, ciprofloxacin had
statistically significantly higher eradication rates (91%) than co-trimoxazole (79%; 95% confidence limit (CL) 5 20.6%, 3.9%) and nitrofurantoin (82%; 95% CL 5 17.1%, 0.9%). Clinical
resolution 410 days after therapy and at the 46 week follow-up was similar among the three
treatment groups. The overall incidence of treatment-emergent adverse events was not significantly different (P 5 0.093) among the three drug regimens, although co-trimoxazole was
associated with a greater number of adverse events than ciprofloxacin (P 0.05). Ciprofloxacin
also caused fewer episodes of nausea than either of the other agents (P 0.01).

dose or 3 day antimicrobial regimens.612 However, other

studies have reported failures following single-dose treatment.1315
Short-course antimicrobial therapy offers several theoretical advantages, including patient convenience and
enhanced compliance, fewer potential adverse reactions,
less potential for emergence of resistant bacteria in the
gastrointestinal tract and lower cost.4,5 Suitable antimicrobial agents for short-course therapy should provide
antibacterial activity against Enterobacteriaceae, as well as
Staphylococcus saprophyticus and Enterococcus spp., and
high and sustained urinary concentrations.
Although several studies have demonstrated the efficacy

Introduction
Urinary tract infections (UTI) in women are one of the
most common reasons for seeking medical advice, resulting
in millions of office visits annually in the USA.1 As with
many other infections, the optimal duration of treatment is
unclear.25 The conventional length of therapy for the management of uncomplicated UTI in women has been 714
days, with cure rates usually exceeding 90%. Numerous
studies with -lactam antibiotics and sulphonamides have
subsequently demonstrated that a majority of episodes,
especially those in women with infrequent infections (i.e.
fewer than two per year), respond favourably to single-

*Corresponding address: 615 East Princeton Street, Orlando, FL 32803, USA. Tel: 11-407-898-2811; Fax: 11-407-898-6044.
Current address: Bristol-Myers Squibb, Wallingford, CT, USA. See Acknowledgements.

67
1999 The British Society for Antimicrobial Chemotherapy

A. Iravani et al.
of single-dose ciprofloxacin in the treatment of uncomplicated UTI in women,1619 fluoroquinolones, as for cotrimoxazole, appear to be more effective when given as a
3 day regimen. 20 We recently published results from three
multicentre studies which evaluated single-dose or shortcourse ciprofloxacin therapy in the treatment of acute,
uncomplicated UTI.21 Short-course (either 3 or 5 day)
therapy with ciprofloxacin (100 mg, 250 mg or 500 mg once
or twice daily) was found to be equivalent to conventional
7 day therapy with either ciprofloxacin or norfloxacin. As
previously reported, single-dose ciprofloxacin therapy
was found to be less effective, both bacteriologically and
clinically, than conventional treatment. From this experience, we concluded that ciprofloxacin 100 mg bd for 3 days
appeared to be the minimum effective dose for the treatment of women with uncomplicated cystitis.
The present study, an extension of our previous work,
was designed to demonstrate that low-dose, short-course
ciprofloxacin is as effective as standard 7 day regimens of
either co-trimoxazole or nitrofurantoin for the management of acute cystitis.

tablets containing 160 mg trimethoprim and 800 mg

sulphamethoxazole, and nitrofurantoin (Macrobid; Proctor and Gamble Pharmaceuticals Inc., Norwich, NY, USA)
was supplied as 100 mg capsules. All patients received one
active or placebo capsule twice daily for 7 days. Patients
were advised to take study medication with the morning
and evening meals. Concomitant antimicrobials were not
permitted during the study period. The urinary analgesic,
phenazopyridine, could be administered up to 24 h following study enrolment, but was not permitted thereafter as it
might have interfered with assessment of reduction of
dysuria.
Antimicrobial effectiveness was evaluated by means of
conventional clinical and laboratory determinations,
including serial urine cultures obtained by the midstream
clean-catch technique. Bacteriological response at the
end of therapy (410 days after treatment) was the primary
efficacy variable. Identification of causative organisms and
susceptibility testing were performed according to standard NCCLS procedures. 22 The safety population (intentto-treat) consisted of all patients who received study
medication and returned for a follow-up visit. Safety of
drug treatment was monitored by clinical observations.

Patients and methods

Efficacy evaluation

Adult females with a primary diagnosis of acute, symptomatic, uncomplicated lower UTI were enrolled from 13
centres. Eligible patients had a positive urine culture within
48 h before treatment (i.e. single organism with a colony
count >103 cfu/mL of urine) accompanied by signs and
symptoms of pyuria (positive dipstick or >10 WBC/mm3
(uncentrifuged urine)), dysuria, and urinary frequency of
,10 days duration. Exclusion criteria included allergy
to study drugs, asymptomatic infection (bacteriuria),
pregnancy or lactation, suspected or known bacteraemia,
clinically significant lower or upper urinary tract obstruction, neurogenic bladder (residual volume .250 mL),
indwelling urinary catheter during the course of therapy,
multiple causative organisms, administration of more than
one dose of antacid per day, administration of an investigational agent within 30 days of study entry or a serum
creatinine of >3.0 mg/dL or a creatinine clearance of
,30 mL/min/1.73 m2.
All subjects provided written informed consent
approved by the individual insititutions Ethics Committee.

For a course of therapy to be judged evaluable for drug

efficacy, the following criteria had to be met: (i) infection
diagnosis must have been established by clinical signs and
symptoms; (ii) the single organism causing infection must
have been isolated from a pre-treatment culture in quantitative counts of >103 cfu/mL; (iii) urine cultures and colony
counts must have been repeated 410 days after therapy;
(iv) study drug must have been taken for seven full days
unless treatment was a failure; (v) no other antimicrobial
agent active against the causative organisms could have
been administered concomitantly with the study drug.
Patients with a pre-therapy causative organism resistant to
one of the study drugs were not withdrawn from the study
if they were clinically improving and/or had a negative
urine culture during therapy.
The effectiveness of the study drug was determined on
the basis of the bacteriological response of the infecting
organism and the clinical response of the patient. The
bacteriological response was assessed on the basis of results
of urine cultures performed before and at the end of
therapy and was defined as eradication (causative organism
, 103 cfu/mL 410 days after completion of therapy); persistence (causative organism present at >103 cfu/mL at any
time after the first 2 days of therapy or up to 10 days after
completion of therapy); superinfection (a new infectioncausing organism present at >103 cfu/mL at any time
during therapy or up to 10 days after completion of
therapy and requiring additional antibacterial therapy); or
indeterminate (not efficacy-valid for any reason, e.g. posttreatment culture not obtained). Bacteriological responses

Study design and antimicrobial therapy

This was a prospective, randomized, double-blind, threearm comparative study. Patients were randomly assigned
to one of three treatment groups. Medications (active or
placebo) were encapsulated in opaque gelatin capsules for
blinding purposes. Ciprofloxacin (Cipro; Bayer Corporation Pharmaceutical Division, West Haven, CT, USA) was
supplied as 100 mg tablets. Co-trimoxazole (Bactrim;
Roche Laboratories, Nutley, NJ, USA) was supplied as
68

Low-dose, short-course ciprofloxacin in UTI

at the 46 week follow-up were defined as continued eradication (causative organism at ,103 cfu/mL); eradication
with relapse (causative organism at ,103 cfu/mL 410 days
after completion of therapy but reappearance of the same
organism (>103 cfu/mL) before or at the 46-week followup); or eradication with reinfection (causative organism at
,103 cfu/mL during therapy or 410 days after therapy, but
appearance of a new infecting organism (>103 cfu/mL)
before or at the 46 week follow-up); and indeterminate
(not efficacy-valid for any reason, e.g. post-treatment
culture not obtained).
At completion of therapy, the clinical response was
defined as resolution (disappearance of signs and symptoms related to the infection such that additional antimicrobial therapy was not required), failure (persistence of
the signs and symptoms of infection up to and including the
410 day post-therapy visit, such that alternative antimicrobial therapy was required), or indeterminate (no evaluation possible). At the 46 week follow-up, clinical response
was defined as continued resolution (continued absence
of all signs and symptoms related to infection), relapse
(reappearance of any signs and symptoms related to infection and requiring antimicrobial therapy) or indeterminate
(no evaluation possible).
If the patient was removed from the study before receiving 7 days of therapy, she was not considered valid for efficacy evaluation unless she was a treatment failure or had
therapy discontinued due to an adverse event. All patients
receiving study drug for any length of time were evaluated
for drug safety (intent-to-treat population). Adverse
events were rated by their severity (mild, moderate, severe
or serious/life threatening), and by the relationship to the
study drug (probable, possible, remote or none).
In addition to evaluation of the bacteriological and clinical response of all patients valid for efficacy analysis, an
intent-to-treat analysis was also conducted for all patients
who received study drug (intent-to-treat population).

smallest enrolment were subsequently pooled, as one

centre had only one valid patient in the co-trimoxazole
group; there were 35 efficacy-valid patients included in this
pooled centre. Bacteriological analyses at the end of
treatment were conducted using two comparisons including the primary comparison (eradication vs persistence
plus superinfection) and a secondary comparison (eradication vs persistence).
The treatment equivalence of the efficacy variables was
evaluated using a sequentially rejective multiple testing
procedure. A two-sided 95% CL was initially calculated for
the smaller estimated difference between ciprofloxacin and
one of the active controls. The two treatments were
declared equivalent at the 2.5% significance level if the
upper CL was <0.10. The ciprofloxacin regimen and the
second control regimen were subsequently compared at
the 5% significance level using a 90% CL. Ciprofloxacin
was considered equivalent to both of the control agents if
the 90% upper CL for the larger treatment difference was
<0.10, or superior if it was equivalent to a control drug, and
the upper limit of a 95% CL for the difference between the
two treatments was ,0.
Categorical variables (i.e. race) were analysed using a
CochranMantelHaenszel test, adjusting for a possible
centre effect. For variables of a continuous nature, a twoway analysis of variance (ANOVA) model was fitted with
centre, treatment and treatment-by-centre interaction as
factors. If the treatment-by-centre interaction was not
significant ( 5 0.05), this term was dropped from the final
ANOVA model. The incidence rates of adverse events
were compared across the three treatment groups using a
x2 test or Fishers exact test if .25% of the cells had
expected values of ,5. If the overall P-value for treatment
difference was <0.20, pairwise comparisons were performed.
Logistic regression was used to determine whether any
selected baseline characteristic was a significant predicator
of bacteriological response (eradication vs persistence plus
superinfection). Initially, demographic variables that were
thought to affect bacteriological response, irrespective of
treatment group, were included in logistic regression
models to assess the ability of that variable to influence outcome. Subsequently, each treatment group was analysed
using a similar logistic model.

Statistical analysis
The primary goal of this study was to demonstrate that the
3 day ciprofloxacin regimen was equivalent to the 7 day cotrimoxazole and nitrofurantoin regimens. If ciprofloxacin
was found to be equivalent to one of the control drugs, then
a test to establish a significant difference over the control
drug was performed. Significance testing for determining
clinical and bacteriological equivalence was one-sided with
a 5% significance level. All other significance tests were
two-sided, with an of 5%. Selected demographic variables and both efficacy parameters were also analysed for
the intent-to-treat population.
For the 410 day post-therapy and 46 week follow-up
for bacteriological and clinical responses, the differences
between eradication and resolution rates were estimated
using a CochranMantelHaenszel weighting procedure to
adjust for a possible centre effect. Three centres with the

Results
Study population
Seven hundred and thirteen adult women were enrolled at
13 study sites. A total of 521 (73%) patients were valid for
efficacy (168 ciprofloxacin, 174 co-trimoxazole, 179 nitrofurantoin) and all 713 patients constituted the intent-totreat population. Among the 192 patients not valid for
efficacy, the majority (67% (n 5 128)) were excluded
because no causative organism was isolated before treat69

A. Iravani et al.
ment. Other reasons for patient invalidity included the
following: required cultures not obtained (n 5 28), entry
criteria violations (n 5 14), inadequate duration of treatment (n 5 12), insufficient pretreatment colony counts
(n 5 3), administration of concomitant antimicrobials (n 5
3), noncompliance with study drug regimen (n 5 2), no
follow-up evaluation after enrolment (n 5 1) or organism
resistant to study drug (n 5 1). The rates of, and reasons
for, invalidity were similar among the three study drug
regimens. Fifty-six patients were prematurely discontinued
from their study medication (15 ciprofloxacin, 21 cotrimoxazole, 20 nitrofurantoin). The most frequent reasons
for discontinuation were adverse events (n 5 21 (three
ciprofloxacin, 11 co-trimoxazole, seven nitrofurantoin; P 5
0.092)), protocol violations (n 5 13), lack of pretherapy
organism (n 5 11) and resistant pretherapy organism (n 5
2).
The baseline demographics and medical characteristics
for females enrolled in the three treatment groups are summarized in Table I. There were no statistically significant
differences among treatment groups with respect to race,
age, health status, previous use of antimicrobials or presence of accompanying diseases. The mean age for the 521
efficacy-valid patients was 34 years; nearly 88% were
Caucasian. Approximately one-third of patients reported
no history of previous UTI. The median duration of symptoms for the current UTI episode was 3 days for patients in
all three treatment groups.
Baseline demographics were similar in the intent-totreat population. There were no statistically significant
differences among study drug groups except with respect to
previous antimicrobial use (P 5 0.035); four (2%) patients
in the ciprofloxacin group, six (3%) in the co-trimoxazole
group and none in the nitrofurantoin group reported
receiving an antimicrobial agent before study entry.

Bacteriological response
The bacteriological response rates determined 410 days
after therapy are presented in Table II. Bacteriological
eradication was similar in the three treatment groups
(ciprofloxacin, 88%; co-trimoxazole, 93%; nitrofurantoin,
86%). The rates of bacteriological eradication were not
significantly different among the three treatment groups
when eradication vs persistence plus superinfection was
evaluated (primary comparison). Using the comparison
eradication vs persistence, ciprofloxacin was found to be
equivalent to nitrofurantoin; however, co-trimoxazole
was superior to ciprofloxacin when superinfection was
excluded from the analysis (95% confidence limits (CL) 5
1.2%, 11.2% for the co-trimoxazoleciprofloxacin comparison). Superinfections were reported in four patients
who had their pre-treatment organism eradicated and in
three other patients with a bacteriological response of persistence 410 days after therapy (one ciprofloxacin, four
co-trimoxazole, two nitrofurantoin). The organisms causing superinfections included Proteus mirabilis (one cotrimoxazole, one nitrofurantoin), Streptococcus spp. (one
co-trimoxazole, one nitrofurantoin), Enterococcus faecalis
(one ciprofloxacin), S. saprophyticus (one co-trimoxazole)
and methicillin-resistant Staphylococcus aureus (one cotrimoxazole).
Continued eradication rates at the 46 week posttherapy evaluation were statistically significantly higher in
the ciprofloxacin group than in the other active control
groups (ciprofloxacin, 91%; co-trimoxazole, 79%; nitrofurantoin, 82%) (95% CL 5 20.6%, 3.9% for the cotrimoxazoleciprofloxacin comparison and 95% CL 5
17.1%, 0.9% for the nitrofurantoinciprofloxacin comparison; Table II). Fifty-five patients (10 ciprofloxacin, 26
co-trimoxazole, 19 nitrofurantoin) had a relapse at weeks

Table I. Demographics and baseline medical characteristics for efficacy-valid patientsa

Ciprofloxacin
(n 5 168)
Age, years
mean 6 S.D.
range
Race, n (%)
Caucasian
African American
Hispanic
other
Absence of previous UTI, n (%)
Duration of symptoms
of current UTI (days)
mean 6 S.D.
range
a

34.5 6 16.6
1882
149 (89)
11 (6)
3 (2)
5 (3)
54 (32)

Co-trimoxazole
(n 5 174)
33.4 6 15.3
1885
154 (89)
14 (8)
4 (2)
2 (1)
62 (36)

3.5 6 2.3
110

3.4 6 2.2
110

None of the differences between the treatment groups were statistically significant.

70

Nitrofurantoin
(n 5 179)
34.2 6 16.7
1885
155 (87)
9 (5)
8 (4)
7 (4)
69 (39)
3.6 6 2.3
110

Low-dose, short-course ciprofloxacin in UTI

Table II. Bacteriological response (n/N (%)) following 3 day ciprofloxacin or 7 day co-trimoxazole
or nitrofurantoin therapy
Time of evaluation/response

Ciprofloxacin

410 days after therapy (end of therapy)

eradication
148/168 (88)a,b
persistence
20/168 (12)
superinfection
0
At 46 week follow-upc
continued eradication
118/130 (91)a,b
eradication plus relapse
10/130 (8)
eradication plus reinfection
2/130 (1)

Co-trimoxazole

Nitrofurantoin

161/174 (93)b
10/174 (6)
3/174 (2)

153/177 (86)a
23/177 (13)
1/177 (1)

113/144 (79)b
26/144 (18)
5/144 (3)

115/141 (82)a
19/141 (14)
7/141 (4)

95% Confidence limits (CL) 5 8.4%, 5.6% (410 days after therapy); 95% CL 5 17.1%, 0.9% (follow-up).
90% CL 5 0.6%, 9.9% (410 days after therapy); 95% CL 5 20.6%, 3.9% (follow-up).
c
Denominators are different from those in the 410 days post-therapy category because patients with persistence at end of
therapy, who prematurely discontinued due to an adverse event, or were lost to follow-up (38 in the ciprofloxacin group, 30
in the co-trimoxazole group and 38 in the nitrofurantoin group) are not included.
a

46 with 60 organisms, including Escherichia coli (one

ciprofloxacin, 25 co-trimoxazole, 19 nitrofurantoin); Kleb siella pneumoniae (one ciprofloxacin, one co-trimoxazole);
Citrobacter diversus (one nitrofurantoin); Proteus sp. (one
ciprofloxacin, two co-trimoxazole); Streptococcus viridans
(one ciprofloxacin); and Enterococcus sp. (one ciprofloxacin, one co-trimoxazole). Fourteen patients (two
ciprofloxacin, five co-trimoxazole, seven nitrofurantoin)
were reported to have re-infections caused by 18 organisms, including E. coli (three nitrofurantoin); Klebsiella
sp. (one ciprofloxacin, six co-trimoxazole); P. mirabilis
(one co-trimoxazole, one nitrofurantoin); S. aureus (one
nitrofurantoin); Group B haemolytic streptococci (one
ciprofloxacin); Group D haemolytic streptococci (one cotrimoxazole); and Enterococcus spp. (one co-trimoxazole,
two nitrofurantoin).
A subset analysis of patients with pre-therapy bacterial
colony counts of >105 cfu/mL was also conducted but
excluded patients from one centre where urine colony
counts were all reported as either .104 or <104 cfu/mL.
Seventy-one per cent (371) efficacy-valid patients fit into
this defined subpopulation (124 ciprofloxacin, 126 cotrimoxazole, 121 nitrofurantoin). Bacteriological eradication 410 days after therapy was reported in 87% of
ciprofloxacin patients, 90% of co-trimoxazole patients and
88% of nitrofurantoin patients. Ciprofloxacin was equivalent to both control drugs for the primary comparison (i.e.
bacteriological eradication) in this subset analysis. Similarly, ciprofloxacin was equivalent to nitrofurantoin, but
was not equivalent to co-trimoxazole when the secondary
comparison, eradication vs persistence, was applied (90%
CL 5 0.8%, 11.3%). At the 46 week follow-up evaluation, however, ciprofloxacin was found to be statistically
superior to co-trimoxazole (91% vs 77%, respectively; 95%
CL 5 23.8%, 4.0%) while maintaining equivalence with
nitrofurantoin (83%).
71

The bacteriological response by organism was also evaluated (Table III). In all three treatment groups, E. coli was
the most frequent pathogen causing infection, as well as
the commonest persisting organism. Organism response at
follow-up (eradication vs eradication with relapse plus
persistence) revealed equivalent overall eradication rates
(ciprofloxacin, 82%; co-trimoxazole, 79%; nitrofurantoin,
76%; 90% CL 5 9.3%, 4.8% for the co-trimoxazole
ciprofloxacin regimen and 95% CL 5 13.8%, 3.2% for the
nitrofurantoinciprofloxacin regimen).
To determine whether selected baseline characteristics
influenced bacteriological response, rates of eradication
and persistence plus superinfection were calculated for
each level of the baseline variables and each treatment
group (Table IV). Logistic regression analysis using the
selected baseline characteristics for the entire pool of efficacy-valid patients failed to reveal any statistically significant effect on bacteriological response. Similar logistic
regression models performed on a treatment-by-treatment
basis found in the nitrofurantoin treatment group that
being Caucasian and <30 years of age positively influenced
bacteriological success. None of the baseline characteristics
examined appeared to influence response in the ciprofloxacin and the co-trimoxazole groups.

Clinical response
Clinical response (resolution of symptoms) 410 days after
therapy was reported in 95%, 95% and 93% of patients
receiving ciprofloxacin, co-trimoxazole and nitrofurantoin
regimens, respectively (Table V). Ciprofloxacin was found
to be statistically equivalent to both 7 day co-trimoxazole
and nitrofurantoin therapies. Of the 27 clinical failures, 17
patients had persisting organisms (seven ciprofloxacin,
three co-trimoxazole, seven nitrofurantoin); seven patients
had their pathogen eradicated (one ciprofloxacin, one co-

A. Iravani et al.
Table III. Bacteriological response (n (%)) by organism
Causative organism/response
E. coli
eradication
persistence
K. pneumoniae
eradication
persistence
P. mirabilis
eradication
persistence
Enterobacter spp.
eradication
persistence
Other Gram-negative bacilli
eradication
persistence
S. saphrophyticus
eradication
persistence
Other staphylococci
eradication
persistence
Streptococcus/Enterococcus spp.
eradication
persistence
Other Gram-positive cocci
eradication
persistence

Ciprofloxacin

Co-trimoxazole

Nitrofurantoin

119 (90)
14 (11)

140 (93)
10 (7)

126 (86)
20 (14)

9 (82)
2 (18)

7 (100)
0

5 (100)
0

4 (100)
0

6 (100)
0

3 (60)
2 (40)

5 (63)
3 (37)

2 (100)
0

5 (100)
0

5 (100)
0

4 (100)
0

5 (83)
1 (17)

2 (67)
1 (33)

2 (100)
0

3 (100)
0

2 (100)
0

3 (100)
0

2 (100)
0

2 (100)
0

0
0

4 (100)
0

0
0

0
0

1 (100)
0

Table IV. Bacteriological eradication (n/N (%)) calculated by race or agea

Variable
Race
Caucasian
non-Caucasian
Age (years)
<30
>30

Ciprofloxacin

Co-trimoxazole

Nitrofurantoin

130/149 (87)
18/19 (95)

141/154 (92)
20/20 (100)

137/152 (90)
16/24 (67)

85/95 (90)
63/73 (86)

96/104 (92)
65/70 (93)

95/108 (88)
58/69 (84)

Includes patients valid for efficacy with a response of eradication, persistence, or superinfection.

trimoxazole, five nitrofurantoin); and three patients had

superinfections (two co-trimoxazole, one nitrofurantoin).
Continued clinical resolution at the 46-week follow-up
was reported in 90% of ciprofloxacin patients, 90% of cotrimoxazole patients and 89% of nitrofurantoin patients
and remained statistically equivalent among the three
treatment regimens.
For the subpopulation with >105 cfu/mL of a single
pathogen, clinical response 410 days after therapy ranged

from 93% to 96% for all three treatment groups; ciprofloxacin was statistically equivalent to the two control
drugs. Similar findings were reported at the 46 week
follow-up visit.
In general, intent-to-treat analyses for bacteriological
and clinical responses also demonstrated statistical equivalence between 3 day ciprofloxacin and both standard 7 day
regimens.

72

Low-dose, short-course ciprofloxacin in UTI

Table V. Clinical response (n/N (%)) following 3 day ciprofloxacin or 7 day
co-trimoxazole or nitrofurantoin therapy
Time of evaluation/response

Ciprofloxacin

410 days after therapy (end of therapy)

resolution
160/168 (95)a,b
improvement
0/168 (0)
failure
8/168 (5)
At 46 week follow-up
continued resolution
132/147 (90)a,b
relapse
15/147 (10)

Co-trimoxazole

Nitrofurantoin

165/174 (95)b
3/174 (2)
6/174 (3)

166/179 (93)a
0/179 (0)
13/179 (7)

137/153 (90)b
16/153 (10)

135/151 (89)a
16/151 (11)

a
Resolution plus improvement vs. failure; 95% confidence limits (CL) 5 7.3%, 2.7% (410 days after
therapy); 90% CL 5 5.4%, 6.4% (follow-up).
b
Resolution plus improvement vs. failure; 90% CL 5 2.2%, 5.2% (410 days after therapy); 95% CL 5
7.1%, 7.3% (follow-up).

similar study population and demonstrated treatment

equivalency to co-trimoxazole and nitrofurantoin, two
standard 7 day regimens frequently prescribed for this
infection. Furthermore, it was shown that low-dose, shortcourse ciprofloxacin treatment was better tolerated than
conventional 7 day co-trimoxazole or nitrofurantoin regimens.
In this study, bacteriological outcomes at the end of
therapy were equivalent when comparing the 100 mg bd
3 day ciprofloxacin regimen (88% eradication) with the
standard 7 day co-trimoxazole (93% eradication) or nitrofurantoin (86% eradication) treatment regimen. Ciprofloxacin, however, was shown to be significantly superior to
both control regimens upon examination of short-term
follow-up (46 weeks) bacteriological eradication rates
(i.e. lower rates of reinfection and relapse). Additionally,
bacteriological and clinical equivalence to co-trimoxazole
and nitrofurantoin was achieved for ciprofloxacin when
evaluating the 521 efficacy-valid patients with pre-treatment colony counts of >103 cfu/mL and the subset of valid
patients (n 5 371) who had pre-treatment colony counts of
>105 cfu/mL. Of possible significance, treatment failures
were associated more frequently in non-Caucasian women
over the age of 30 years who received nitrofurantoin.
No demographic factors were associated with treatment
failure for patients given either ciprofloxacin or cotrimoxazole.
This study also proved the contention that low-dose,
short-course ciprofloxacin treatment has a better safety
profile than standard 7 day therapy with either co-trimoxazole or nitrofurantoin. Rates of drug-related nausea and
rash were found to be significantly different among the
three treatment groups. Although the adverse events
reported following ciprofloxacin were similar to other
fluoroquinolones (i.e. headache and nausea), ciprofloxacin
had significantly fewer drug-related episodes of nausea
than co-trimoxazole or nitrofurantoin. There was also a

Safety and adverse events

Seven hundred and thirteen patients enrolled in this multicentre study were valid for safety. Sixty-eight (28%) of the
ciprofloxacin patients, 90 (38%) of the co-trimoxazole
patients and 80 (34%) of the nitrofurantoin patients
reported at least one adverse event (P 5 0.093). The rate of
overall adverse events was significantly higher in the cotrimoxazole group than in the ciprofloxacin group (P
, 0.05). Forty-eight (20%) ciprofloxacin, 63 (26%) cotrimoxazole and 60 (25%) nitrofurantoin recipients experienced at least one drug-related adverse event (P 5 0.215).
Most events were mild to moderate in severity and few
patients required premature discontinuation of study drug
(three ciprofloxacin, 11 co-trimoxazole, seven nitrofurantoin; P 5 0.092).
The most common drug-related adverse event was
nausea, reported in 56 patients (8%). Ciprofloxacin was
associated with statistically significantly fewer episodes of
nausea (3%; 8/239) than were co-trimoxazole (9%; 22/238)
and nitrofurantoin (11%; 26/236; P < 0.05). Headache, the
second most common event (n 5 50; 7%), was reported in
similar rates across the three treatment groups. Rash was
reported more frequently in patients given co-trimoxazole
than in those given nitrofurantoin (4% (9/238) and 0.4%
(1/236), respectively; P < 0.05); three ciprofloxacin patients
(1%) reported episodes of rash. Vaginitis was reported
more often following therapy with co-trimoxazole (2%;
5/238) than with ciprofloxacin (0%; 0/239; P < 0.05).

Discussion
Short-course, low-dose (100 mg bd) ciprofloxacin therapy
has been previously shown to represent the minimum
effective dosage regimen in the treatment of acute, uncomplicated UTI in women.16,21 This study confirmed the efficacy of this low-dose, 3 day ciprofloxacin regimen in a
73

A. Iravani et al.
trend of fewer episodes of rash following ciprofloxacin
(n 5 3) than co-trimoxazole (n 5 8) therapy. Although not
statistically significant, fewer patients receiving profloxacin
had therapy discontinued as a result of an adverse event
than either co-trimoxazole or nitrofurantoin recipients
(three, 11 and seven, respectively).
The establishment of a minimum effective antimicrobial
dose in the treatment of infection, especially when the illness occurs commonly and is not life-threatening, is
paramount to minimizing emergence of resistance, adverse
events and healthcare costs. Previous studies have generally shown an inferior bacteriological success rate with
single-dose regimens than with 7 day regimens in the management of acute, uncomplicated UTI. The 3 day, low-dose
(100 mg bd) ciprofloxacin regimen in this study was bacteriologically and clinically equivalent to standard 7 day
co-trimoxazole and nitrofurantoin regimens and provided
the best safety profile for the treatment of acute cystitis
in women.

4. Hooton, T. M. & Stam, W. E. (1991). Management of acute

uncomplicated urinary tract infection in adults. Medical Clinics of
North America 75, 33957.
5. Johnson, J. R. & Stamm, W. E. (1989). Urinary tract infections in
women: diagnosis and treatment. Annals of Internal Medicine 111,
90617.
6. Charlton, C. A., Crowther, A., Davies, J. G., Dines, J., Jackson,
G. E., Mann, P. G. et al. (1981). Three-day and one-day chemotherapy for urinary tract infections in general practice. Journal of
Antimicrobial Chemotherapy 8, 40912.
7. Iravani, A. (1994). Short-regimen lomefloxacin therapy in urinary
tract infections. Pharmacology and Therapeutics April, 27S34S.
8. Iravani, A. (1993). Multicenter study of single-dose and multipledose fleroxacin versus ciprofloxacin in the treatment of uncomplicated urinary tract infections. American Journal of Medicine 94,
Suppl. 3A, 89S96S.
9. Buckwold, F. J., Ludwig, P., Harding, G. K., Thompson, L.,
Slutchuk, M., Shaw, J. et al. (1982). Therapy for acute cystitis in
adult women. Randomized comparison of single-dose sulfisoxazole
vs trimethoprimsulfamethoxazole. Journal of the American Medical Association 247, 183942.
10. Iravani, A., Pryor, N. D. & Richard, G. A. (1983). Treatment of
urinary tract infections with varying regimens of sulfisoxazole. Journal of Urology 130, 4847.

Acknowledgements
This work was presented in part at the Fifth International
Symposium on New Quinolones, Singapore, August 1994
and at the American College of Clinical Pharmacy,
Orlando, FL, February 1995. We thank Adrienne Block,
Jonathan Harris, Teresa Tartaglione, Amy Plofker and
Joan Hinchcliffe for editorial contributions.
Members of the Urinary Tract Infection Group: Wayne
Harper, Wake Internal Medicine, Raleigh, NC; David
Sikes, East Pasco Specialty Care, Zephyrhills, FL; Thomas
W. Littlejohn, Keith Van Zandt and Susan Donahue, Piedmont Research, Winston-Salem, NC; J. Daniel Scott and
Carol Janes, R/D Clinical Research, Lake Jackson, TX;
Jeffery E. Heck, Charles Margolis, Kathleen Downey,
Philip M. Diller, Robert Burt, Toby Acheson and Mary
Beth VonderMeulen, University of Cincinnati College of
Medicine, Cincinnati, OH; Franklin C. Lowe and Richard
Carlson, New York, NY; Betty Bolick, University of Michigan University Health Service, Ann Arbor, MI; David L.
Williams and Cheryl Benedict, CRCC, Atlantic Institute of
Clinical Research, Daytona Beach, FL; Thomas Nolen,
Columbiana Clinic, Columbiana, AL; William A. Christmas, University of Vermont Student Health Center,
Burlington, VT, USA.

11. Osterberg, E., Aberg, H., Hallander, H. O., Kallner, A. & Lundin,
A. (1990). Efficacy of single-dose versus seven-day trimethoprim
treatment of cystitis in women: a randomized double-blind study.
Journal of Infectious Diseases 161, 9427.
12. Counts, G. W., Stamm, W. E., McKevitt, M., Running, K.,
Holmes, K. K. & Turck, M. (1982). Treatment of cystitis in women
with a single dose of trimethoprimsulfamethoxazole. Reviews of
Infectious Diseases 4, 48490.
13. Fang, L. S., Tolkoff-Rubin, N. E. & Rubin, R. H. (1978). Efficacy
of single-dose and conventional amoxicillin therapy in urinary-tract
infection localized by the antibody-coated bacteria technique. New
England Journal of Medicine 298, 41316.
14. Rubin, R. H., Fang, L. S., Jones, S. R., Munford, R. S., Slepack,
J. M., Varga, P. A. et al. (1980). Single-dose amoxicillin therapy
for urinary tract infection. Multicenter trial using antibody-coated
bacteria localization technique. Journal of the American Medical
Association 244, 5614.
15. Ronald, A. R., Conway, B. & Zhanel, G. G. (1990). The value of
single-dose therapy to diagnose the site of urinary infection. Chemotherapy 36, Suppl. 1, 29.
16. Gellerman, H. J., Grote, J., Peters-Haertel, W. & Verbeek, H.
(1988). Short-course treatment with ciprofloxacin of uncomplicated
infections of the female urinary tract. Medical World News 39,
158691.
17. Garlando, F., Rietiker, S., Tauber, M. G., Flepp, M., Meier, B. &
Luthy, R. (1987). Single-dose ciprofloxacin at 100 versus 250 mg for
treatment of uncomplicated urinary tract infections in women.
Antimicrobial Agents and Chemotherapy 31, 3546.

References
1. Kunin, C. M. (1994). Urinary tract infections in females. Clinical
Infectious Diseases 18, 110.

18. Fischbach, F., Loos, W., Hugo, R. V., Machka, K. & Graeff, H.
(1985). Ciprofloxacin in the treatment of patients with symptomatic
and asymptomatic urinary tract infections: a comparative study of
single-dose application versus three-day treatment concerning
efficacy and safety. In Proceedings of the 14th International Congress of Chemotherapy, Tokyo, Japan. International Society of
Chemotherapy.

2. Iravani, A. (1991). Advances in the understanding and treatment

of urinary tract infections in young women. Urology 37, 50311.
3. Norrby, S. R. (1990). Efficacy and safety of antibiotic treatment in
relation to treatment time. Scandanavian Journal of Infectious Disease 74, Suppl., 2629.

74

Low-dose, short-course ciprofloxacin in UTI

19. Raz, R., Rottensterich, E., Hefter, H., Kennes, Y. & Potasman,
I. (1989). Single-dose ciprofloxacin in the treatment of uncomplicated urinary tract infection in women. European Journal of Clinical
Microbiology and Infectious Diseases 8, 10402.

21. Iravani, A., Tice, A. D., McCarty, J., Sikes, D. H., Nolen, T., Gallis, H. A. et al. (1995). Short-course ciprofloxacin treatment of acute
uncomplicated urinary tract infection in women: the minimum effective dose. Archives of Internal Medicine 155, 48594.

20. Andriole, V. T. (1991). Use of quinolones in treatment of

prostatitis and lower urinary tract infections. European Journal of
Clinical Microbiology and Infectious Diseases 10, 34250.

22. National Committee for Clinical Laboratory Standards. (1993).

Performance Standards for Antimicrobial Disk Susceptibility Tests:
Approved Standard M2-A5. NCCLS, Villanova, PA.

75