Journal of Antimicrobial Chemotherapy (2000) 46, Suppl.

S1, 4952

JAC

Survey on antibiotic usage in the treatment of urinary tract infections

Kurt G. Naber*
Klinikum St Elisabeth Straubing GmbH, St Elisabeth Strasse 23, D-94315, Straubing, Germany
Ninety-two clinical microbiologists, infectious disease clinicians and clinicians with an interest
in the management of urinary tract infections (UTIs) participated in an interactive session
concerning the management of acute uncomplicated lower UTI. The antibacterials considered
most appropriate as first-line agents were trimethoprim, co-trimoxazole, pivmecillinam, nitrofurantoin and fluoroquinolones. The current level of usage of fluoroquinolones for lower UTIs
was, however, considered inappropriate by most delegates from a societal perspective, in
terms of spread of resistance and potential impact on the environment.

Questions were posed by a moderator and the delegates

used a key-pad to select an answer to each question. The
overall response to each question was analysed and presented to the participants for information and discussion.
The answers, which represented the delegates perceived
use of antibacterials, were compared with market data on
actual use as provided by International Medical Statistics
(IMS), Pinner, UK.

Introduction
Guidelines for the appropriate and rational prescribing
of antibacterial agents are being produced increasingly
frequently. While antibiotic-prescribing policies for use
locally in a particular hospital have existed for many years,
newer guidelines are more comprehensive, in that they are
more detailed and are not intended just for local use. They
include not only recommendations concerning which antibacterials to prescribe, but also appropriate doses, duration
of therapy and broader aspects of management, including
laboratory investigations.13 Current guidelines for the
management of urinary tract infections (UTIs) are summarized elsewhere in this supplement.4
With the increasing promulgation of guidelines, it is
instructive to compare expert opinion on disease management with the actual usage of antimicrobials in practice.
This report summarizes the results of an interactive session
involving delegates at an international symposium on
UTIs.

Results
Frequency of acute uncomplicated lower UTI
Before discussing the specific treatment of acute uncomplicated lower UTI, delegates were asked whether the number of prescriptions for this condition per annum per
million inhabitants in western Europe and Canada was
c.10 000, c.100 000 or c.250 000 prescriptions. In fact, in
1999, there were 46 million prescriptions per annum for
acute uncomplicated lower UTI in western Europe and
Canada, corresponding to a mean of 110 000 prescriptions
per million inhabitants per annum. The delegates showed a
good awareness of this: 14 (15%) of the delegates thought
there were c.10 000 prescriptions/year, 59 (64%) thought
there were c.100 000/year and 19 (21%) thought there were
c.250 000 prescriptions/year.

Materials and methods

Ninety-two symposium delegates, including microbiologists,
infectious disease specialists and clinicians with an interest in
therapy of UTIs, participated. This was an international
group, 34 (37%) of the delegates coming from Nordic countries (Denmark, Norway, Sweden or Finland), 19 (21%)
from the UK or Eire, 13 (14%) from Germany, Austria or
Holland, 10 (11%) from Canada, eight (9%) from France,
Switzerland or Belgium, three (3%) from Italy, Spain,
Portugal or Greece and five (5%) from other countries.

First-line empirical treatment of acute uncomplicated

lower UTI
Symposium delegates were asked to indicate which of the
following was their first-line empirical therapy for acute

*Corresponding author. Tel: 49-94-21-710-1700; Fax: 49-94-21-710-1015; E-mail: naberk@klinikum-straubing.de

49
2000 The British Society for Antimicrobial Chemotherapy

K. G. Naber
trimethoprim, co-trimoxazole, pivmecillinam, nitrofurantoin, norfloxacin or ciprofloxacin. Delegates believed that
co-trimoxazole and trimethoprim were the most popular
treatments in Europe and Canada (Table II). This was not
the case in 1998/9 (Table II). While there was considerable
use of these agents, the quinolones, particularly norfloxacin
and ciprofloxacin, were the most popular antibacterial
agents in the treatment of acute uncomplicated lower UTI
in 1999.

uncomplicated lower UTI: trimethoprim, co-trimoxazole,

pivmecillinam, nitrofurantoin, fluoroquinolones, cephalosporins or others. Twenty-six (30%) of the delegates
chose trimethoprim, 18 (19%) pivmecillinam, 17 (18%) cotrimoxazole, 10 (11%) nitrofurantoin, nine (10%) fluoroquinolones, three (3%) cephalosporins and nine (10%)
others.

Antibiotic treatment for acute uncomplicated lower

UTI

Pivmecillinam usage in Nordic countries

Delegates were asked to estimate the proportion of patients

in their country with acute uncomplicated lower UTI
treated with fluoroquinolones. The options provided were
010%, 1120%, 2130%, 3140%, 4150% or 5171%.
There was a broad range of estimates concerning fluoroquinolone usage in acute uncomplicated UTI. Forty-two
(46%) of the delegates estimated that fluoroquinolones
were used in 10% of patients with acute uncomplicated
lower UTIs. Seventeen (18%) estimated their usage in
1120%, 13 (14%) in 2130%, 13 (14%) in 3140% and
seven (8%) in 50% of patients with acute uncomplicated
lower UTIs.
The actual usage of antibiotics, expressed as a percentage
of the total number of prescriptions for acute uncomplicated
lower UTI, is shown in Table I. Fluoroquinolones (and other
quinolones) are very widely used in the treatment of acute
uncomplicated lower UTI. They are particularly heavily
prescribed in Italy, Spain, Portugal, France, Switzerland and
Belgium. There is less, but still considerable, use in Greece,
Germany, Austria, The Netherlands and Canada. In the UK
and Finland, use of these agents is lower.
Delegates were asked to state which of the following was
the most frequently used antibiotic in western Europe and
Canada for the treatment of acute uncomplicated UTI:

Pivmecillinam is a widely used treatment for acute uncomplicated lower UTI in Nordic countries. Participants were
asked to estimate whether 510%, 1120%, 2125%,
2630% or 30% of patients with acute uncomplicated
UTI were treated with pivmecillinam in Scandinavia
(Norway, Denmark and Sweden). Sixty-two delegates
responded to this question; eight delegates (13%) estimated that 510% of patients with lower UTIs would be
prescribed pivmecillinam, 23 (37%) chose the option
1120%, 11 (18%) chose 2125%, eight (13%) chose
2630% and 12 (19%) chose 30% . Thus the delegates
estimates of pivmecillinam usage in Scandinavia varied
widely, between 510% of patients and 30% of patients.
The actual usage of pivmecillinam is shown in the Figure.
In Scandinavia, between 15% and 30% of patients with
acute uncomplicated UTI are treated with pivmecillinam.

Length of treatment for acute uncomplicated lower

UTI
Delegates were asked to choose whether the current average length of treatment with co-trimoxazole, ciprofloxacin
and fosfomycin trometamol for UTI was 13 days, 57

Table I. Antibacterials used (percentage of prescriptions) in the treatment of uncomplicated UTI. Figures are derived
from data provided by International Medical Statistics, Pinner, UK
Country
Antibacterial

IT

ES

PT

GR

FR

CH

BE

DE

AT

NL

UK

FI

CA

(Fluoro)quinolone
Trimethoprim
Co-trimoxazole
Broad-spectrum penicillin
Cephalosporin
Fosfomycin
Nitrofurantoin
Pivmecillinam
Sulphonamide
Other

65
0
5
3
4
18
2
0
0
3

73
0
4
6
6
8
1
0
0
2

61
0
19
8
3
2
6
1
0
0

37
0
17
7
24
1
2
0
0
12

57
0
9
6
4
16
7
1
0
0

64
0
21
5
2
5
2
0
0
1

63
0
5
5
1
10
12
0
0
4

30
1
46
2
3
2
9
0
6
1

40
17
12
8
10
8
2
1
0
2

28
27
10
7
0
0
27
0
0
1

7
56
0
12
19
0
6
0
0
0

12
35
8
1
6
2
18
16
0
2

33
1
34
6
4
0
21
0
0
1

IT, Italy; ES, Spain; PT, Portugal; GR, Greece; FR, France; CH, Switzerland; BE, Belgium; DE, Germany; AT, Austria; NL, The Netherlands;
UK, United Kingdom; FI, Finland; CA, Canada.

50

Survey on antibiotic usage in treating UTIs

days, 810 days or 1114 days. As expected, delegates
believed that short (3 days) courses of a fluoroquinolone,
such as ciprofloxacin, would be used (Table III). Results for
fosfomycin, for which single-dose treatment is advocated,
were similar (Table III).

on the agents actually used, delegates were asked whether

they felt that the current usage of fluoroquinolones in acute
uncomplicated lower UTI was good or bad from a
societal perspective, in terms of spread of resistance and
impact on the environment. Seventy-seven (84%) of the
delegates considered the current usage of fluoroquinolones
to be bad and 10 (11%) felt it was good. Five delegates
(5%) did not express an opinion.

Attitudes to the use of fluoroquinolones in acute

uncomplicated lower UTI
After expressing their views concerning appropriate antibacterial therapy in uncomplicated UTI and observing data

First-line empirical treatment for acute uncomplicated

lower UTI
At the end of the session, delegates were asked again to
select, from the same choices as previously, the agent they
would recommend as first-line empirical treatment for acute
uncomplicated lower UTI. Thirty-two delegates (35%)
chose trimethoprim, 27 (29%) pivmecillinam, 11 (12%)
nitrofurantoin, 10 (11%) co-trimoxazole, two (2%) fluoroquinolones, two (2%) cephalosporins and eight (9%)
others.

Discussion
There was a preponderance of delegates from the Nordic
countries, so the responses may show a bias towards the
attitudes of Scandinavian microbiologists and infectious
disease specialists. The agents recommended by symposium delegates as first-line empirical therapy for acute
uncomplicated lower UTI were trimethoprim, pivmecillinam, co-trimoxazole, nitrofurantoin, fluoroquinolones
and cephalosporins. The inclusion of pivmecillinam in this

Figure. Pivmecillinam usage for uncomplicated UTI in

Norway ( - ), Denmark (- - -) and Sweden ().

Table II. Most popular treatments for uncomplicated UTI in western Europe and Canada.a Data were obtained from
International Medical Statistics, Pinner, UK
Actual usage
total number of
prescriptions
(103)
Norfloxacin
Co-trimoxazole
Ciprofloxacin
Trimethoprim
Fosfomycinc
Nitrofurantoin
Ofloxacinc
Pipemidic acidc
Amoxycillin  clavulanic acidc
Lomefloxacinc
Pivmecillinam

7764
5600
4671
2807
2833
2505
1427
1416
1112
887
no data

percentage
change
1998 to 1999

percentage of
prescriptions for
uncomplicated UTI

number (%) of delegatesb

selecting as most widely
used treatment for UTI

3.9
0.4
9.8
2.1
16.0
0.8
13.1
2.1
0.9
15.3
no data

20.7
14.9
12.4
7.5
7.5
6.7
3.8
3.8
3.0
2.4
no data

16 (18%)
35 (39%)
16 (18%)
21 (24%)

1 (1%)

Data obtained from Austria, Belgium, Canada, Finland, France, Germany, Greece, Italy, The Netherlands, Portugal, Spain, Switzerland and UK.
Eighty-nine of 92 delegates expressed an opinion.
c
Participants were not asked to comment on usage of these agents.
b

51

K. G. Naber
Table III. Estimated average length of antibiotic treatment in acute cystitis
Number (%) delegates selecting average length of treatment with (days)
Treatment
duration (days)
13
57
810
1114
Others

co-trimoxazolea

ciprofloxacinb

fosfomycin trometamolc

42 (58%)
28 (38%)
3 (4%)
0 (0%)
0 (0%)

72 (90%)
3 (4%)
0 (0%)
0 (0%)
7 (6%)

29 (36%)
48 (60%)
3 (4%)
0 (0%)
0 (0%)

Eighty of 92 delegates expressed an opinion.

Seventy-three of 92 delegates expressed an opinion.
c
Eighty-two of 92 delegates expressed an opinion.
b

list reflects the above-mentioned bias, as pivmecillinam is

not used in most of Europe.
Actual fluoroquinolone use was much greater than most
delegates expected. There was a general concern amongst
participants about the amount of fluoroquinolone usage for
uncomplicated UTI. Nearly 90% of the microbiologists and
infectious disease clinicans in the survey felt that the level
of current usage was inappropriate from a societal perspective. Although there was general agreement that these are
effective for treating UTI, it was a general view that these
agents were inappropriate first-line treatments, because of
the risk of increased resistance, and that they should be
reserved for complicated and refractory UTI and pyelonephritis.
The general trend towards a shorter duration of therapy
for acute uncomplicated lower UTI5 was endorsed by this
survey of expert opinion.
The results of the survey were somewhat thoughtprovoking to participants, particularly regarding what
many felt was excessive use of fluoroquinolones. This was
apparent when delegates were asked again at the end of the
session to recommend first-line empirical treatment for
acute uncomplicated UTI. On the second occasion, there
was an increase in the recommendation of pivmecillinam
as a first-line treatment. The number of delegates who
recommended fluoroquinolones for empirical use in UTI

decreased from 10% to 2% after the delegates had seen

data on the actual usage of these agents. This attitude is in
general agreement with that of Piddock,6 who recommended limited use of fluoroquinolones for first-line
therapy in order to prevent the spread of resistance in
pathogenic bacteria.

References
1. Hooton, T. M. (1990). Practice guidelines for urinary tract infection in the era of managed care. International Journal of Antimicrobial Agents 11, 2415.
2. Warren, J. W., Arbutyn, E., Hebel, J. R., Johnson, J., Schaeffer,
A. J. & Stamm, W. E. (1999). Guidelines for the antimicrobial treatment of uncomplicated acute bacterial cystitis and acute
pyelonephritis in women. Clinical Infectious Diseases 29, 7458.
3. Henning, C. & Bengtsson, L. (1997). The treatment of acute urinary tract problems. Lakartidningen 94, 238790.
4. Naber, K. G. (2000). Treatment options for acute uncomplicated
cystitis in adults. Journal of Antimicrobial Chemotherapy 46, Suppl.
1, 237.
5. Tice, A. D. (1999). Short-course therapy of acute cystitis: a brief
review of therapeutic strategies. Journal of Antimicrobial Chemotherapy 43, Suppl. A, 8593.
6. Piddock, L. J. V. (1998). Fluoroquinolone resistance. British
Medical Journal 317, 102930.

52