The Author (2005). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.

org

doi:10.1093/fampra/cmh720

Family Practice Advance Access originally published online on 7 January 2005

Management of urinary tract infections in female general practice patients

Eva Hummers-Pradiera, Ann Marit Ohsea, Matthias Kochb, Wolfgang R Heizmannc and Michael M Kochena
Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012

Hummers-Pradier E, Ohse AM, Koch M, Heizmann WR and Kochen MM. Management of urinary tract infections in female general practice patients. Family Practice 2005; 22: 7177. Background. Though guidelines for the management of urinary tract infections (UTI) exist in several European countries, little is known about GPs adherence, and the appropriateness of their management with regard to antibiotic resistance. Objectives. To describe German GPs management of female patients with symptoms of UTI, to assess the diagnostic accuracy of dipsticks in a German general practice setting, to develop diagnostic prediction rules for culture-confirmed UTI, and to compare the adequacy of empirical treatment strategies and GPs actual prescriptions. Methods. In 36 (of 118 invited) teaching general practices, urine cultures and resistance testing were performed during 4 months on all symptomatic patients. GPs completed a questionnaire on each patients symptoms, risk factors and treatment. Adequacy of different treatment approaches was calculated based on culture results. Results. 445 adult women (76% of all patients) were included, with a median age of 53 years. Complicating factors were present in 27%. Urine culture revealed UTI in 77%. GPs diagnostic accuracy, using both dipsticks and clinical impressions, was low. A positive nitrite test, dysuria and older age were the only predictive factors of culture-confirmed UTI, however the negative predictive value of dipsticks is low (35%). Empirical treatment of all symptomatic patients with either nitrofurantoin or fluoroquinolones would result in a higher rate of appropriate therapies than the individualized approach chosen by the GPs. Conclusion. Most patients with urinary symptoms were not treated according to current guidelines, and GPs diagnostic and therapeutic accuracy was low. Empirical treatment of all symptomatic patients is probably the most effective policy, but implies unnecessary antibiotic prescriptions. Keywords. Anti-bacterial agents/therapeutic use, family practice, female, physicians practice patterns, sensitivity and specificity, urinary tract infection.

Introduction
Urinary tract infections (UTIs) are common in women, and a frequent reason to prescribe antibiotics in general practice. Guidelines on management have been published in several European countries, and in the US. Recommendations for the decision to treat and first line antibiotics are summarized in Table 1.16
Received 10 June 2004; Accepted 27 Spetember 2004. aDepartment of General Practice, University of Gttingen, Humboldtallee 38, 37073 Gttingen, bMedical partnership Wagner Stibbe Kast Bispink & Partner, Werner-von-SiemensStr. 10, 37077 Gttingen and cInstitute for Microbiology and Infectiology, Mahlower Str. 24, 12049 Berlin, Germany. Correspondence to Dr Eva Hummers-Pradier, Department of General Practice, University of Gttingen, Humboldtallee 38, 37073 Gttingen, Germany; Email: ehummer@gwdg.de

Empiric treatment with first choice antibiotics (trimethoprim, nitrofurantoin, in the US: cotrimoxazole) is usually recommended, either for all symptomatic women without additional risk factors, or after using dipsticks to identify patients with a high probability of UTI. However, recently, there has been some discussion on appropriate diagnosis and targeting of antiobiotic prescriptions to contain rising resistance levels in urinary pathogens.79 The aims of our study were: to describe German GPs management of female patients with symptoms of UTI; to assess the diagnostic accuracy of dipsticks in a German general practice setting; to develop diagnostic prediction rules for a microbiologically confirmed UTI; 71

72
TABLE 1 Country Germany1 The Netherlands2

Family Practicean international journal

International guidelines on management of uncomplicated urinary tract infections in women Treat in presence of Symptoms and both nitrite and leucocytes on dipstick Symptoms and nitrite on dipstick or dipstick negative but bacteria on sediment microscopy or dipslide Symptoms (urine examination not recommended) Symptoms and either leucocytes or nitrite on dipstick or leucocytes and bacteria in native urine microscopy Symptoms and history of UTI (phone management) or symptoms and leucocytes in dipstick or microscopy Symptoms (phone management) or symptoms and pyuria First choice drugs (empical treatment) Trimethoprim or nitrofurantoin Nitrofurantoin or trimethoprim Trimethoprim or nitrofurantoin Trimethoprim or nitrofurantoin or cefalexin Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012 Cotrimoxazole Cotrimoxazole or trimethoprim

Norway3 UK4 USA/Canada5 USA6

to estimate the appropriateness of different approaches: empiric therapy of all symptomatic patients or treatment decisions based on dipstick results; and to assess the appropriateness of GPs therapy with regard to antibiotic resistance.

Methods
This survey is part of a larger study on urinary tract infection in Germany, and methods have already been described elsewhere.10 The local ethics review board had no objections to the study. All 118 teaching general practices of the Department of General Practice, University of Gttingen, were invited for this study and 36 (31%) agreed to participate (8 female GPs, 14 working in group practices with 24 partners). During the study period of 4 months (November 2000 February 2001), active participation was encouraged by regular telephone monitoring of the practices. To maximise generalizability and to reflect daily practice, we choose rather open inclusion criteria for a consecutive sample: all patients presenting to the participating practices in whom the GPs suspected UTI based on symptoms were to be recruited prospectively, including those with risk factors, comorbidity or recent antibiotic treatment. Only patients with an obvious other diagnosis explaining their symptoms (i.e. vaginitis) should be excluded. Patients were to be managed as usual according to the GPs judgement. The use of dipsticks as well as prescription of empirical treatment was at GPs discretion. Each patients age, sex, current symptoms and risk or complicating factors as well as results of dipstick tests and diagnostic procedures (if performed) and treatment were documented on a short, structured form identified by a patient code number. The reference standard for this study was a conventional urine culture. In addition to their usual proceeding and for the purpose of this study, GPs were required to order a culture for all patients (regardless of

dipstick results) before beginning treatment, but to wait for the culture results only if this would have been their usual policy. According to current recommendations, GPs were to sample freshly voided urine only, but midstream sampling was not required.1,4,11 Urine samples were stored in sterile containers supplied by the laboratory and kept refrigerated until processing the same day at the laboratory. A urine culture was performed and antibiotic susceptibility tested in case of bacterial growth, using internationally recommended standard procedures.12,13 All cultures and susceptibility tests were performed by three trained microbiology technicians in the same specialised laboratory (Medical Partnership Wagner Stibbe Kast Bispink & Partners). Technicians were informed of any particular question or information the GPs had noted on the order form (usually nothing). However, they had no access to the documentation form, dipstick results, or other clinical information. Susceptibility testing was performed in samples with more than 102 CFU/ml and less than 3 pathogens, though international literature now suggests that 102 CFU/ml are consistent with UTI.14,15 Therefore, we did not restrict our analysis to traditionally defined high count UTI (105 CFU/ml), but also included cultures yielding low count bacteriuria (102 CFU/ml) or mixed growth in our definition of UTI. Culture results were labelled with the patients code numbers and communicated to the department of general practice. Participating GPs were informed of culture results assigned to patients names. Documentation forms were collected by the laboratorys transport service together with the urine samples for the GPs convenience. They were then sent to the department of general practice without any further processing and without disclosing patients identity. There, all data were entered into SAS, Version 8;16 patients documentation and laboratory results were linked with the patient code number. Descriptive statistics, 2 2 contingency tables and logistic regression models [odds ratios (OR), 95% confidence intervals (CI)] were calculated in SAS. Age (younger than 50, 5074,

Management of urinary tract infections

73

older than 74), dysuria, urgency/ frequency, flank pain, fever, relapse (within 2 weeks of last episode) or recurrent UTI (more than 2 weeks since last episode), presence of any additional risk factors, as well as leucocytes and nitrite on dipsticks were used as independent variables to compare GPs diagnostic decision with a prediction of culture-confirmed UTI. In order to assess the adequacy of different treatment approaches, dummy variables were used. The treatment prescribed by the study GPs was assessed using a dummy variable indicating resistance to the antibiotic chosen. Generally, not treating culture-confirmed UTI with antibiotics was considered inadequate, as well as treatment with an antibiotic to which the pathogen was resistant (cultures with intermediate susceptibility to fluoroquinolones were considered susceptible). Antibiotic treatment for patients with sterile urine was also considered inadequate. Not treating patients with sterile urine and treating patients with UTI with an antibiotic to which the pathogen is susceptible is considered adequate. If information was missing, i.e. either dipsticks or susceptibility testing had not been performed or the antibiotic prescribed had not been specified by the GP, adequacy was defined as unknown. Treatment duration was not considered when assessing adequacy, as no precise recommendations are available for elderly patients.

TABLE 2

Characteristics of adult female patients with suspected UTI ( n = 445) 53 (3371) 45% 33% 22% 12% 23% 89% 55% 52% 17% 14% 12% 27% 9% 7% 3% 3% 3% 4% 1% 12% Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012

Age (median, interquartile range) 50 years 5074 years 75 years Relapse within 14 days since last episode Reinfection 14 days since last episode Symptoms Urgency/frequency Dysuria Kidney/flank pain Suprapubic pain Others Additional risk factors Diabetes Antibiotics in the last 2 weeks Indwelling catheter History of urinary surgery Neurologic voiding dysfunction Renal failure Pregnancy Others

TABLE 3

GPs management of women with suspected UTI ( n 445)

Results
Of the 118 invited GPs, 36 (31%) participated in the study (8 women, 14 working in group practices with 24 partners). 585 patients of both sexes were recruited within 4 months. Only the subgroup of adult women (76%, n = 445) is considered here. Results of male patients have been analysed and published separately,10 as well as a detailed analysis of factors predicting antibiotic resistance. Symptoms were documented on the study sheet in 89% of the women; median duration of symptoms was 3 days. Patients characteristics, symptoms and risk factors are presented in Table 2. GPs diagnostic procedures, diagnoses and prescriptions are presented in Table 3. Median antibiotic treatment duration was 5 days, irrespective of patients age and presence of complicating factors. 70% of young patients with uncomplicated UTI were treated for longer than 3 days. On the other hand, 25% of patients of any age with additional risk factors (complicated UTI) received antibiotics for 3 days or less. One fifth (19.3%) of the patients were not prescribed antibiotics though their GPs had diagnosed UTI. These patients were less likely to have dipsticks positive for nitrite (OR 0.38, 95% CI 0.190.76) and leucocytes (OR 0.48; 95% CI 0.250.91) than those who had been prescribed antibiotics. Patients who had flank pain, but in whom GPs had diagnosed UTI rather than pyelonephritis were less likely to receive antibiotics than other patients with

Diagnostic procedures Dipstick test Sediment microscopy Physical examination Ultrasound of urinary tract Referral Control consultation recommended GPs diagnosis UTI Urethral syndrome (Suspected) pyelonephritis Others Healthy/no diagnosis GPs prescriptions None Antibiotics Cotrimoxazole Fluoroquinolones Nitrofurantoin Trimethoprim Cefixim/cefuroxime Others Spasmolytics Herbal drug Other/not specified Duration of antibiotic treatment (median/days)a Uncomplicated, age 50 Uncomplicated, age 5074 Uncomplicated, age 74 Complicated (any age)
a

92% 44% 38% 15% 3% 43% 64% 16% 5% 7% 8% 36% 56% 46% 33% 2% 13% 2% 4% 2% 9% 3% 5 5 5 5

Patients with suspected pyelonephritis excluded.

UTI (OR 2.94; 95% CI 1.456.25). Treated and untreated patients with GP-diagnosed UTI did not differ with regard to other symptoms, age, recurrent UTI or risk factors.

74

Family Practicean international journal

Urine culture results were available for 430 patients (97%); results are shown in Table 4. Bacterial species were identified in the 282 patients with 103 cfu/ml or more and no more than two pathogens: 67.7% were

infected with Escherichia coli, 10.3% with enterococci, 9.6% with Proteus spp., 7.4% with Streptococcus agalactiae, several others were found in 14 urine samples, respectively. Prediction rule for UTI GPs apparently diagnosed UTI based on both their clinical assessment and dipstick tests, which had been performed in 91.6% of all patients. Detailed documentation of a stepwise approach was not available for feasibility reasons. Table 5 illustrates the diagnostic value of dipsticks with regard to culture results. Table 6 compares predictors for GPs diagnosis of infection (either UTI or pyelonephritis) and cultureconfirmed bacteriuria. Fever (OR 6.44, 95% CI 1.1536.03) and flank pain (OR 22.06, 95% CI 7.1368.32) were the only significant predictors for a diagnosis of pyelonephritis; however, one in 5 patients with either fever or flank pain was diagnosed as not having pyelonephrits (data not shown). Using cultureconfirmed UTI (102 CFU/ml) as a gold standard, GPs diagnosis had a sensitivity of 70.8% and a specificity of 38.8%, the positive predictive value (PPV) was 79.7%, the negative predictive value (NPV) 28.2%. Adequacy of different treatment approaches Figure 1 illustrates the appropriateness of treatment decisions for all patients in this study in a flow chart. 60.2% of the 332 patients with culture-confirmed UTI received antibiotics, and 42.9% of those with sterile urine (n = 98). 39.8% of all women with UTI received no treatment. In 24% of the women with UTI who were prescribed antibiotics, pathogens were resistant to the treatment and this occurred significantly more often in patients with complicating factors (52.1 versus 22.5%; OR 3.74; 95% CI 1.847.62).

TABLE 4

Results of urine cultures (available for n 430 patients) n % 22.8% 53.5% 8.1% 4.0% 11.6%

Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012

Sterile urine UTI 102 cfu/ml, single growth UTI UTI 102 102 cfu/ml, 2 pathogens cfu/ml, 3 pathogens

98 230 35 17 50

Low count UTI 102 cfu/ml

Resistance levels in all cultures with more than 102 cfu/ml and less than 3 pathogens (n = 298a) Antibiotic Amoxicillin Co-amoxiclav Cefazolin Cefixime Gentamicin Ofloxacinb Ciprofloxacinb Co-trimoxazole Trimethoprim Nitrofurantoin
a

% resistant pathogens 36.9 28.5 38.6 16.1 21.5 6.7 6.4 31.5 33.6 13.1

Refers to 265 patients; in 33 patients two pathogens were tested for susceptibility. b 17% of all pathogens had intermediate susceptibility to ofloxacin and ciprofloxacin. As fluoroquinolones are likely to be clinically effective for UTI in these patients, they were considered susceptible.

TABLE 5

Dipstick test results and diagnostic value with regard to any UTI 102 CFU/ml, including mixed growth Totals Nitrite positive 121 10 131 39.3 88.4 92.4 28.9 3.39 92.4 Nitrite negative 187 76 263 Leucocytes positive 221 46 267 71.8 46.5 82.8 31.5 1.34 82.7 Leucocytes negative 87 40 127 Both positive At least one positive 235 46 281 76.3 46.5 83.6 35.4 1.43 83.6

UTI Sterile urine Totals Sensitivity (%) Specificity (%) PPVb (%)

308 86 394a

107 10 117 34.7 88.4 91.5 27.4 2.99 91.5

NPVc(%) Pos. likelihood ratio Post-test probability (%)

a Dipstick and cultures had been performed in all patients. Urine culture was not available for 15 patients; dipsticks had not been performed in 36 patients. b Positive predictive value. c Negative predictive value.

Management of urinary tract infections

75

A simulation of different therapeutic strategies is presented in Table 7; and the adequacy of empiric treatment with several antibiotics of either all patients with suspected UTI or patients with a positive nitrite test is compared with the treatment prescribed by the GPs in our study.

Discussion
Our cross-sectional study combines results of systematic urine cultures with an observational survey of GPs
TABLE 6 Comparison of factors predicting GPs diagnosis and culture results (multivariate logistic regression; significant predictors in bold type) Independent variable UTI or pyelonephritis diagnosed by GP OR (95% CI) 2.77 (1.395.55) 3.53 (2.086.00) 1.04 (0.561.93) 7.59 (4.2913.42) 6.39 (2.9014.09) 2.28 (1.343.87) 1.67 (0.743.74) 18.12 (1.72190.85) 1.66 (0.713.89) 2.30 (1.164.56) 1.36 (0.951.96) Culture-confirmed UTI 102 cfu/ml OR (95% CI) 3.41 (1.876.21) 1.57 (0.972.54) 1.16 (0.671.99) 1.97 (1.223.17) 0.44 (0.230.79) 0.87 (0.551.38) 0.88 (0.46 1.70) 2.37 (0.589.68) 0.61 (0.301.25) 1.51 (0.862.67) 1.48 (1.082.02)

Nitrite on dipstick Leucocytes on dipstick Complicating factors Dysuria Flank pain Urgency/frequency Suprapubic pain Fever Relapse Reinfection Older age (50, 5074, 74)

management in of women with suspected UTI, allowing us to assess the appropriateness of GPs prescriptions. GPs diagnosed UTI on a basis of both dipsticks and a clinical impression of symptoms, resulting in relatively low positive and negative predictive values when compared to culture results. Dipsticks can confirm UTI, but are not very helpful in identifying patients with sterile urine. Patients with uncomplicated UTI were often treated for longer than recommended, and secondchoice antibiotics were prescribed to a large proportion of all patients. Both under-treatment of patients with UTI and over-treatment of patients with sterile urine occurred frequently. Similar non-adherence to UTI guidelines has been described in other countries.17 GPs participating in our survey were not routinely involved in research, they practice in both rural and urban settings. Though there may be selection bias concerning GPs, their patients are not likely to differ from patients in non-participating practices. Typical for general practice studies in countries without practice lists, we do not know the precise catchment rate of our study, but had to rely on participating GPs to include all their eligible patients.18,19 We attempted to ensure active participation though regular telephone monitoring. The number of included patients (of both sexes) corresponds to the practice prevalence reported in other German studies on UTI.20,21 All urine cultures and susceptibility tests were performed in a single laboratory, but dipsticks were assessed by the individual GPs (or their practice staff), who reported their reading. Though GPs reading of urinary dipsticks is known to vary, our approach reflects practice reality better than standardised reading in a study centre (which was not done for feasibility reasons).22,23

Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012

445 patients

249 antibiotics

196 no antibiotics

7 no culture available

200 UTI

42 sterile urine

56 sterile urine

132 UTI

8 no culture available

148 UTI > 10 cfu/ml

20 UTI 10 cfu/ml

32 mixed growth

82 UTI > 10 cfu/ml

30 UTI 10 cfu/ml

20 mixed growth

111 susceptible

9 susceptible

11 antibiogram

33 resistant

15 resistant

9 no antibiogram

4 antibiotic not specified

8 no antibiogram, 3 species

FIGURE 1 Adequacy of all patients management with regard to the presence of culture-confirmed UTI and susceptibility to the individual antibiotic prescribed in each case

76
TABLE 7

Family Practicean international journal

Calculated adequacy of different treatment approaches ( n = 445 patients) Adequately treated patients n (%) Inadequately treated patients n (%) Adequacy cannot be judged, missing information n (%) 47 (10.6) 86 (19.3) 86 (19.3) 86 (19.3) 113 (25.4) 113 (25.4) 113 (25.4)

Treatment option

Treatment as prescribed by GPs Empirical treatment nitrofurantoin Empirical treatment trimethoprim Empirical treatment fluoroquinolone Nitrofurantoin if nitrite positive Trimethoprim if nitrite positive Fluoroquinolone if nitrite positive

176 (39.5) 222 (49.9) 168 (37.8) 241 (54.2) 165 (37.1) 141 (31.7) 175 (39.3)

222 (49.9) 137 (30.8) 191 (42.9) 118 (26.5) 167 (37.5) 191 (42.9) 157 (35.3)

GPs used dipsticks in most patients, and their results significantly contributed to the diagnostic decision, with GPs relying more on leucocytes than the nitrite test. Additionally, several symptoms and earlier UTI episodes were taken into consideration. However, this mixed approach resulted in a poorer positive predictive value than dipsticks alone, and the negative predictive values were similarly low. Culture-confirmed UTI was significantly predicted by a positive nitrite test, and to a lesser degree by the symptom dysuria and older age. Leucocytes on dipstick are not helpful. Due to low sensitivity and negative predictive values, dipsticks can confirm a diagnosis in symptomatic patients, but are not very reliable in excluding UTI.24 It is somewhat surprising that neither reported fever nor flank pain were significantly associated with bacteriuria. In line with current recommendations, GPs had associated both symptoms with pyelonephritis, but decided against this diagnosis in a few patients. Possibly, GPs were aware that co-existing common conditions like viral infections or low back pain could blur the picture. Only about one third of all patients in our study were adequately treated, and in half of all patients either the decision to prescribe an antibiotic or the antibiotic chosen was inappropriate. Our cross-sectional survey did not allow us to determine clinical effectiveness of treatment or evolution of symptoms in untreated patients with UTI or patients receiving antibiotics to which their pathogens were resistant. As sometimes clinical results have shown to be better than in vitro susceptibility would predict,7,25 it is possible that some of our inadequately treated patients had a favourable

clinical outcome. Additionally to the apparent difficulty in targeting antibiotic prescriptions, treatment duration was shorter than recommended1,2,4,14 in one quarter of patients with additional risk factors, and longer than necessary in most young women with uncomplicated UTI.26 In our comparison of different treatment approaches to the patients in this study, empirical treatment with either nitrofurantoin or fluoroquinolones proved superior, and empirical treatment with trimethoprim was equivalent to the individualized approach preferred by the study GPs. Treatment of all symptomatic patients has been shown to be the most cost-effective option,27 but costs of rising resistance levels were not considered.28 However, this would result in inadequate treatment for one quarter to one third of all patients, including unnecessary antibiotic exposure of patients who did not have UTI (22% in our study). This should be a reason for concern in presence of high resistance levels for all common antibiotics.29,30,31 In our study, pathogens proved resistant to the antibiotic the patient had received in almost a quarter of all cases. GPs seemed to be aware of resistance problems and choose to prescribe fluoroquinolones in one third of cases. Though still highly effective at the time of our study, the high prevalence of intermediate susceptibility indicates a dropping susceptibility level: in Germany, the level of resistance against ciprofloxacin in E.coli increased from 7.7% in 1998 to 14.5 % in 200132 and similar tendencies have been observed in other countries.7,33 One possibility to reduce unnecessary antibiotic use is to treat only patients with a positive nitrite test. This would have reduced the number of inadequately treated patients in our study (as dipsticks were performed for only 90% of the patients, however, adequacy could not be judged in many cases). Due to the low negative predictive value of dipsticks, cultures should then be considered for symptomatic patients with a negative nitrite test. Ordering cultures for all patients and delaying antibiotic treatment until results are available could limit over-treatment.8 However this would increase direct costs, and often cultures would have no impact on management. In many patients without additional risk factors, UTI seems to be a self-limiting condition: A survey in a British practice has shown that few women with urinary symptoms actually consult a doctor;34 one trial in Belgium has shown that half of the patients were free of symptoms after 3 days of placebo.35 Symptomatic treatment of uncomplicated UTI may be an option which merits further research.

Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012

Acknowledgements
We thank all participating GPs for their efforts and co-operation. We are grateful to Dr Thomas Fischer, Gttingen, for his comments on the manuscript.

Management of urinary tract infections

77

Declaration
Funding: the medical laboratory partnership Wagner, Stibbe, Kast, Bispink & Partners sponsored the urine cultures and susceptibility testing. Ethical approval: the local ethics review board had no objections to the study. Conflicts of interest: none.

16

17

18

19

20

References
1 2

10

11

12

13

14

15

Hummers-Pradier E, Kochen MM. Dysuria. DEGAM-guideline No. 1. [in German] Z Allg Med 2000; 76: 3548. Timmermans AE, Baselier PJAM, Winkens RAG et al. NHG standard: urinary tract infection. [in Dutch] Huisarts Wet 1999; 42: 613622. Flottorp S, Oxman AD, Cooper JG, Hjortdahl P, Sandberg S, Vorland LH. Guidelines for diagnosis and treatment of acute urinary tract problems in women [in Norwegian]. Tidsskr Nor Laegeforen 2000; 120: 17481753. Prodigy guidance. Urinary tract infections (lower)women, 2002. http://www.prodigy.nhs.uk/guidance.asp?gt=uti%20(lower) %20-%20women (accessed on June 14, 2004.) University of Michigan Health System. UMHS urinary tract infection guideline. Ann Arbor (MI): University of Michigan Health System; 1999. http://www.guideline.gov/summary/ summary.aspx?ss=15&doc_id=2284&nbr=1510&string=urinary %20AND%20tract%20AND%20infection%20AND%20women (accessed on June 14, 2004.) Institute for Clinical Systems Improvement (ICSI). Uncomplicated urinary tract infection in women. Bloomington (MN): Institute for Clinical Systems Improvement (ICSI); 2002. http://www. guideline.gov/summary/summary.aspx?ss=15&doc_id=3675& nbr=2901&string=urinary%20AND%20tract%20AND%20 infection%20AND%20women (accessed on June 14, 2004.) Gupta K, Hooton TM, Stamm WE. Increasing antimicrobial resistance and the management of uncomplicated communityacquired urinary tract infections. Ann Intern Med 2001; 135: 4150. Davey P, Steinke D, MacDonald T, Phillips G, Sullivan F. Not so simple cystitis: how should prescribers be supported to make informed decisions about the increasing prevalence of infections caused by drug-resistant bacteria? Br J Gen Pract 2000; 50: 143146. Stamm WE. An epidemic of urinary tract infections? N Engl J Med 2001; 345: 10551057. Hummers-Pradier E, Ohse AM, Koch M, Heizmann WR, Kochen MM. Urinary tract infections in men. Int J Clin Pharmacol Ther 2004; 7: 360366. Hummers-Pradier E, Kochen MM. Urinary tract infections in adult general practice patients. Br J Gen Pract 2002; 52: 752761. Clarridge JE, Pezzlo MT, Vosti KL, Weissfeld AS (co-ordinating ed.), Cumitech 2A. Laboratory diagnosis of urinary tract infections. Washington DC: American Society of Microbiology; 1987. Anonymus, Medical Microbiology and Immunology: diagnostic procedures [in German]. 3rd Edition. Berlin: DIN, Deutsches Institut fr Normung e.V (German Institute for Standardisation); 2000. Stamm WE, Hooton TM. Management of urinary tract infection in adults. N Engl J Med 1993; 329: 13281334. Kunin CM, White LV, Hua TH. A reassessment of the importance of low-count bacteriuria in young women with acute urinary symptoms. Ann Intern Med 1993; 119: 454460.

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

SAS Institute Inc. SAS/STAT. Users guide, Version 8, vol. 3. Cary, NC: SAS Institute Inc.; 1999. Kahan E, Kahan NR, Chinitz DP. Urinary tract infection in womenphysicians preferences for treatment and adherence to guidelines: a national drug utilization study in a managed care setting. Eur J Clin Pharmacol 2003; 59: 663668. Bell-Syer SE, Moffett JA. Recruiting patients to randomized trials in primary care: principles and case study. Fam Pract 2000; 17: 187191. Wilson S, Delaney BC, Roalfe A et al. Randomised controlled trials in primary care: case study. Br Med J 2000; 321: 2427. Gulich M, Bux C, Zeitler HP. The DEGAM guidelines Dysuria by the German Society of General Practice and Family Medicine (DEGAM)possible consequences of the implementation in general practice [in German]. Z Arztl Fortbild Qualitatssich 2001; 95: 141145. Hummers-Pradier E, Beyer M, Gerlach F, Kochen MM. A feasibility study as part of the development of guidelines for general practice in Germany. Eur J Gen Pract 2001; 7: A6. Winkens RA, Leffers P, Trienekens TA, Stobberingh EE. The validity of urine examination for urinary tract infections in daily practice. Fam Pract 1995; 12: 290293. Christiaens TCM, Meyere MD, Derese A. Disappointing specificity of the leucocyte-esterase test for the diagnosis of urinary tract infection in general practice. Eur J Gen Pract 1998; 4: 144147. Verest LF, van Esch WM, van Ree JW, Stobberingh EE. Management of acute uncomplicated urinary tract infections in general practice in the south of The Netherlands. Br J Gen Pract 2000; 50: 309310. Raz R, Chazan B, Kennes Y et al. Empiric use of trimethoprimsulfamethoxazole (TMP-SMX) in the treatment of women with uncomplicated urinary tract infections, in a geographical area with a high prevalence of TMP-SMX-resistant uropathogens. Clin Infect Dis 2002; 34: 11651169. Kahan NR, Chinitz DP, Kahan E. Longer than recommended empiric antibiotic treatment of urinary tract infection in women: an avoidable waste of money. J Clin Ther 2004; 29: 59. Fenwick EA, Briggs AH, Hawke CI. Management of urinary tract infection in general practice: a cost-effectiveness analysis. Br J Gen Pract 2000; 50: 635639. Coast J, Smith RD, Millar MR. An economic perspective on policy to reduce antimicrobial resistance. Soc Sci Med 1998; 46: 2938. Christiaens TC, Digranes A, Baerheim A. The relation between sale of antimicrobial drugs and antibiotic resistance in uropathogens in general practice. Scand J Prim Health Care 2002; 20: 4549. Magee JT, Prichard EL, Fitzgerald KA, Dunstan FDJ, Howard AJ. Antimicrobiotic prescribing and antibiotic resistance in community practice: retrospective study, 19968. Br Med J 1999; 319: 12391240. McIsaac WJ, Low DE, Biringer A, Pimlott N, Evans M, Glazier R. The impact of empirical management of acute cystitis on unnecessary antibiotic use. Arch Intern Med 2002; 162: 600605. Kresken M, Hafner D, Schmitz FJ, Wichelhaus TA. Resistance situation of clinically important infectious pathogens in Germany and central Europe [in German]. Paul-Ehrlich-Society for Chemotherapy. Bonn: Antiinfectives Intelligence; 2003. Schaeffer AJ. The expanding role of fluoroquinolones. Am J Med 2002; 113 Suppl 1A: 45S54S. Jolleys JV. The reported prevalence of urinary symptoms in women in one rural general practice. Br J Gen Pract 1990; 40: 335337. Christiaens TC, De Meyere M, Verschraegen G, Peersman W, Heytens S, De Maeseneer JM. Randomised controlled trial of nitrofurantoin versus placebo in the treatment of uncomplicated urinary tract infection in adult women. Br J Gen Pract 2002; 52: 729734.

Downloaded from http://fampra.oxfordjournals.org/ at Grigore T.Popa Medicine and Pharmacy University from Iasi on November 21, 2012