674
Views of General Practice A PROSPECTIVE STUDY OF URINARY-TRACT INFECTIONS IN A DUTCH GENERAL
PRACTICE R. GAYMANS M. I. HAVERKORN†
H. A. VALKENBURG* W. R. O. GOSLINGS
would not influence the prevalence figures. Women with Significant A.B. were asked to come back at bimonthly intervals for re-examination of urine. During the follow-up period onh one urine culture was determined at each examination. The same procedure was applied to women with a symptoms: urinary-tract infection. As the reliability of the first culture for all species of gram-negative organisms was 90% (see later’ this was considered sufficient for the purpose of this study and dec reased the work-load both for the general practitioner and the women concerned. The group of women with significant u was matched in age by a control group not having A.B. at
screening.
Department of Microbial Diseases, University Hospital, Leiden, The Netherlands Dutch general practice the adult Summary female population was screened for asymptomatic bacteriuria (A.B.) by repeated urine culture after vulval cleansing. The prevalence of significant A.B. was 4·7% and increased with age. Women with significant A.B. were followed up for one year. All symptomatic urinary-tract infections were recorded during the same period (incidence 59 per 1000 population). Women with significant A.B. at screening were divided into three groups according to the pattern of the follow-up results: transitory A.B., symptomatic A.B., and persistent A.B. The last group differed from the penultimate group with regard to the site of their urinary-tract involvement, In
a
symptomatic cases having predominantly upper-urinarytract involvement and persistent cases lower-urinaryinfection. In the matched control group the acquisition-rate of both symptomatic and asymptomatic bacteriuria was over 12%, a figure similar to the percentage of women present in the practice population during one year with transient, symptomatic, and persistent A.B. Screening for A.B. in the general non-pregnant female population is not advocated at present. Screening and treatment of existing A.B. should be carried out in pregnant women who run an increased risk. tract
INTRODUCTION
THE natural history of asymptomatic bacteriuria occurring in women above the age of fourteen was studied in a Dutch general practice for one year. The study was primarily designed to evaluate the persistent and transitory presence of significant numbers of microorganisms in the urine of women who were found to have asymptomatic bacteriuria during a screening survey and were followed up without treatment for the subsequent year. The use of periodic screening for asymptomatic bacteriuria for the prevention of more serious kidney disease such as chronic pyelonephritis was assessed. SUBJECTS
AND METHODS
Study Population
population above the age of fourteen (1862 persons) belonging to the general practice of R.G. and his father, F.H.C. Gaymans, was included in the study. This practice is situated in the south of the Netherlands in a predominantly rural area. Some of the population work in nearby industries. Of the 1862 women 1758 participated (94.4%). Pregnant women were not excluded from the prevalence study, The whole female
which covered
one
year. It
was
Urinalysis A questionnaire dealing with demographic data and a medical history pertinent to the urinary tract was completed by all before their visit to the survey centre. The women were instructed how to produce a midstream urine sample after two vulval cleansings. The urine was immediately cultured in duplicate by R.G. according to the method of Leigh and Wit liams’ on McConkey agar. If, after overnight incubation at 37°C, the culture yielded > 10 microorganisms/ml of one spe cies or of mixed growth or between 104 and 103 bacteria/ml of one type, the same urine was cultured again the next day, after overnight refrigeration at 4°C, by a serial dilution method on McConkey agar. If the colony count was > 10’ bacteria/mlof either pure or mixed growth with a maximum of two species, the strain(s) were typed and the respondent was asked to come back within two weeks for another urine culture. In the sub-
women
jects with gram-negative microorganisms in which only one species was observed at both examinations within two weeks,
repoducibility of the results of the first culture was 90c (for Escherichia coli even 93%), as judged by the results of the second urine culture. Microscopical urinalysis was done on the non-stained wet urine sediment (magnification 250 times) and on a gram-stained dried droplet of non-centrifuged urine (magnification 1000 times). the
Serum Analysis
and creatinine was determined in subjects with and the matched control group at an early stage of the follow-up period. In some subjects with significant A.B., the direct agglutination titre was assayed against the homologous E. coli strain. Serum
urea
significant
A.B.
Urological Examination Patients with insufficient response to treatment for theo clinical symptoms and persons with persistent A.B. were urologically examined at the end of the follow-up period by means of an intravenous pyelogram (i.v.p.) and bladder inspection,
Definition ofA.B. Significant A.B. was defined as the presence of two separate urine cultures yielding .10 microorganisms in pure culture 0: maximum of two species in the same subject. Transitory was considered to be present when during the follow-up perror of one year the subject became negative during at least one bimonthly follow-up period without treatment. Persistents defined as the continuous presence of ≥105 microcr ganisms/ml in the same woman during the one-year follow-ut Symptomatic A.B. occurred when a subject with A.B. developer symptoms of urinary-tract infection during the follow was
period. RESULTS
assumed that seasonal factors
Prevalence *Present address:
Department of Epidemiology, Erasmus University, Rotterdam. †Present address: Department of Clinical Microbiology and Antibiotic Treatment, Erasmus University, Rotterdam.
of Significant A.B.
Of the 1758women examined, 115 had a positive ,t:: ( 105 microorganisms/ml) at the first examinat!o
ture
675 the basis of the statistical analysis of the probability of the presence or absence of the missing E. coli strain. This procedure was applied in 19 cases with one missing follow-up examination and 5 cases with three known follow-up examinations. In 5 of the 58cases antimicrobial therapy was given for reasons not related to a urinary-tract infection. These cases were excluded from the analysis of the
6 of these were not re-examined due to migration from the area, refusal to return, or administrative errors. Of the 109 women examined a second time 13 were negative on second culture. In 5 women the strain of the second urine specimen differed from the first culture, but the follow-up results indicated the presence of a double infection. In 8 women symptoms of urinary-tract infection developed within the one or two weeks between the two examinations and the second urine cul-
corrected
yielded the same microorganisms. This reduced the total number of women with significant A.B. to 88 (83 with a single species and 5 with a double infection). 90% of the isolated bacteria were E. coli. The age distribution of women with significant A.B. differed significantly from those not having bacteriuria 0001
Views of General Practice A PROSPECTIVE STUDY OF URINARY-TRACT INFECTIONS IN A DUTCH GENERAL
PRACTICE R. GAYMANS M. I. HAVERKORN†
H. A. VALKENBURG* W. R. O. GOSLINGS
would not influence the prevalence figures. Women with Significant A.B. were asked to come back at bimonthly intervals for re-examination of urine. During the follow-up period onh one urine culture was determined at each examination. The same procedure was applied to women with a symptoms: urinary-tract infection. As the reliability of the first culture for all species of gram-negative organisms was 90% (see later’ this was considered sufficient for the purpose of this study and dec reased the work-load both for the general practitioner and the women concerned. The group of women with significant u was matched in age by a control group not having A.B. at
screening.
Department of Microbial Diseases, University Hospital, Leiden, The Netherlands Dutch general practice the adult Summary female population was screened for asymptomatic bacteriuria (A.B.) by repeated urine culture after vulval cleansing. The prevalence of significant A.B. was 4·7% and increased with age. Women with significant A.B. were followed up for one year. All symptomatic urinary-tract infections were recorded during the same period (incidence 59 per 1000 population). Women with significant A.B. at screening were divided into three groups according to the pattern of the follow-up results: transitory A.B., symptomatic A.B., and persistent A.B. The last group differed from the penultimate group with regard to the site of their urinary-tract involvement, In
a
symptomatic cases having predominantly upper-urinarytract involvement and persistent cases lower-urinaryinfection. In the matched control group the acquisition-rate of both symptomatic and asymptomatic bacteriuria was over 12%, a figure similar to the percentage of women present in the practice population during one year with transient, symptomatic, and persistent A.B. Screening for A.B. in the general non-pregnant female population is not advocated at present. Screening and treatment of existing A.B. should be carried out in pregnant women who run an increased risk. tract
INTRODUCTION
THE natural history of asymptomatic bacteriuria occurring in women above the age of fourteen was studied in a Dutch general practice for one year. The study was primarily designed to evaluate the persistent and transitory presence of significant numbers of microorganisms in the urine of women who were found to have asymptomatic bacteriuria during a screening survey and were followed up without treatment for the subsequent year. The use of periodic screening for asymptomatic bacteriuria for the prevention of more serious kidney disease such as chronic pyelonephritis was assessed. SUBJECTS
AND METHODS
Study Population
population above the age of fourteen (1862 persons) belonging to the general practice of R.G. and his father, F.H.C. Gaymans, was included in the study. This practice is situated in the south of the Netherlands in a predominantly rural area. Some of the population work in nearby industries. Of the 1862 women 1758 participated (94.4%). Pregnant women were not excluded from the prevalence study, The whole female
which covered
one
year. It
was
Urinalysis A questionnaire dealing with demographic data and a medical history pertinent to the urinary tract was completed by all before their visit to the survey centre. The women were instructed how to produce a midstream urine sample after two vulval cleansings. The urine was immediately cultured in duplicate by R.G. according to the method of Leigh and Wit liams’ on McConkey agar. If, after overnight incubation at 37°C, the culture yielded > 10 microorganisms/ml of one spe cies or of mixed growth or between 104 and 103 bacteria/ml of one type, the same urine was cultured again the next day, after overnight refrigeration at 4°C, by a serial dilution method on McConkey agar. If the colony count was > 10’ bacteria/mlof either pure or mixed growth with a maximum of two species, the strain(s) were typed and the respondent was asked to come back within two weeks for another urine culture. In the sub-
women
jects with gram-negative microorganisms in which only one species was observed at both examinations within two weeks,
repoducibility of the results of the first culture was 90c (for Escherichia coli even 93%), as judged by the results of the second urine culture. Microscopical urinalysis was done on the non-stained wet urine sediment (magnification 250 times) and on a gram-stained dried droplet of non-centrifuged urine (magnification 1000 times). the
Serum Analysis
and creatinine was determined in subjects with and the matched control group at an early stage of the follow-up period. In some subjects with significant A.B., the direct agglutination titre was assayed against the homologous E. coli strain. Serum
urea
significant
A.B.
Urological Examination Patients with insufficient response to treatment for theo clinical symptoms and persons with persistent A.B. were urologically examined at the end of the follow-up period by means of an intravenous pyelogram (i.v.p.) and bladder inspection,
Definition ofA.B. Significant A.B. was defined as the presence of two separate urine cultures yielding .10 microorganisms in pure culture 0: maximum of two species in the same subject. Transitory was considered to be present when during the follow-up perror of one year the subject became negative during at least one bimonthly follow-up period without treatment. Persistents defined as the continuous presence of ≥105 microcr ganisms/ml in the same woman during the one-year follow-ut Symptomatic A.B. occurred when a subject with A.B. developer symptoms of urinary-tract infection during the follow was
period. RESULTS
assumed that seasonal factors
Prevalence *Present address:
Department of Epidemiology, Erasmus University, Rotterdam. †Present address: Department of Clinical Microbiology and Antibiotic Treatment, Erasmus University, Rotterdam.
of Significant A.B.
Of the 1758women examined, 115 had a positive ,t:: ( 105 microorganisms/ml) at the first examinat!o
ture
675 the basis of the statistical analysis of the probability of the presence or absence of the missing E. coli strain. This procedure was applied in 19 cases with one missing follow-up examination and 5 cases with three known follow-up examinations. In 5 of the 58cases antimicrobial therapy was given for reasons not related to a urinary-tract infection. These cases were excluded from the analysis of the
6 of these were not re-examined due to migration from the area, refusal to return, or administrative errors. Of the 109 women examined a second time 13 were negative on second culture. In 5 women the strain of the second urine specimen differed from the first culture, but the follow-up results indicated the presence of a double infection. In 8 women symptoms of urinary-tract infection developed within the one or two weeks between the two examinations and the second urine cul-
corrected
yielded the same microorganisms. This reduced the total number of women with significant A.B. to 88 (83 with a single species and 5 with a double infection). 90% of the isolated bacteria were E. coli. The age distribution of women with significant A.B. differed significantly from those not having bacteriuria 0001
