Diagnosis of Coliform Infection in Acutely Dysuric Women

• Published in: The New England journal of medicine Vol. 307; no. 8; pp. 463 - 468
• Main Authors: Stamm, Walter E, Counts, George W, Running, Katherine R, Fihn, Stephen, Turck, Marvin, Holmes, King K
• Format: Journal Article
• Language: English
• Published: United States Massachusetts Medical Society 19.08.1982
• Subjects: Acute Disease; Bacteria; Bladder; Chlamydia; Chlamydia trachomatis; Coliforms; Diagnosis; Diagnostic tests; Enterobacteriaceae - isolation & purification; Enterobacteriaceae Infections - diagnosis; Enterobacteriaceae Infections - microbiology; Female; Humans; Intubation; Laboratories; Medical diagnosis; Organisms; Patients; Potash; Potassium; Public health; Specimen Handling; Urethra; Urinary bladder; Urinary tract; Urinary tract infections; Urinary Tract Infections - diagnosis; Urinary Tract Infections - microbiology; Urination Disorders - microbiology; Urine; Urine - microbiology; Urogenital system; Vagina
• ISSN: 0028-4793; 1533-4406
• DOI: 10.1056/NEJM198208193070802

We reevaluated conventional criteria for diagnosing coliform infection of the lower urinary tract in symptomatic women by obtaining cultures of the urethra, vagina, midstream urine, and bladder urine. The traditional diagnostic criterion, ≥10 5 bacteria per milliliter of midstream urine, identified only 51 per cent of women whose bladder urine contained coliforms. We found the best diagnostic criterion to be ≥10 2 bacteria per milliliter (sensitivity, 0.95; specificity, 0.85). Although isolation of <10 5 coliforms per milliliter of midstream urine has had a low predictive value in previous studies of asymptomatic women, the predictive value of the criterion of ≥10 2 per milliliter was high (0.88) among symptomatic women in whom the prevalence of coliform infection exceeded 50 per cent. In view of these findings, clinicians and microbiologists should alter their approach to the diagnosis and treatment of women with acute symptomatic coliform infection of the lower urinary tract. (N Engl J Med. 1982; 307:463–8.) SINCE the benchmark studies of Kass 1 2 3 more than 20 years ago, a quantitative culture of midstream urine containing ≥10 5 bacteria per milliliter has been the criterion employed by nearly all physicians to establish the diagnosis of urinary-tract infection. These studies established that 95 per cent of women with acute pyelonephritis have bacteriuria of ≥10 5 bacteria per milliliter, and that this number of organisms, when repeatedly isolated from the midstream urine specimens of asymptomatic women, reliably distinguishes contamination from true infection. 1 2 3 However, the majority of women evaluated by physicians for possible urinary-tract infection have neither acute pyelonephritis nor asymptomatic bacteriuria but . . .