Does Asymptomatic Bacteriuria Predict Mortality and Does Antimicrobial Treatment Reduce Mortality in Elderly Ambulatory Women?

15 May 1994 | Volume 120 Issue 10 | Pages 827-833

Objective: To determine whether asymptomatic bacteriuria in elderly ambulatory women is a marker of increased mortality and, if so, whether it is because of an association with other determinants of mortality or because asymptomatic bacteriuria is itself an independent cause, the removal of which might improve longevity.

Design: A cohort study and a controlled clinical trial of the effect of antimicrobial treatment.

Setting: A geriatric center and 21 continuing care retirement communities.

Participants: Women without urinary tract catheters.

Measurements: Urine cultures every 6 months (the same organism at 10⁵ colony-forming units or more per mL on two midstream urine specimens defined asymptomatic bacteriuria), comorbidity, and mortality.

Results: In the observational study, infected residents (n = 318) were older, and sicker, and had higher mortality (18.7 per 100 000 resident-days) than uninfected residents (n = 1173; 10.1 per 100 000 resident-days). However, in a multivariate Cox analysis, infection was not related to mortality (relative risk, 1.1; P > 0.2), whereas age at entry and self-rated health (score 1 [excellent] to 4 [bad or poor]) were strong predictors. In the clinical trial, mortality in 166 treated residents (13.8 per 100 000 resident-days) was comparable to that of 192 untreated residents (15.1 per 100 000 resident-days); the relative rate was 0.92 (95% CI, 0.57 to 1.47). The cure rates among treated and untreated residents were 82.9% and 15.6%, respectively.

Conclusion: Urinary tract infection was not an independent risk factor for mortality, and its treatment did not lower the mortality rate. Screening and treatment of asymptomatic bacteriuria in ambulatory elderly women to decrease mortality do not appear warranted.

Author and Article Information

From the Medical College of Pennsylvania, Philadelphia Department of Veterans Affairs Medical Center, and the University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania.
Requests for Reprints: Elias Abrutyn, MD, Infectious Diseases, Department of Veterans Affairs Medical Center (111 ID), 38th and Woodland Avenues, Philadelphia, PA 19104-4594.
Acknowledgments: The authors thank Karen Cassel, Anne Dreibelbis, Caroline J. Fediw, Janet Lennett, Arthur J. Osborne, and Audrey Stein for their assistance and the residents and staff of the communities without whom this study would not have been possible: The Beaumont, Cadbury, Cathedral Village, Cokesbury Village, Dunwoody Village, Foulkeways at Gwynedd, Granite Farms Estates, Gwynedd Estates, Lima Estates, Logan Square East, Marins Run, Medford Leas, Normandy Farm Estates, Pennswood Village, Philadelphia Geriatric Center, Pine Run, Protestant Home of Philadelphia, Rydal Park, Spring House Estates, Springfield, Stapeley in Germantown, and Waverly Heights.
Grant Support: In part by National Institutes of Health TNH award AG03934.