Screening for urinary tract infection in women with hyperemesis gravidarum

The aim of this study was to evaluate urine microscopy, dipstick analysis and urinary symptoms in screening for urinary tract infection (UTI) in hyperemesis gravidarum (HG).

A prospective cross-sectional study was performed on women at first hospitalization for HG. A clean-catch mid-stream urine sample from each recruit was sent for microscopy (for bacteria, leucocytes and erythrocytes), dipstick analysis (for leukocyte esterase, nitrites, protein and hemoglobin) and microbiological culture. The presence of current urinary symptoms was elicited by questionnaire. UTI is defined as at least 10(5) colony-forming units/mL of a single uropathogen on culture. Screening test parameters were analyzed against UTI.

UTI was diagnosed in 15/292 subjects (5.1%). Receiver-operator characteristic curve analysis of microscopic urine leucocytes revealed area under the curve=0.64, 95% confidence interval (CI) 0.5-0.79, P=0.063 and erythrocytes area under the curve=0.53, 95%CI 0.39-0.67, P=0.67 for UTI indicating the limited screening utility of these parameters. Microscopic bacteriuria (likelihood ratio [LR] 1.1, 95%CI 0.7-1.5) and urine dipstick leukocyte esterase (LR 1.4, 95%CI 1.1-1.8), nitrites (LR 2.3, 95%CI 0.3-17.2), protein (LR 1.0, 95%CI 0.7-1.6) and hemoglobin (LR 0.8, 95%CI 0.4-1.5) were not useful screening tests for UTI in HG. Elicited symptoms were also not predictive of UTI.

Urine microscopy, dipstick analysis and urinary symptoms were not useful in screening for UTI in HG. UTI should be established by urine culture in HG before starting antibiotic treatment.