The diagnostic accuracy of the point-of-care urine dipstick test in detecting urinary tract infection among symptomatic patients in Nairobi County, Kenya

Abstract

AbstractBackgroundUrinary tract infections (UTIs) are among the most frequently diagnosed bacterial infections and constitute a large proportion of workload in clinical microbiology laboratories. Urine culture is the confirmatory test for UTI. However, most primary care settings routinely use the more rapid, less labour-intensive dipstick. This study assesses the characteristics of a urine dipstick test in predicting a positive urine culture and how best it can be used in resource constrained settings despite its limitations.MethodA cross-sectional study was conducted at two level-V health facilities in Nairobi County, Kenya. Adults and children presenting with clinical symptoms of UTI were enrolled after obtaining written informed consent. Midstream urine samples were collected. Urinary dipstick was used to identify Nitrites (NIT) and leucocyte esterase (LE) production. Urine was cultured on Cystine Lactose Electrolyte Deficient agar, blood agar and MacConkey agar; and incubated at 370C for 24 hours. Urine cultures with pure bacterial colony counts of ≥104cfu/ml were classified as “positive” for UTI.ResultsOf a total of 552 participants enrolled into the study, 124 (23%) were urine culture positive. Prior medication use was associated with culture negativity. With urine culture as the reference standard, urinary dipstick sensitivity was poor overall (using either LE+ or NIT+ to confer a positive dipstick result still only achieved sensitivity of 66.9%). Using combined NIT+ and LE+ to confer a positive dipstick result had the highest specificity (99.2%), alongside a positive predictive value of 91.1%, and a positive likelihood ratio of 35.6. A NIT+ test alone showed highest concordance with urine culture results (percentage agreement: 86%) but still had a Cohen’s Kappa value of only 0.5, conferring weak agreement overall.ConclusionDipstick test is a poor surrogate of urine culture. However, the test may be suitable as a ‘rule-out’ test to exclude UTI, and avoid antibiotic prescription, when both NIT and LE are negative. Although dipstick continues to be in use in resource constrained settings, poor concordance with urine culture results highlights a need for better near patient tests to diagnose UTI and guide antibiotic decision-making.