Accuracy of Urine Dipstick Test in Detecting Asymptomatic Bacteriuria Among Pregnant Women Receiving Antenatal Care at Kenyatta National Hospital, Nairobi, Kenya.

Abstract

Background: When not detected and treated asymptomatic bacteriuria (ASB) in pregnancy is associated with development of symptomatic urinary tract infections, preterm deliveries, low birth weight infants, intrauterine growth restriction, preterm premature rupture of membranes and pre-eclampsia. Routine screening of pregnant women is necessary to avert the adverse outcomes, unlike in the general population where the disease is considered benign. The quantitative urine culture, the “gold standard” test for detection of ASB, is time consuming, expensive, requires special equipment and trained personnel hence not routinely available, especially in low resource settings. Although the dipstick test is readily available, cheaper, easier to perform and interpret, its accuracy and role in detecting ASB in pregnancy in this setting has not been evaluated. Methodology: This was a cross sectional study conducted among pregnant women without symptoms of urinary tract infection, who were receiving routine antenatal care at Kenyatta National Hospital, in Nairobi Kenya. Clean catch, mid-stream, voided urine specimens from 132 eligible participants were subjected to concurrent dipstick and bacteriologic culture. Markers of ASB in urine dipstick (presence of either or both leucocyte esterase (LE) and nitrites) were compared with culture. Accuracy of urine dipstick, as measured from the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio(PLR) and negative likelihood ratio(NLR) in detecting ASB was estimated using culture as the "gold standard " test. Results: Out of 320 women screened, 132(41%) were found eligible. Prevalence of ASB was 6.9%. The sensitivity, specificity, PPV and NPV was 66.7%, 74.4%, 16.2% and 96.8% for LE; 44.4%, 97.5%, 57.1% and 95.9% for nitrite; 22.2%, 100%, 100% and 94.5% for either LE or nitrite; and 88.9%, 71.9%, 19% and 98.9% for both LE and nitrite respectively. The xiii PLR and NLR for LE was 2.61 and 0.45 whereas that of nitrite was 17.76 and 0.37 respectively. Conclusion: A negative urine dipstick test very likely rules out ASB in pregnancy and the need for routine culture. On the other hand, a positive dipstick test has low accuracy in detecting ASB in pregnancy and requires confirmatory testing with culture. Treatment for ASB based on positive dipstick alone would expose a large number of women to unnecessary antibiotics and their associated side effects. Recommendation: A urine dipstick test should be done for all pregnant women as an initial screening test to exclude those without ASB. However, all dipstick positive cases should undergo confirmatory testing with culture instead of giving empirical treatment. This testing approach will ensure a high diagnostic performance, prevent unnecessary administration of antibiotics and reduce laboratory costs and workload because not all urine samples will be subjected to routine culture.