In Bangladesh, a number of screening tests for asymptomatic bacteriuria in pregnancy are in practice. The objective of this study was to assess the validity and cost-effectiveness of these screening tests. A total of 600 apparently healthy pregnant mothers were included in this study. The validity of the screening tests was calculated against the urine culture as 'gold standard'. Incremental cost-effective ratio between the screening test methods and the least costly method (microscopic urine analysis) was calculated. Bacterial count/oil-immersion field in Gram-stained smear of urine was the most sensitive (91.7%) and specific (97.2%). Incremental cost per additional positive cases of bacterial count, leukocyte esterase and combination of leukocyte esterase and nitrite were US$3, US$25 and US$23, respectively. Gram staining may be the alternative approach to traditional routine urinalysis for the screening of asymptomatic bacteriuria during pregnancy in clinical practice in Bangladesh, as well as other developing countries.
A total of 1,661 pregnant women aged between 13 and 45 years were screened for bacteriuria by urine culture. Of the 1,661 culture results, 615 (37%) yielded no growth; 728 (43.8%) yielded no significant growth (presence of <10(5) organisms/ml urine of one or more types of bacteria); 286 (17.2%) yielded mixed growth (presence of >10(5) organisms/ml urine of more than one type of bacteria) and only 32 (1.9%) showed significant growth (presence of >10(5) organisms/ml urine of a single bacterium). Urine microscopy was also conducted. Two hundred and twenty-four (13.5%) specimens had >10 white blood cells/ml urine, of which 66 had >100 white blood cells; 13 were from the significant growth group. Three hundred and seventy-four (22.5%) specimens showed the presence of bacteria, 42 (2.5%) had red blood cells, 370 (22.3%) had epithelial cells, 58 (3.5%) had crystals, and 14 (0.8%) had yeasts. The most common bacterium isolated was Escherichia coli (12; 40%); the others included group B Streptococcus (5; 15%), Klebsiella spp (5; 15%), Diphtheroids (2), and Candida albicans (2). Fifty-two percent of tested strains were sensitive to ampicillin; 24 of 28 strains (85.7%) were sensitive to ciprofloxacin; all 7 strains tested were sensitive to nitrofurantoin and all 20 strains tested were sensitive to cotrimoxazole; 14/20 (70%) and 16/17 (94.1%) were sensitive to cephalexin and cefuroxime respectively. This study shows that asymptomatic bacteriuria does occur in pregnant women, albeit at a very low rate in an urban setting like Cheras. Urine microscopy is not specific and only serves as a guide to bacteriuria. The commonest causative organisms are those from the gastrointestinal tract and vagina. The antibiogram showed that cefuroxime and cephalexin are likely to be effective in treating bacteriuria: ampicillin must be reserved for Gram-negative organisms. For Gram-positive organisms, of which Group B Streptococcus is important, ampicillin is still effective in vitro. Nitrofurantion and cotrimoxazole have excellent activity in vitro and should be considered for therapy. 17.2% of the urine culture yielded mixed growth: likely to indicate that contamination of urine specimens still happens despite the strict instructions given to patients about the collection of a midstream urine specimen. Proper collection, appropriate transport, and the early processing of urine specimens remain essential.
First-void urine samples collected from sexually transmitted diseases (STD) clinic patients were examined by a nested polymerase chain reaction (PCR) and a commercial enzyme immunoassay (IDEIA Chlamydia) for the diagnosis of Chlamydia trachomatis urethritis or cervicitis. The primers for the PCR amplified a target in the major outer membrane protein (MOMP) gene in C trachomatis while the IDEIA detected genus-specific chlamydial lipopolysaccharide. Discrepant results were resolved by retesting urine specimens with a second (plasmid-based) PCR and taking urethral or endocervical swab results into consideration. For 231 men (chlamydial prevalence 20.4%), the sensitivity, specificity, positive and negative predictive values were 59.6%, 99.5%, 96.6% and 90.6% for urine IDEIA, 68.1%, 99.5%, 97% and 92.4% for urethral swab IDEIA and 97.9%, 99.5%, 97.9% and 99.5% for urine PCR. The corresponding rates for 66 women (chlamydial prevalence 54.6%) were 19.4%, 100%, 100% and 50.8% for urine IDEIA, 86.1%, 96.7%, 96.9% and 85.3% for endocervical swab IDEIA and 91.7%, 93.3%, 94.3% and 90.3% for urine PCR. Hence, in a high prevalence population, the urine IDEIA was a suitable alternative to the male urethral swab IDEIA but significantly less sensitive than the endocervical swab IDEIA. The urine PCR was, however, much more sensitive than the urine IDEIA for both men and women and could replace the endocervical swab IDEIA for the diagnosis of chlamydial cervicitis.