Performance of microscopy compared to conventional PCR in identification of soil-transmitted helminth infections among antenatal women in a low-prevalence setting

Traditional microscopy-based methods may provide inaccurate estimates of Soil transmitted helminth (STH) infections in mild intensity of infection. Therefore, we aimed to determine the prevalence of STH infections using molecular diagnostic methods and compare the diagnostic performance of microscopy with polymerase chain reaction (PCR) in stool samples collected from pregnant women in primary care settings in Puducherry, India. A singleplex PCR assay was developed to detect three species of STHs, namely Ascaris lumbricoides, Necator americanus, and Ancylostoma duodenale, by targeting the internal transcribed spacer regions (ITS1 and ITS2) of 5.8S rRNA. The PCR generated 420, 662, and 515 base pairs of DNA for the respective organisms. In addition to singleplex PCR, wet and concentration microscopy techniques were used. The results were expressed as percentages with 95% confidence intervals, and the diagnostic performance of microscopy was compared with PCR in terms of sensitivity, specificity, and positive, negative predictive values and kappa statistics. Among the 650 pregnant women included, 48.8% were aged 25 years or less, 59% were primigravida, and half were from rural areas. The overall prevalence of any STH infection was higher in PCR compared to microscopy (8.9% vs. 7.2%). The prevalence of Ascaris lumbricoides was higher by microscopy (5.4% vs 2.6%), while the prevalence of Necator americanus was higher by PCR (6.3%) than by microscopy (1.8%). No species of Ancylostoma duodenale was detected. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of microscopy for detecting any STH infection was 22.4%, 94.3%, 27.7%, and 92.5%, respectively. The agreement between microscopy and PCR for the identification is as follows: for any STH infection, k ​= ​0.12, Ascaris k ​= ​0.16, and Necator k ​= ​0.20, respectively. The prevalence of any STH infection identified by PCR was higher than microscopy, and the agreement between the two methods was poor.