Walson, Judd LArndt, Michael Benjamin2013-04-172015-12-142013-04-172012Arndt_washington_0250O_11245.pdfhttp://hdl.handle.net/1773/22469Thesis (Master's)--University of Washington, 2012University of Washington Abstract Impact of Increased Helminth Diagnostic Sensitivity on Estimation of Risk Factors and Outcomes Michael B. Arndt Chair of the Supervisory Committee: Assistant Professor Judd Walson Global Health, Medicine, Pediatrics, Epidemiology Background: Traditional methods utilizing microscopy for the detection of helminth infections have limited sensitivity to detect infection, particularly in populations with lower helminth prevalence and burden. Newer polymerase chain reaction (PCR) assays may enhance detection of helminth infections and improve identification of risk factors for infection. However, these methods may detect low level helminth infections with limited impact on clinical outcomes. Methods: This cross-sectional study was nested within a randomized clinical trial (RCT) conducted at 3 HIV Care sites in Kenya. We performed three microscopy methods and real-time multiplex PCR for the detection of Ascaris lumbricoides, hookworm spp., Strongyloides stercoralis, and Schistosoma spp. in stool. Sensitivity for each diagnostic modality to detect helminths was calculated using positive results from microscopy and PCR as the gold standard positive. PCR cycle threshold is the amplification cycle in which the fluorescent probe signal level exceeds background fluorescence, and correlates with parasite DNA load. We utilized relative risk regression and linear regression models to evaluate the association between helminth infection as detected by either microscopy or PCR and potential risk factors or outcomes. Results: Of 307 adults surveyed, 61 (19.1%) and 21 (6.8%) were positive for one or more helminth species by PCR and microscopy, respectively (p<0.0001). PCR was markedly more sensitive (96.8%) than the microscopy techniques (33.3%) (p<0.0001), and had higher negative predictive values (99.2% vs. 85.3%). Increased infection intensity based on Ct was associated with diagnosis by microscopy (OR=1.24; 95% CI: 1.14, 1.36). PCR-identified helminth infections were significantly associated with farming occupation (RR 1.57, 95% CI: 1.02, 2.40), piped water source (RR 0.26, 95% CI: 0.07, 0.99), and with completion of primary education (RR 0.65, 95% CI: 0.43, 0.99). Microscopy-detected helminth infection was not significantly associated with any of these risk factors. Microscopy-detected helminth infections were significantly associated with lower hematocrit (mean difference of -3.56, 95% CI: -6.27, -0.85) and higher eosinophil counts (+630.92, 95% CI: 121.52, 1140.33). In contrast, these associations were not observed for PCR-detected infections, however infections diagnosed with either PCR (RR=2.42, 95% CI: 1.02, 5.76) or microscopy (RR=2.92, 95% CI: 1.29, 6.60) were associated with increased risk of eosinophilia (>500 cells/µl). Conclusion: PCR detection of helminths had greater sensitivity and NPVs compared to microscopy. PCR assays enabled identification of more potential risk factors for infection while microscopy was better than PCR in discriminating infections with clinical outcomes, such as anemia and eosinophilia due to its specific detection of higher burden infections.application/pdfen-USCopyright is held by the individual authors.diagnostics; helminths; parasites; PCR; Polymerase chain reaction; risk factorEpidemiologyParasitologyPublic healthepidemiologyThesis