Diagnostic accuracy of loop-mediated isothermal amplification (LAMP) for screening malaria in peripheral and placental blood samples from pregnant women in Colombia

Abstract Background Pregnant women frequently show low-density Plasmodium infections that require more sensitive methods for accurate diagnosis and early treatment of malaria. This is particularly relevant in low-malaria transmission areas, where intermittent preventive treatment is not recommended. Molecular methods, such as polymerase chain reaction (PCR) are highly sensitive, but require sophisticated equipment and advanced training. Instead, loop mediated isothermal amplification (LAMP) provides an opportunity for molecular detection of malaria infections in remote endemic areas, outside a reference laboratory. The aim of the study is to evaluate the performance of LAMP for the screening of malaria in pregnant women in Colombia. Methods This is a nested prospective study that uses data and samples from a larger cross-sectional project conducted from May 2016 to January 2017 in three Colombian endemic areas (El Bagre, Quibdó, and Tumaco). A total of 531 peripheral and placental samples from pregnant women self-presenting at local hospitals for antenatal care visits, at delivery or seeking medical care for suspected malaria were collected. Samples were analysed for Plasmodium parasites by light microscopy (LM), rapid diagnostic test (RDT) and LAMP. Diagnostic accuracy endpoints (sensitivity, specificity, predictive values, and kappa scores) of LM, RDT and LAMP were compared with nested PCR (nPCR) as the reference standard. Results In peripheral samples, LAMP showed an improved sensitivity (100.0%) when compared with LM 79.5% and RDT 76.9% (p < 0.01), particularly in afebrile women, for which LAMP sensitivity was two-times higher than LM and RDT. Overall agreement among LAMP and nPCR was high (kappa value = 1.0). Specificity was similar in all tests (100%). In placental blood, LAMP evidenced a four-fold improvement in sensitivity (88.9%) when compared with LM and RDT (22.2%), being the only method, together with nPCR, able to detect placental infections in peripheral blood. Conclusions LAMP is a simple, rapid and accurate molecular tool for detecting gestational and placental malaria, being able to overcome the limited sensitivity of LM and RDT. These findings could guide maternal health programs in low-transmission settings to integrate LAMP in their surveillance systems for the active detection of low-density infections and asymptomatic malaria cases