High-resolution melting analysis reveals low Plasmodium parasitaemia infections among microscopically negative febrile patients in western Kenya

Abstract

Background: Microscopy and rapid diagnostic tests (RDTs) are common tools for diagnosing malaria, but are deficient in detecting low Plasmodium parasitaemia. A novel molecular diagnostic tool (nPCR-HRM) that combines the sensitivity and specificity of nested PCR (nPCR) and direct PCR-high resolution melting analysis (dPCR-HRM) was developed. To evaluate patterns of anti-malarial drug administration when no parasites are detected, nPCR-HRM was employed to screen blood samples for low parasitaemia from febrile patients without microscopically detectable Plasmodium infections in a rural malaria-endemic setting.

Methods: Blood samples (n = 197) were collected in two islands of Lake Victoria, Kenya, from febrile patients without Plasmodium detectable by microscopy or RDTs. 18S rRNA gene sequences were amplified from extracted DNA by nPCR-HRM, nPCR, and dPCR-HRM to detect and differentiate Plasmodium parasites. The limits of detection (LoD) were compared using serial dilutions of the WHO International Standard for P. falciparum DNA. Data on administration of anti-malarials were collected to estimate prescription of anti-malarial drugs to patients with and without low parasitaemia Plasmodium infections.

Results: The coupled nPCR-HRM assay detected Plasmodium parasites with greater sensitivity (LoD = 236 parasites/mL) than either nPCR (LoD = 4,700 parasites/mL) or dPCR-HRM (LoD = 1,490 parasites/mL). Moreover, nPCR-HRM detected and differentiated low-parasitaemia infections in significantly greater proportions of patients than did either nPCR or dPCR-HRM (p-value <0.001). Among these low-parasitaemia infections, 67.7% of patients were treated with anti-malarials, whereas 81.5% of patients not infected with Plasmodium parasites were treated with anti-malarials.

Conclusions: The enhanced sensitivity of nPCR-HRM demonstrates limitations of differential febrile illness diagnostics in rural malaria endemic settings that confound epidemiological estimates of malaria, and lead to inadvertent misadministration of anti-malarial drugs. This is the first study that employs low-parasitaemia Plasmodium diagnostics to quantify the prescription of anti-malarial drugs to both non-malaria febrile patients and patients with low-parasitaemia Plasmodium infections. nPCR-HRM enhances low-parasitaemia malaria diagnosis and can potentially surmount the deficiencies of microscopy and RDT-based results in determining low-parasitaemia Plasmodium infection rates for evaluating malaria elimination efforts. The findings highlight the need for improved differential diagnostics of febrile illness in remote malaria endemic regions.

Figure 1

Comparison of molecular techniques for low-parasitaemia Plasmodium detection among samples from microscopically negative febrile patients. Total Plasmodium and P. falciparum prevalence among patient samples as determined by nPCR-HRM, dPCR-HRM, and nPCR. Error bars indicate the upper limit of the 95% CI.

Figure 2

Representative Plasmodium HRM profiles from a subset of the study samples. Distinct melting profiles, represented by change in fluorescence units with increasing temperatures (df/dt) of patient samples with P. malariae, P. falciparum, and P. ovale infections.

Figure 3

Anti-malarial treatment of febrile patients with and without low parasitaemia Plasmodium infections identified by nPCR-HRM. Numbers of patients treated or not treated with anti-malarial drugs in Sena and Tom Mboya Health Centres.