Uncomplicated malaria as a risk factor for COVID-19 duration and severity in western Kenya and Burkina Faso (MALCOV) a prospective cohort study

BACKGROUND: The relationship between malaria and COVID-19 varies across different clinical scenarios; historical malaria exposure might protect against severe COVID-19, whereas co-infection in hospitalised patients with severe disease might increase mortality. Interactions between non-severe malaria and COVID-19 remain poorly understood. We conducted a cohort study among COVID-19 patients of all ages in western Kenya and Burkina Faso to assess the effects of acute, uncomplicated Plasmodium falciparum malaria co-infection on COVID-19 outcomes in ambulatory patients. 

METHODS: Participants with laboratory-confirmed SARS-CoV-2 infection (positive rapid antigen test or reverse transcription quantitative real-time PCR [RT-qPCR]) were tested for malaria by rapid antigen tests with confirmatory microscopy. Patients with COVID-19 and malaria co-infection received artemether-lumefantrine or pyronaridine-artesunate. COVID-19 symptom course was assessed daily using FLU-PRO Plus (a validated patient-reported outcome instrument) until day 14. Viral load was measured by RT-qPCR on days 0, 3, 7, 14, and 28. The primary endpoint was time to symptom resolution on the FLU-PRO Plus. Analyses were adjusted for country, age, disease severity, and viral load. 

FINDINGS: Between Jan 8, 2021 and Jan 24, 2022, we screened 5161 participants and recruited 756 with COVID-19. 742 participants with valid malaria tests were enrolled, of which 151 (20%) had malaria co-infection and the remaining 591 (80%) did not have malaria. Patients with malaria were younger (49 [32%] aged <15 years) than those without malaria (35 [6%]; p<0·0001). Time to symptom resolution was similar between those with malaria (median 9 days [IQR 5-13]) and those without (10 days [IQR 6-13]; adjusted hazard ratio [aHR] 1·14 [95% CI 0·91-1·42]; p=0·26). Three (2%) patients with malaria and nine (2%) without malaria were hospitalised; two (1%) with malaria and three (1%) without malaria died, four from acute respiratory distress syndrome and one (in the no malaria group) from perforated peptic ulcer complicated by anaemia. Participants with malaria more frequently reported moderate-to-severe symptoms at enrolment (68% vs 60%; p=0·074), but overall symptom duration was similar (adjusted incidence rate ratio 0·95 [95% CI 0·86-1·05]; p=0·31). Previous malaria exposure significantly modified outcomes, with patients with malaria co-infection and previous exposure having faster symptom clearance than those without previous exposure (pinteraction=0·042). SARS-CoV-2 clearance was slower in the malaria group by day 7 (aHR 0·69 [95% CI 0·51-0·94]; p=0·017) but was similar between groups by day 28 (adjusted risk ratio 0·99 [95% CI 0·79-1·24]; p=0·95). INTERPRETATION: This study shows that acute uncomplicated malaria co-infection does not adversely affect COVID-19 progression when appropriately treated. Moreover, serological evidence confirms that previous lifelong malaria exposure might provide some protection, with exposed individuals having faster symptom resolution.