Viral Load Is Key in Predicting Ebola Mortality

The amount of Ebola virus in an infected person’s blood is a strong predictor of death, according to a recent study by Ourmar Faye, of the Arbovirus and Viral Hemorrahagic Fever Unit at the Pasteur Institute of Dakar in Senegal, and colleagues. The results of their investigation were published in PLOS Medicine on December 1, 2015.

The researchers set out to find out whether the amount of Ebola in the blood (viremia) of patients with Ebola virus disease (EVD) could be used to evaluate the case fatality ratio (CFR). Currently, CFR varies “over time and space for reasons that are not fully understood.”

Data for 699 patients who were admitted into hospitals in the Conakry area of Guinea between March 1, 2014 and February 28, 2015 with EVD were analyzed. The mean age of the patients was 31 years, and 47% were female. The researchers “restricted the univariable analysis of viremia and outcome to the 548 (78%) patients whose samples were collected within 7 days of symptom onset, a time period when viremia appeared to be stable.”

The patients were divided into categories of low, intermediate, and high viremia. The corresponding CFRs were 21%, 53%, and 81%, meaning that viremia is a “strong predictor of death that can partly explain variations in the CFR of EVD.” From March to July of 2014, the CFR of the study population increased from 35% to 49% which “coincided with an order of magnitude increase in mean viremia and a surge in the proportion of patients in the high viremia group.”

The results explain some of the variation in CFR for EVD patients, and further gives scientists an additional tool for testing prospective treatments. The researchers suggest, “As Ebola viremia is a strong predictor of mortality, comparing patient outcomes within the same viremia category (stratified comparison) will be the most efficient for testing the efficacy of a new treatment.” Treatments may have different effects depending on the patient’s viremia.

This study only included people who were hospitalized, and so the results may not apply to a wider population. Also, the analysis was retrospective and there was no active follow-up.