By Will Boggs MD
NEW YORK (Reuters Health) – A score consisting of factors present at or during the days after triage can predict the severity of Ebola virus disease (EVD) and thereby identify patients most in need of intensive treatment, according to new findings.
“As with all clinical scoring systems, it provides an objective means to assess and predict patient severity,” Dr. Mary-Anne Hartley from GOAL Global in Dublin, Ireland, and University of Lausanne in Switzerland told Reuters Health by email. “This is especially important when clinicians are under pressure to prioritize their patient interactions within the strict time limitations imposed by the high-risk environment of the ‘red zone.'”
Dr. Hartley’s team analyzed clinical and epidemiological data from 158 EVD patients admitted to the Ebola Treatment Center in Port Loko, Sierra Leone, to construct a statistically weighted disease scoring and staging system.
Factors significantly contributing to increased EVD severity at triage included age younger or older than 5 to 25 years, the presence of muscle pain or disorientation, and a combination of referral time and viral load, the team reports February 2 in PLoS Neglected Tropical Diseases.
“Something that was particularly surprising to me was that our initial, raw analysis showed that patients reporting earlier to the Ebola Treatment Center (ETC) had an unchanged (and, actually, slightly higher death rate) than those who delayed treatment,” Dr. Hartley said.
“Thankfully, deeper analysis showed that those presenting earlier had equal viral loads to those presenting later, making a strong case that the ‘early presenters’ had a more quickly developing disease, and were perhaps predisposed to more severe outcomes,” she said. “Recalculating the impact of referral time (now normalized to viral load) revealed the much more anticipated outcome that each day of delayed treatment increased the risk of death by 12%: a statistic which may encourage future health-care seeking behavior.”
Factors present after triage (daily) that were associated with worse EVD severity included age <5 years or >25 years, the presence of disorientation or hemorrhage, and fewer days spent in the ETC.
The final fatal outcome discriminative power was 89.12% for the triage score and 97.04% for the daily score.
“By better identifying the small percentage of patients in need of more intensive monitoring and treatment, high-cost interventions (such as regular biochemistry and dialysis) could become financially and practically feasible options even with the extremely limited public health resources available in the countries most at risk for uncontrolled Ebola outbreaks,” Dr. Hartley said. “This could work towards closing the gap between the death rate seen amongst patients treated in high-income countries (18%) vs. that seen in Sierra Leone (60%).”
“I would like to emphasize that all prognostic tools carry the risk of becoming self-fulfilling prophecies if incorrectly used as an indicator for palliation: dooming severely ill patients to death when the score is not reflective of clinical advances,” she said. “This score is a tool to predict clinical severity alone and should never be used to extinguish hope or effort.”
Dr. Hartley added, “Evolution is inevitable. The Ebola virus will evolve (in or out of step) with our ability to treat it and the community reaction to control it. In this changing environment, clinical scores should be equally malleable. In preparation for this, we are constructing a machine-learning tool that learns from new patients to maintain its predictive accuracy.”
A companion paper describing the interaction between EVD and malaria will appear shortly.
Full-page printable templates of the scorecards can be found online.
PLoS Negl Trop Dis 2017.