Date Published: September 23, 2015
Publisher: Public Library of Science
Author(s): Xin Zhang, Yihui Rong, Lijian Sun, Liming Liu, Haibin Su, Jian Zhang, Guangju Teng, Ning Du, Haoyang Chen, Yuan Fang, Wei Zhan, Alex B. J. Kanu, Sheku M. Koroma, Bo Jin, Zhe Xu, Haihan Song, Matthew Kasper. http://doi.org/10.1371/journal.pntd.0004113
Abstract: BackgroundThe Ebola virus causes an acute, serious illness which is often fatal if untreated. However, factors affecting the survival of the disease remain unclear. Here, we investigated the prognostic factors of Ebola virus disease (EVD) through various statistical models.Methodology/Principal FindingsSixty three laboratory-confirmed EVD patients with relatively complete clinical profiles were included in the study. All the patients were recruited at Jui Government Hospital, Sierra Leone between October 1st, 2014 and January 18th, 2015. We first investigated whether a single clinical presentation would be correlated with the survival of EVD. Log-rank test demonstrated that patients with viral load higher than 106 copies/ml presented significantly shorter survival time than those whose viral load were lower than 106 copies/ml (P = 0.005). Also, using Pearson chi-square test, we identified that chest pain, coma, and viral load (>106 copies/ml) were significantly associated with poor survival of EVD patients. Furthermore, we evaluated the effect of multiple variables on the survival of EVD by Cox proportional hazards model. Interestingly, results revealed that patient’s age, symptom of confusion, and viral load were the significantly associated with the survival of EVD cases (P = 0.017, P = 0.002, and P = 0.027, respectively).Conclusions/SignificanceThese results suggest that age, chest pain, coma, confusion and viral load are associated with the prognosis of EVD, in which viral load could be one of the most important factors for the survival of the disease.
Partial Text: In the year of 2014, Ebola virus disease (EVD) was quickly widespread and caused the whole world to pay attention [1,2]. By the end of 2014, more than eleven thousand cases were reported from West African countries such as Guinea, Sierra Leone, Liberia, Senegal, Nigeria, and Mali [1,2]. However, Ebola did not stop in West Africa only, it has gone globally as cases were diagnosed in the United States and Spain [1,2].
In this study, we identified significant differences between survivors and non-survivors in terms of chest pain and coma. Moreover, the p value was close to 0.05 for symptoms such as diarrhea, anorexia and fever. These data indicate that the current Ebola outbreak is similar to previous ones [15,16]. Meanwhile, we observed some differences. First of all, EVD used to be called Ebola hemorrhagic fever, but most cases did not show bleeding or just had little fever in the current Ebola outbreak. It might be due to different strains of the Ebola virus. Secondly, occurrence of chest pain, coma and confusion were statistically significantly correlated with EVD death by Pearson’s chi-squared test (Table 1), whereas only confusion showed correlation to the survival of EVD by Cox proportional hazard model (Table 2). There could be two reasons causing the discrepancy: 1) sample size was too small; 2) Cox proportional hazard model takes survival time into consideration. Different survival time had different effects on the model, even though the clinical outcome of the patients was the same, whereas one-way ANOVA analysis only considers the clinical outcome of the patients.