Symptomatology of Case-Patients in the 2012 African Filovirus Outbreaks — Potential for the Detection of Imported Cases in The United States

BACKGROUND:   A Marburg Hemorrhagic Fever (MHF) case imported into the United States from Uganda in 2008 demonstrated the direct effect that Filovirus outbreaks can have on the U.S. population.  In 2012, four Filovirus outbreaks were detected in Uganda and the Democratic Republic of Congo, including one MHF and three Ebola Hemorrhagic Fever (EHF) outbreaks.  A thorough understanding of the clinical and historical presentation of Filovirus case-patients is crucial to detecting cases that may enter the country from endemic areas.  The symptomatology of cases in the 2012 outbreaks was examined, and the NDSS CSTE Viral Hemorrhagic Fever (VHF) case definition evaluated for its ability to detect the cases.

METHODS:   Suspect case epidemiologic and clinical information was collected in a standardized case report form.  Confirmatory blood testing was performed by Polymerase Chain Reaction (PCR) and antigen, IgM, and IgG serology.  Symptom frequency was analyzed for all four outbreaks.  Confirmed and probable cases were examined against the case definition.  The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for the case definition’s ability to detect confirmed and probable cases as suspects, and the sensitivity calculated for its ability to correctly classify laboratory confirmed cases as confirmed.

RESULTS:   Of 108 confirmed and probable cases examined, the majority displayed the following symptoms:  fever (92.6%), appetite loss/anorexia (81.5%), vomiting (78.7%), intense fatigue (77.8%), headache (70.4%), diarrhea (64.8%), abdominal pain (60.2%), and muscle and/or joint pain (57.4%).  Only 43.5% exhibited unexplained hemorrhage.  Ten cases were missed by the case definition, which demonstrated 90.7% sensitivity, 94.7% NPV, 28.7% specificity, and 18.0% PPV for classifying outbreak cases as suspects, and 51.4% sensitivity for correctly classifying confirmed cases.

CONCLUSIONS:   The symptoms displayed by the majority of Filovirus case-patients are included in the case definition except intense fatigue and appetite loss/anorexia.  Including these symptoms in the case definition should be considered, since two cases displaying only fever, fatigue, and appetite loss would have been missed.  The additional eight cases that would have been missed had no fever reported or unknown fever history. The majority of cases displayed non-hemorrhagic symptoms.  Sensitivity and NPV of the case definition could be improved.  The specificity and PPV were appropriately low, since case elimination should occur through diagnostic testing.  The case definition lacks the ability to classify retrospectively confirmed convalescent cases identified by antibody detection.  U.S. clinicians should carefully consider patients with appropriate travel/exposure history and symptoms as possible Filovirus cases.