Marburgviruses: An Update

Caterina M. Miraglia, DC, MLS(ASCP)


Lab Med. 2019;50(1):16-28. 

In This Article


Differential diagnosis includes malaria, gram-negative septicemia, plague, typhoid fever, leptospirosis, and rickettsial diseases.[22] Health care workers in nonendemic areas should be aware that MVD could be imported by travelers and should suspect filoviral infection as the cause of fever in a patient with history of recent travel to an endemic area, when the clinical status of the patient quickly worsens and when hemorrhage is present.[14] If VHF is suspected, local and state health departments and the Centers for Disease Control and Prevention Special Pathogens Branch must be notified promptly.[12]

MVD and EVD are indistinguishable based on signs, symptoms, and pathologic characteristics; laboratory testing is needed to identify the causative agent.[22] During the early disease stage, diagnosis can be achieved by antigen-capture enzyme-linked immunosorbent assay (ELISA), IgM-capture ELISA, real-time polymerase chain reaction (RT-PCR), or virus isolation, which must be performed in a biosafety level (BSL)–4 laboratory. In late disease/convalescence, IgG-capture ELISA can be used to determine the diagnosis. Retrospective diagnosis in dead people or animals can be accomplished by viral isolation, RT-PCR of blood or tissues, and immunohistochemistry staining of tissues.[52] The Centers for Disease Control and Prevention (CDC) recommends testing acute paired specimens at time points 0 to 4 and 4 to 9 days in a BSL-4 reference laboratory using the various methods described herein.[12]

For ELISA (antibody or antigen detection), whole blood, serum, or plasma can be used. For polymerase chain reaction (PCR) methods, whole blood, tissues, serum, or plasma are appropriate; however, whole blood or tissue is preferable.[53] Heparin anticoagulant tubes cannot be used because heparin interferes with PCR testing.[54] Tissue specimens are used for immunohistochemistry to detect viral antigens and should be fixed in formalin.[53] Filoviral morphology can be evaluated via electron microscopy. Culture and electron microscopy techniques require BSL-4 facilities and are time consuming. Nucleic acid detection is sensitive and specific, results are obtained quickly, and testing can be performed at the site of an outbreak. Oral and nasal swabs can also be tested via PCR but are less sensitive, and those with a negative result must be retested using a more sensitive assay.[54]

Numerous improvements have recently been made to filoviral laboratory assays to enhance sensitivity, specificity, and turnaround time, and to increase availability and ease of use in the field during an outbreak. The RealStar Filovirus Screen RT-PCR Kit 1.0 (QIAGEN N.V.), which detects all ebolaviruses and marburgviruses by targeting the L gene, was evaluated for analytical sensitivity. Using the LightCycler 480 II (F. Hoffman-La Roche, Ltd.), 11 to 67 copies of filoviral RNA per reaction were measured, with a probability of 95%, which provides suitable analytical sensitivity.[55]

A single-step RT-PCR assay was developed for detection of ebolaviruses and marburgviruses. Primers for each were designed for the nucleocapsid protein. The assay had high analytical sensitivity and specificity, with no cross-reactivity of the primers for other filoviruses. The limit of detection for MARV was 10−7 and for RAVV, it was 10−5. The total assay time took slightly longer than 1 hour, allowing for rapid identification.[56]

A multiplex PCR/LDR (ligase detection reaction) assay detects category A agents, VHF viruses (including EBOV, SUDV, RESTV, MARV), and variola. The assay detects 32 different viral isolates. The limit of detection for MARV was 53 RNA copies per mL. The assay still must be clinically validated.[57]

Screening tests for acute febrile illness outbreaks and for use for surveillance are important tools; a TaqMan Array card (Thermo Fisher Scientific) was created for these purposes. It tested for 26 pathogens known to cause acute febrile illness in Africa, including MARV. Analytical specimens and 1050 clinical specimens were screened using RT-PCR, and the assay was compared to individual assay RT-PCR. The lower limit of detection was 104 copies per mL of blood and had an overall 88% sensitivity and 99% specificity. The assay took 2.5 hours to perform and tested 6 to 8 specimens at a time.[58]

Full viral genome sequencing allows for more accurate identification. A microarray assay using probes of EBOV, SUDV, RESTV, MARV, and RAVV was tiled on a microarray chip, enabling the full viral genome in a specimen to be compared to the full genomes of viruses.[59]

A multiplex suspension bead array assay was developed by conjugating oligonucleotides specific for ebolaviruses and marburgviruses to magnetic beads. The assay was able to accurately and rapidly recognize EBOV, SUDV, TAFV, MARV, and RAVV in virally spiked human serum specimens.[60]

Standardized and validated point-of-care testing is needed for use at the site of an outbreak. A prototype antigen-capture ELISA was produced for the VP40 of EBOV, SUDV, and MARV. A polyclonal antibody for MARV VP40 was tested for its ability to bind to MARV VP40 in human serum via ELISA; the limit of detection was 1 ng per mL. A prototype lateral flow immunoassay (LFI) was also developed for VP40 of EBOV, SUDV, and MARV. Strips contained polyclonal antibodies to MARV VP40; they were tested with serial dilutions of human serum containing MARV VP40. The test line was detectable at serial dilutions between 300 and 0.1 μg per mL. These tests had limits of detection that could allow them to be used for evaluating clinical specimens.[61]