In 2004, a virologist at USAMRIID was working in a BSL-4 laboratory with mice that had been infected 2 days before with a mouse-adapted variant of the Zaire species of Ebola virus (ZEBOV) (2). The virulence and infectious dose of this variant of ZEBOV are unknown in humans; wild-type virus has a case-fatality rate of up to 90% (3).

The person had been following standard procedure, holding the mice while injecting them intraperitoneally with an immune globulin preparation. While the person was injecting the fifth mouse with a hypodermic syringe that had been used on previous mice, the animal kicked the syringe, causing the needle to pierce the person’s left-hand gloves, resulting in a small laceration. The virologist immediately squeezed the site to force the extravasation of blood. After decontamination of the blue suit in the chemical shower, the injured site was irrigated with 1 liter of sterile water and then scrubbed with povidone-iodine for 10 minutes.

In terms of exposure risk, the needle was presumed to be contaminated with virus-laden blood, although it was suspected that low levels of virus were present on the needle. The animals had not yet manifested signs of infection, and much contamination may have been removed mechanically when the needle pierced the gloves. The local decontamination of the site also reduced potential for infection.

USAMRIID medical, scientific, and executive staff concluded that the person with potential exposure warranted quarantine in the MCS. Contact plus airborne precautions (gown, gloves, N95 mask, eye protection) were used, with a plan to upgrade to BSL-4 precautions for signs or symptoms of illness. These extra precautions were instituted while the patient was asymptomatic for several reasons: 1) the timing of initial clinical manifestations with regard to potential for shedding virus were not known for this specific isolate in human infection; 2) there was interest in ensuring all infection control procedures were being followed appropriately in advance of clinical illness; and 3) there was interest in reducing any potential confounders, such as a caregiver transmitting a febrile respiratory infection to the patient, which might lead to unnecessary procedures or additional isolation. The person was monitored for routine vital signs; daily laboratory studies (coagulation studies, blood counts, chemistries, viral isolation, D-dimer) and regular physician assessments were performed.

Over the next several days, discussions were held with several internationally recognized filovirus experts regarding potential treatments or postexposure prophylaxis options. Local and state public health officials were also notified. The consensus opinion was that there was no safe, readily available source of immune plasma and little evidence existed to support its use. Emergency investigational new drug (IND) protocols were established for treatment with recombinant nematode protein (rNAPc2) and antisense oligomers, with the intention to consider implementation only if the patient demonstrated evidence of infection.

Ultimately, none of the 5 mice had confirmed viremia at the time of the incident. The patient did not become ill or seroconvert and was discharged after 21 days. The story received national and local media attention (4,5).

In planning for USAMRIID (established in 1969), the decision was made to include a maximum containment (now termed BSL-4) capability to care for 2 personnel who may have been exposed to a biologic agent. This MCS would also be available for managing occupational exposures at USAMRIID or other government agencies to diseases requiring containment (R. McKinney, pers. comm.).

The 1,288-square-foot MCS includes 2 patient rooms and a treatment room and is equipped for intensive care monitoring, ventilator use, and teleradiology capability. It has an independent ventilation system and a chemical shower for decontaminating caregivers’ encapsulating suits (identical to those worn in the BSL-4 laboratories), and it is isolated from adjacent areas by doors fitted with airtight gaskets. This latter feature earned the facility the moniker “the slammer,” which was popularized in the book The Hot Zone (6).

The MCS is staffed by USAMRIID personnel, with augmentation and specialized care provided by nearby medical center staff, who serve on a medical augmentation team (7). Passage of consumables and supplies in and out of the suite occurs through 3 conduits: a double-door autoclave, an ultraviolet light passbox, and a disinfectant dunk tank, which enables decontamination and transport of specimens to external areas within the facility for laboratory analysis. The MCS is maintained under negative pressure. Air undergoes HEPA filtration upon entry to and exit from the facility. The septic system links into USAMRIID’s laboratory sewer system, which undergoes steam sterilization.

Twenty-one patients have been considered candidates for admission to the MCS (Table 1) (7,8). Eighteen were USAMRIID investigators and 3 were from elsewhere. Four patients (3, 6, 7, and 18) would likely not be admitted today: 2 involved dengue virus (now a BSL-2 pathogen) and 1 each involved Japanese encephalitis B virus and Rift Valley fever virus (BSL-3 pathogens with licensed and investigational vaccines, respectively) (1).

Three patients (6, 7, and 15) were managed in an improvised manner. The MCS was unavailable for 1 patient because the unit was undergoing maintenance. Thus, the observation was conducted in another set of rooms without high-level containment features. Two other patients (16 and 19) were deemed low risk; consequently, isolation was permitted under more conventional conditions.

Of the remaining 14 admissions after potential exposure to BSL-4 viruses, 8 involved percutaneous injury and 6 involved potential aerosol exposure. Eight persons (5 evaluated for exposure to Lassa virus, 2 for Machupo virus, and 1 for Junin virus) received immune plasma (prepared from recovered patients in virus-endemic areas); 1 patient potentially exposed to Lassa virus also received intravenous ribavirin. No patient developed disease or seroconverted. From 1985 through 2003, no potential exposure in a USAMRIID BSL-4 laboratory was deemed a high enough risk to require quarantine.

Laboratories remain a potential venue for exposures to BSL-4 viruses (1). Filoviruses, in particular, have been associated with laboratory-acquired infection, being first identified after exposure to African green monkeys in Marburg, Germany, in 1967 (9). Since then, laboratory-acquired Ebola virus infections have occurred in England (10) and Côte d’Ivoire (11). The death of a Russian researcher in 2004 (12) from laboratory-acquired Ebola virus infection and USAMRIID’s recent experience demonstrate the seriousness of this issue.

Although USAMRIID periodically manages potential laboratory exposures (13–16), potential exposures in BSL-4 laboratories are rare. Nonetheless, as more facilities conduct research on viruses requiring BSL-4 containment, cases such as the one presented herein may become more commonplace.

The decision to place someone in quarantine is a difficult one. When this potential exposure occurred, several alternatives were considered: sending the patient home with periodic home or clinic assessments, admission to a medical center, and admission to the MCS. Some less serious potential exposures had been managed with twice-a-day vital sign assessments and exclusion from the laboratory. The current situation appeared to present a much higher risk to the person than past potential exposures: the patient had a break in the skin caused by a potentially contaminated needle, a route of infection known to transmit efficiently and associated with enhanced risk for death (17). Therefore, the most reasonable approach was determined to be quarantine in the MCS, thus enabling closer monitoring than could be provided at home and ensuring the safety of caregivers and family members.

Quarantine in a hospital was considered. This option presents certain safety challenges in an unprepared facility, including safe handling, transport, and analysis of laboratory specimens within the hospital; safe disposal of waste; potential reluctance of hospital staff (unfamiliar with viral hemorrhagic fevers) to care for such an infected person; and lack of a specific area within the hospital configured for handling of this type of patient.

There are advantages and disadvantages to using the MCS as a stand-alone medical facility. It enables close monitoring separate from other patients (thus eliminating risk of nosocomial spread and cross-contamination); its personnel are already trained in managing a patient in containment; public access is limited; a proven system is in place for waste disposal; and an on-site containment laboratory (with the ability to culture virus or perform sophisticated diagnostic testing under containment conditions) reduces risk for infection of clinical laboratory personnel and contamination of laboratory equipment. Disadvantages of a stand-alone facility include lack of ready access to consultative physicians; critical care nursing; radiologic and other imaging studies; blood products, medications, and resuscitative procedures; and other services available at a large medical center. Activities undertaken to compensate for these deficiencies include staffing the MCS with intensive care and infectious disease physicians and having other consultants available, as needed. Moreover, ventilator and dialysis machines and blood product and laboratory support can be kept on stand-by status within or near the facility. However, it is easier to have an isolation unit located with or within a major medical center, as has been conducted elsewhere (18).

One might question whether a facility such as the MCS is the most appropriate place to isolate an infected patient. The US Centers for Disease Control and Prevention (CDC) advises that patients with viral hemorrhagic fevers can be managed safely in a conventional hospital with, at the most, airborne and contact precautions (19–21). Those patients with potential exposures (close or high-risk contacts of infected persons) who are not ill should be placed under surveillance with twice-a-day temperature checks, and a period of observation is appropriate (19). However, setting(s) for this observation period were not specified. Home observation works well for potential exposures deemed low risk; however, waiting for a patient at higher risk for highly hazardous or contagious diseases to manifest fever at home is not ideal.

The MCS was constructed with the premise that certain rare situations might call for extraordinary precautions to isolate victims of severe contagious diseases. These precautions reduce risk for a virus such as Ebola being introduced into the community by minimizing risk for nosocomial spread and optimizing known effective infection control practices. Although these precautions are useful for filoviruses, they may be more useful for other viral hemorrhagic fevers that are transmitted more readily by the aerosol route or are potentially adaptable to local animal reservoirs. The facility may provide some reassurance to the community (and thus serve to lessen public anxiety related to a filovirus exposure) and to laboratory researchers that there is a place for their care if they become infected. A patient with a filovirus infection in an unprepared medical facility would be handled as safely as possible, using CDC guidelines (if the disease were recognized). It is acknowledged that BSL-4–like infection control precautions may not be necessary for observation or illness. However, most clinical experience managing filovirus infections is from sub-Saharan Africa, where increased temperature and humidity may reduce stability of viruses in aerosol (22). Contrast that environment with a US hospital where air is cool, dry, and recycled within the facility, a setting potentially more conducive to airborne virus spread.

A laboratory-acquired case of Sabia virus provides an example of how a patient infected with a BSL-4 agent might be managed safely in a community hospital but also provides cause for caution and heightened vigilance, especially among facilities that might receive patients referred from containment laboratories (23). The Sabia virus–infected scientist did not report the initial potential exposure and only came to the attention of healthcare providers after 5 days of illness. He was evaluated initially at a tropical medicine clinic and was subsequently referred to an emergency department. Although the department was notified that the arriving patient might have been infected with an arenavirus, there was a 12-hour delay before heightened infection control measures (specific for managing a viral hemorrhagic fever patient) were instituted. A total of 142 persons were identified as potential case-contacts, including 61 workers in the hospital clinical laboratory. Although no secondary cases occurred, potential risk and anxiety of contacts, as well as costs of an investigation by 3 agencies (CDC, the Connecticut Department of Public Health, and Yale University) and a 6-week period of surveillance, argue in favor of 1) an aggressive program of reporting and evaluating any mishap or potential exposure occurring in a containment laboratory; 2) use of facilities familiar with and prepared in advance, when possible, for managing a similar patient; and 3) a preestablished method for surveillance and site for potential quarantine of high-risk exposures.

USAMRIID is not unique in foreseeing the need for a special isolation unit. Emory University and the University of Nebraska maintain special isolation wards for patients with potentially contagious, highly hazardous diseases, and recommendations on design and planning for biocontainment patient care units have recently been published (18).

In our recent case, the patient voluntarily entered the MCS. Had the patient refused to be quarantined, decision and authority on forcible quarantine would have rested with local or state health departments. Although there is some variability in local and state regulations, authority to enforce isolation and quarantine derives from the states’ power to “safeguard the health, safety, and welfare of its citizens” (24).

There are no approved treatments or postexposure prophylaxis regimens for filovirus infections. Use of passive immunotherapy was considered; however, no studies support a definite benefit, and no readily available safe source of such products exists. On the basis of limited data indicating improved survival in rhesus macaques challenged with ZEBOV and treated with rNAPc2, an emergency IND protocol was obtained for using rNAPc2 (25). Another emergency IND protocol was obtained for use of antisense oligonucleotides on the basis of demonstrated safety with these compounds for other indications (26). Use of either protocol was without proven safety or efficacy in Ebola virus–infected humans. Both products were available for therapeutic use had the patient developed infection with clinical manifestations that warranted aggressive treatment.

Subsequent studies have demonstrated promise for treatment of ZEBOV infections with antisense oligonucleotides and small interfering RNAs (27,28). An effective vaccine would reduce inherent hazards in working with these viruses. There have been some recent developments with virus-vectored vaccines with and without naked DNA vaccine priming (29–31). A phase I human study of such a vaccine is ongoing (32).

Given increasing interest in construction of additional laboratories for study of BSL-4 agents, potential exists for clinicians to manage an occupational exposure to these viruses. Our experience led us to formulate a stepwise approach that might help others plan for and manage similar incidents.

There are few institutions in the United States currently capable of working with viruses that require BSL-4 containment, although the list is expected to expand in the near future. This article highlights medical issues and provides management considerations on the basis of USAMRIID’s experience related to a recent potential exposure to a filovirus. The expectation is that as other facilities contemplate conducting research with BSL-4 pathogens, this report may enable them to improve their preparation for potential exposures in the future.