Abstract: On August 8, 1994, a 46-year-old virologist working alone in a biosafety-level-3 laboratory used a high-speed centrifuge to clarify a harvest of infected Vero cells containing Sabiá virus. The centrifuge contained six 250-ml bottles in a rotor with an intact O-ring to seal the contents during centrifugation. Each screw-capped polycarbonate bottle contained approximately 200 ml of tissue-culture fluid. The centrifuge was run at 10,000 rpm for 10 minutes (10,200×g) at a temperature setting of 4°C. The virologist observed no indication of a problem during the centrifugation process. On opening the lid of the rotor to remove the centrifuge bottles, he noted that the outside of one bottle was wet and that fluid had leaked into the bottom of the rotor. No obvious break was identified at the time, and the virologist was wearing a surgical mask, a disposable solid-front gown, and gloves. He had no abrasions or scratches on his hands.
The virologist used a
second pair of gloves during the decontamination of the rotor, but did
not wear a positive air-purifying respirator, although it was available.
He decontaminated the spillage by pouring a concentrated solution of
sodium hypochlorite (5.25 percent) directly into the rotor bucket as
well as inside and outside the bottle that had leaked. The combined
bleach and liquid in the rotor were then absorbed with paper towels.
After the incident, the virologist continued working alone in the
laboratory for another three to four hours. All his protective garments
as well as other contaminated material in the laboratory were
autoclaved. Initially, he did not report the incident because he
believed that no exposure to virus had occurred.
On August 16, 1994, the virologist noted myalgias, a mild headache, a stiff neck, and fever while driving home to New Haven, Connecticut, after a weekend visit to Boston. He treated himself with ibuprofen for two days before seeking medical care. On questioning, he described recrudescences of Plasmodium vivax infection that had never been treated with primaquine. He was concerned that this fever could represent a relapse of malaria. He initially did not recall any serious laboratory exposures. On physical examination he appeared mildly ill, with a temperature of 37.6°C (99.8°F), a pulse of 89 beats per minute, and a blood pressure of 130/80 mm Hg. The only remarkable features were mild conjunctival injection and shotty cervical nodes in the anterior chain. Laboratory studies performed that afternoon revealed a hematocrit of 42 percent, a white-cell count of 2600 per cubic millimeter, a platelet count of 138,000 per cubic millimeter, and an alanine aminotransferase level of 63 U per liter; urinalysis revealed moderate proteinuria (2+). After a smear proved negative for malaria, further review of possible infectious exposures led the patient to recall the August 8 laboratory incident with Sabiá virus.The patient was immediately hospitalized and treated with intravenous ribavirin at a dosage used by the Centers for Disease Control and Prevention (CDC) for other arenavirus infections (a loading dose of 30 mg per kilogram of body weight, followed by a dose of 15 mg per kilogram every six hours for four days, and then by a dose of 7.5 mg per kilogram three times daily for six days). Pretreatment blood samples were sent for viral culture and examination by the polymerase chain reaction (PCR) for the presence of Sabiá virus RNA. PCR testing for Sabiá virus was reported to be positive on hospital day 2. The reverse-transcription PCR technique produces a fragment of 180 base pairs by using one primer specific for arenavirus in combination with one specific for Sabiá virus. Controls consisted of Sabiá virus RNA extracted from infected cell monolayers and normal human serum (NEJM, 1995).