1 table omitted
ON OCTOBER 17 and October 23, 1997, a man in Texas and a man in New Jersey, respectively, died from rabies. This report summarizes the clinical features of these cases and the epidemiologic investigations by the Texas Department of Health and the New Jersey State Department of Health and Senior Services, which indicated that a bat-associated variant of the rabies virus was responsible for infection in both cases.
On October 3, a 71-year-old man from Houston, Texas, developed malaise, anorexia, and sharp left-side face and ear pain that radiated to his chest. On October 7, his evaluation as an outpatient included a computerized tomography (CT) scan, which suggested left frontal and sphenoid sinusitis with normal brain parenchyma, and a laryngeal examination, which revealed left vocal-cord paralysis. On October 8, he was admitted to a hospital in Houston for further evaluation of generalized pruritus, agitation, confusion, and fever; treatment of sinusitis; and possible alcohol withdrawal. He was anxious and tremulous but mentally coherent and was treated empirically with antibiotics and benzodiazepines. During the next 2 days, he developed fever of 103.6 F (39.8 C), ocular motor paralysis, myoclonic tremors, and dysphagia, manifested by an inability to swallow his saliva. Notable laboratory findings included a peripheral white blood cell (WBC) count of 11,000/µL (normal: 4000-10,000/µL), cerebrospinal fluid (CSF) protein of 67 mg/dL (normal: <45 mg/dL), and CSF WBC count of 17/µL (normal: <5/µL). An electroencephalogram showed mild bilateral cerebral dysfunction.
On October 11, he became hypotensive and hypopneic, necessitating mechanical ventilation. On October 12, rabies was suspected; a sample of the patient's serum was sent to a commercial laboratory for evaluation for rabies virus neutralizing antibodies, and he was placed in respiratory isolation. By October 16, he was comatose with flaccid extremities. On October 17, he was found to have absent brainstem reflexes, ventilator support was withdrawn, and the patient died.
After previous unsuccessful attempts by physicians to elicit from the patient and his wife a history of animal exposure, on October 12, the wife reported that the patient had had recent contact with a bat. On August 3, while sleeping in a motel in Harrison County, the patient had been awakened by a bat that had settled on his left shoulder. He removed and disposed of the bat. The patient's wife had examined his skin immediately after removing the bat and had not detected a bite wound. She did not recall whether the bat had been clinging directly to the patient's skin or to a shirt he may have been wearing. Investigation by state and local health officials revealed that, in the bathroom of the patient's motel room, a separation between the wall and the ceiling connected the bathroom to the attic. In addition, several openings to the outside were noted in the attic, but there was no evidence of current or previous occupation by bats.
The serum sample sent to the commercial laboratory on October 12 did not demonstrate evidence of rabies virus neutralizing antibodies. However, on October 18, rabies was diagnosed by the direct fluorescent antibody (DFA) test from postmortem brain samples tested by the Houston Department of Health and Human Services. The diagnosis was confirmed at CDC using the DFA test and reverse transcriptase polymerase chain reaction (RT-PCR) on brain tissue samples. Nucleotide sequence analysis of viral RNA implicated a variant associated with silver-haired (Lasionycteris noctivagans) and eastern pipistrelle (Pipistrellus sp.) bats.
A total of 46 persons (four personal contacts and 42 health-care workers) received postexposure prophylaxis (PEP) because of possible percutaneous or mucous membrane exposure to the patient's saliva or CSF.
On October 12, a 32-year-old man from Warren County, New Jersey, developed an aching sensation in his right shoulder and neck. These symptoms persisted and progressed to include vomiting, chills, and a sore throat, prompting a visit on October 13 to an emergency department where he received oral antibiotics and an anesthetic throat spray. After presenting to his primary physician with additional complaints of fever, insomnia, agitation, and dysphagia, he was admitted to the hospital on October 14. The patient developed dysarthria, hallucinations, myalgias, and fever of 104 F (40 C) and was transferred to a referral hospital on October 15. At the referral hospital, laboratory findings included a peripheral WBC count of 10,800/µL, a creatine phosphokinase of 2500 U/L (normal: 35-185 U/L), CSF protein of 61 mg/dL, and a CSF WBC count of 1/µL. A CT scan was reported normal, and he was started on empiric broad-spectrum antibiotics for his febrile syndrome. The following day, he was electively intubated for airway protection and started on treatment for possible tetanus and herpes encephalitis.
On October 17, rabies was suspected, and serum, saliva, CSF, tears, and nuchal skin biopsy specimens were submitted to CDC for testing; serum and CSF samples were negative for rabies virus neutralizing antibodies. On October 20, the nuchal biopsy specimen tested positive for rabies antigen by the DFA test, which was later confirmed by nested RT-PCR. Saliva and tears also were positive by nested RT-PCR. Nucleotide sequence analysis of viral RNA implicated a variant associated with the silver-haired (L. noctivagans) and eastern pipistrelle (Pipistrellus sp.) bats. The patient developed severe hypotension and renal failure, and he died on October 23. On autopsy, brain tissue specimens collected for evaluation tested positive for rabies virus by RT-PCR.
On initial presentation, the patient reported exposure to his two pet parakeets but not to other animals. Additional history obtained from his wife revealed that on two separate occasions in early July, bats had been found in the living room of the patient's home. The patient had captured the bats by hand by using a cloth and had then released them outside. The patient and his wife did not recall whether he had been bitten. Investigation by state and local health officials revealed that the patient and his family had resided in a second-floor apartment of an old wooden house in poor repair. The attic had numerous openings around the chimneys and between the roofing slates. Although bats were not present at the time of inspection in October, the amount of bat guano present in the attic suggested that a colony of up to 200 bats had been present in the attic during the summer.
PEP was administered to 50 persons (eight personal contacts, 22 health-care workers at the first hospital, and 20 health-care workers at the second hospital) because of possible percutaneous or mucous membrane exposure to the patient's saliva.
A Payne, DrPH, J Nix, Bur of Animal Regulation and Care, Houston Dept of Health and Human Svcs; D Shaff, MD, A Chapman, MD, B Ghorayeb, MD, G Keith, MD, K Lamb, Columbia Spring Br Medical Center, Houston; B Nix, DVM, G Johnson, J Johnson-Minter, MD, H Guidry, MD, J Mahlow, DVM, M Kelley, MD, D Simpson, MD, State Epidemiologist, Texas Dept of Health. D Allegra, MD, E McManus, MD, G Desai, MD, Northwest Covenant Medical Center, Dover; J Hawk, Warren County Dept of Health, Washington; F Sorhage, VMD, H Ellis, MD, New Jersey State Dept of Health and Senior Svcs. Viral and Rickettsial Zoonoses Br, Div of Viral and Rickettsial Diseases, National Center for Infectious Diseases; and an EIS officer, CDC.
Four cases of human rabies were documented in the United States during 1997, including the two described in this report. This report describes the first case of human rabies in New Jersey since 1971 and the 12th case in Texas during the same period. Of those 12 humans in whom rabies had been diagnosed in Texas, seven (58%) involved variants of rabies virus associated with dog and coyote populations and three (25%), including the case in this report, involved variants of the rabies virus associated with bats. Since 1980, a total of 21 (58%) of the 36 cases of rabies diagnosed in the United States have been associated with bat variants of the rabies virus, and the silver-haired bat/eastern pipistrelle rabies virus variant has accounted for 15 (71%) of the 21.
Bats are increasingly implicated as important wildlife reservoirs for variants of rabies virus transmitted to humans. Recent epidemiologic data suggest that transmission of rabies virus can occur through minor bites from bats.1 The limited injury inflicted by a bat bite2 (compared with lesions caused by terrestrial carnivores) and situations in which the exact exposure history is unavailable may limit the ability of health-care providers to determine the risk for rabies resulting from an encounter with a bat.
In all potential human exposures involving bats, the bat in question should be safely collected, if possible, and submitted through local or state health departments for rabies diagnosis. PEP is recommended for all persons who have sustained bite, scratch, or mucous membrane exposures to a bat unless the bat is available for testing and is negative for evidence of rabies. Although both cases described in this report had histories of contact with a bat, neither reported being bitten. Of the 21 cases of human rabies reported since 1980 that were caused by bat-associated rabies virus variants, only one had a definite history of a bat bite. Therefore, PEP also is appropriate even in the absence of a demonstrable bite, scratch, or mucous membrane exposure in situations in which there is reasonable probability that such contact occurred (e.g., a sleeping person awakens to find a bat in the room or an adult witnesses a bat in the room with a previously unattended child or mentally disabled or intoxicated person). This recommendation used in conjunction with current Advisory Committee for Immunization Practices (ACIP) guidelines3 should maximize the ability of health-care providers to respond to situations in which accurate exposure histories may not be obtainable and should minimize inappropriate PEP.
ACIP recently recommended a change in the administration of human rabies immune globulin (HRIG) for PEP: as much as possible of the full dose of HRIG should be thoroughly infiltrated into and around the wound(s). Any remaining volume should be administered intramuscularly at a site distant from vaccine inoculation.
Because bat rabies has been documented in the 49 continental United States4 and reduction of bat populations is not a feasible, practical, or desirable strategy for rabies control in bats, human and domestic animal contact with bats should be minimized. Bats should be physically excluded from houses and surrounding structures by sealing potential entrances.5 In addition, bats should never be handled by untrained and unvaccinated persons without safety precautions and should never be kept as pets.
In both of the cases in this report, many health-care workers received PEP. Strict adherence to universal precaution procedures should minimize the number of persons who need PEP because of exposures of mucous membranes or nonintact skin to potentially infectious body fluid.6
Human Rabies—Texas and New Jersey, 1997. JAMA. 1998;279(6):421-422. doi:10.1001/jama.279.6.421