Rabies is a viral infection that causes acute, progressive encephalitis and is considered to be universally fatal. However, during 2004, an unvaccinated Wisconsin patient received a new medical treatment and became the first documented survivor of rabies* who had not received preexposure vaccination or postexposure prophylaxis (PEP), suggesting the possibility of successful future interventions.1 This report describes two recent patients with rabies who were treated using therapy similar to that used for the Wisconsin patient; both treatments were unsuccessful. The report also describes the concomitant epidemiologic investigations by the Indiana State Department of Health (ISDH), California Department of Health Services (CDHS), and CDC, and the local public health responses in Marshall County, Indiana, and San Joaquin and Alameda counties in California. The findings in this report underscore the continuing need for enhanced clinical awareness of possible rabies exposure to ensure prompt PEP and timely diagnosis of rabies, especially if treatment is attempted.
On September 30, 2006, a girl aged 10 years had pain in her right arm, and her parents noticed a skin eruption on her trunk and extremities. On October 3, she began vomiting and had increased arm pain and occasional arm numbness. During her initial visit to her family's primary health-care provider on October 4, radiographs of her arm and clavicle were normal. Three to five days after her initial symptoms began, the patient's speech became difficult to understand, and she had a decreased appetite, sore throat and neck pain, and temperature of 101°F. She became irritable and agitated. A rapid Group A streptococcal antigen test and slide heterophil antibody assay were negative on October 6. The patient was hospitalized on October 7 at a community hospital, where she was found to have difficulty swallowing secretions. Her tongue had a whitish coating and was protruding from her mouth. Her complete blood count and electrolytes were normal. She was prescribed methylprednisolone for possible glossitis and fluconazole for mucosal candidiasis.
On October 8, neurologic involvement became more evident, and the attending physician arranged for transfer to a university-affiliated tertiary care pediatric hospital. On arrival at the pediatric hospital, the patient was irritable, with intermittent moments of alertness, altered mental status, and lethargy. She had slurred speech and difficulty swallowing secretions and complained of a drowning sensation. Because of difficulty breathing, low oxygen saturation, and excess secretions, the patient was intubated and placed on a mechanical ventilator. A lumbar puncture was performed, indicating a white blood cell (WBC) count of 26 cells/mm3 (normal: 0-7 cells/mm3), a red blood cell (RBC) count of 1 cell/mm3 (normal: 0 cells/mm3), a protein level of 28 mg/dL (normal: 15-45 mg/dL), and a glucose level of 89 mg/dL (normal: 40-70 mg/dL). Vancomycin, cefotaxime, and acyclovir were administered for the presumptive diagnosis of meningoencephalitis. On the second day of hospitalization, the patient experienced episodes of lethargy, somnolence, generalized skin flushing (associated with vancomycin administration), and hypersalivation.
Initial interviews of family members indicated that the patient frequently was exposed to healthy-appearing household cats and dogs but to no other animals. On the third day of hospitalization, the patient's primary-care physician told staff members at the pediatric hospital that a babysitter suggested the patient might have sustained an animal scratch or bite during June 2006. Family members did not know what type of animal might have scratched her. However, in spite of her endotracheal intubation, the patient was able to indicate that a bat had scratched or bit her. On the same day, serum, saliva, cerebrospinal fluid, and a skin biopsy from the nape of the neck (nuchal sample) were sent to CDC for rabies virologic testing, and a serum rabies-virus–specific antibody test was positive. Reverse transcription–polymerase chain reaction (RT-PCR) performed on saliva and skin samples also were positive for rabies virus amplicons, and direct fluorescent antibody (DFA) staining of the skin biopsy was positive for detection of rabies virus antigens. The patient had not received a rabies vaccine or rabies PEP.
After rabies was confirmed, the Wisconsin rabies treatment protocol1 was initiated, including antiexcitatory and antiviral therapy with phenobarbital, midazolam, ketamine, and amantadine with aggressive supportive care. On the sixth day of hospitalization, ribavirin was administered intravenously, under a Food and Drug Administration (FDA) emergency use investigational new drug protocol.† Coenzyme Q10, l-arginine, tetrahydrobiopterin, and vitamin C also were administered in an attempt to replenish neurotransmitter substrates. During hospitalization, the patient experienced multiple complications, including increased intracranial pressure, bouts of diabetes insipidus, syndrome of inappropriate secretion of antidiuretic hormone (SIADH), reversible pancreatitis secondary to ribavirin, intracranial venous sinus thrombosis, and cerebral and cerebellar herniation. In spite of a reduction in sedation drugs, the patient never regained consciousness. Because of a deteriorating clinical condition and poor prognosis, life support was withdrawn. The patient died on November 2, 2006, on the twenty-sixth day of hospitalization. Rabies virus antigen was detected in brain tissue collected postmortem.
The patient's mother reported that in mid-June, the girl had awakened her during the night and said that a bird or bat had flown into her bedroom window and bitten her. The mother saw a small mark on the girl's arm, which the mother washed and treated with an over-the-counter first aid treatment. The mother then went to the girl's bedroom to see whether an animal was present. Finding none, she assumed that the incident was a nightmare, not uncommon for the girl. Approximately 2-3 days later, an older sibling took a dead bat away from the family cat; however, the mother did not associate this event with the previous incident and did not seek rabies PEP for the girl. The mother later reported that at the time of the incident, a bedroom window was probably open without a screen in place.
After genetic sequencing of amplicons obtained from the patient's skin and saliva, on October 14, CDC characterized the infecting agent as a rabies virus variant associated with the silver-haired bat, Lasionycteris noctivagans. ISDH recommended rabies PEP for persons who had been exposed to the patient's saliva, from 7 days before onset of initial symptoms through the time of her death. A total of 66 persons received PEP, including seven members of the patient's immediate family, a health-care worker at the patient's primary-care site, nine staff members at the first hospital in which she was treated, an ambulance service worker, 17 staff members at the pediatric hospital, and 31 persons from the patient's school and community.
On November 15, 2006, a boy aged 11 years had sore throat, fatigue, and fever (101°F). He was taken to his pediatrician's office on November 16 for a previously scheduled childhood vaccination related to his recent immigration from the Philippines on October 2, 2006. He received a diagnosis of pharyngitis and was prescribed amoxicillin; the vaccinations were deferred. That evening, the boy was taken to a hospital emergency department (ED) with chest tightness, dysphagia, and insomnia. He had tachycardia (128 beats/min) and hypertension (148/99 mmHg) but no fever; his respiratory rate and oxygen saturation level were normal.
During the next several hours in the ED, the boy experienced irregular lip and mouth movements, hallucinations, and agitation. Rabies-associated signs such as aerophobia, hydrophobia, profuse salivation, and copious oral secretions were noted, and he was transported to a tertiary care pediatric hospital. Because the possibility of rabies was raised by providers at the referring hospital, infection-control measures were initiated at the pediatric hospital, including contact and droplet precautions.
The patient was admitted to the pediatric intensive care unit (PICU). He had profuse salivation and required tracheal intubation, and he experienced intermittent altered mental status. In the ED and subsequently in the PICU, the patient experienced hemodynamic instability associated with sedative administration, and he required cardiac resuscitation. An electrocardiogram showed sinus tachycardia with diffuse ST–T wave changes, and an echocardiogram indicated high systemic vascular resistance and secondary cardiomyopathy. A lumbar puncture indicated a WBC count of 8 cells/mm3, RBC count of 0 cells/mm3, a protein level of 25 mg/dL, and a glucose level of 128 mg/dL. On the basis of the patient's history, clinical signs, and symptoms, he received ketamine and midazolam infusions on the first hospital day for control of dysautonomia, as described in the Wisconsin rabies treatment protocol.1‡
Overnight, after consultation with the California Department of Health Services (CDHS), samples were obtained for rabies diagnosis, including corneal impressions, cerebrospinal fluid, serum, and saliva. On the second hospital day, the samples were sent to CDC and the CDHS Viral and Rickettsial Disease Laboratory. Rabies virus antigens were detected in the corneal impressions by DFA on November 18. After receiving this result, physicians from the hospital consulted with CDHS, CDC, and the physicians who developed the Wisconsin protocol therapy, and intravenous ribavirin and enteral amantidine, tetrahydrobiopterin, and coenzyme Q10 were administered.
The patient's family was asked about possible animal exposures. Although the parents were unaware of any specific incidents, two siblings recalled that the patient had been bitten by a dog approximately 2 years previously, when he was living in the Philippines. He did not receive rabies PEP at that time.
At CDC, rabies virus RNA was detected by RT-PCR in patient saliva samples obtained on the third hospital day. The gene sequences were similar to those of a canine rabies virus variant from the Philippines. Rabies virus antigen was detected by DFA in a nuchal biopsy that was obtained on the fourth hospital day. Serology for detection of rabies virus antibodies was performed daily by an indirect immunofluorescent assay and was negative until the twelfth hospital day, when an immunoglobulin G (IgG) titer of 1:128 was detected. The following day, the serum IgG titer increased to 1:256, and the immunoglobulin M (IgM) titer was 1:10. Rabies virus IgG was first detected in the cerebrospinal fluid on the fourteenth hospital day, with a titer of 1:8. A repeat nuchal biopsy was performed on the twentieth hospital day to assess viral clearing; rabies virus antigen was detected, but the staining intensity was less prominent and had a less organized pattern than the previous biopsy.
The patient experienced multiple complications while hospitalized, including autonomic lability, SIADH, renal insufficiency, superficial thrombophlebitis of the left lower extremity, cerebral artery spasm, subclinical seizures, mild pancreatic enzyme elevation attributed to ribavirin, and progressive heart block requiring transvenous pacing. Cerebral perfusion was monitored daily via transcranial Doppler. Additional anticonvulsant therapy was initiated on the eighteenth hospital day because of seizures. A continuous electroencephalogram (EEG) indicated bursts of electrical brain activity followed by little brain activity (i.e., burst-suppression pattern), and by the twenty-first hospital day indicated that almost no brain activity remained. Transcranial Doppler sonography results remained within normal limits.
On the twenty-fourth hospital day, the patient had diabetes insipidus, and the EEG indicated almost no electrical brain activity. A cranial computed tomography scan on the same day indicated loss of differentiation of the grey-white boundary of the brain and diffuse cerebral edema. Transcranial Doppler sonography indicated a high resistive index with no diastolic flow. Midazolam infusion was discontinued. After discussions between the family and the care team, life support was withdrawn on December 13, the twenty-seventh hospital day, and the patient died. Rabies virus antigen was detected in brain tissue collected postmortem.
To identify exposures to the patient and the need for rabies PEP among identified contacts, CDHS distributed rabies assessment tools (which vary according to exposure scenario) to the public health department in San Joaquin County, where the patient first became ill, and in Alameda County, the location of the pediatric hospital. All 13 members of the patient's family were identified as potentially exposed because they had shared food and drink with the patient and reported contact with the patient's saliva; all received PEP. None of the three staff members at the primary-care physician's office or two emergency-transport personnel were determined to require PEP. Health-care workers at the hospital in San Joaquin County were interviewed, and eight of 22 elected to begin PEP because of potential contact with saliva. No additional interventions were deemed necessary in San Joaquin County. Twenty-four health-care workers were interviewed at the pediatric hospital in Alameda County. Three received PEP for potential exposure to saliva through mucous membranes or uncovered breaks in the skin.
JC Christenson, MD, Riley Hospital for Children; BM Holm, MD, S Lechlitner, Marshall County Health Dept; JF Howell, DVM, Mona Wenger, Indiana State Dept of Health. A Roy-Burman, MD, CJ Hsieh, MD, L LaBar, A Petru, MD, S Davis, MD, N Simons, MD, Children's Hospital and Research Center at Oakland; P Apolinario, MD, San Joaquin General Hospital; K Furst, MD, San Joaquin County Public Health Dept; B Benjamin, MD, Alameda County Public Health Dept; LJ Christie, MD, D Schnurr, PhD, CA Glaser, MD, DVM, J Rosenberg, MD, B Sun, DVM, California Dept of Health Services. RE Willoughby, MD, Children's Hospital of Wisconsin, Milwaukee. CE Rupprecht, VMD, PhD, Div of Viral and Rickettsial Diseases, National Center for Zoonotic, Vector-Borne, and Enteric Diseases, CDC.
During 2000-2006, a total of 19 of the 24 human rabies cases reported in the United States were acquired indigenously.2 The history of exposure to a bat given by the patient in Indiana described in this report and the identification of a specific bat rabies virus variant support the explanation of indigenous rabies acquisition via bat bite. In contrast, the history of a dog bite in the patient from California described in this report, even though the bite occurred years before his immigration to the United States, suggests acquisition of rabies in the Philippines; this is supported by isolation and identification of a specific canine-associated rabies virus variant found in the Philippines. Typical rabies incubation periods vary from 1 to 3 months after exposure, but longer intervals have been documented.3
Human rabies is preventable with proper wound care and timely, appropriate administration of human rabies immune globulin and rabies vaccine before onset of clinical symptoms.4 PEP is recommended for all persons who have been bitten or scratched by an animal suspected to have rabies virus and for all persons whose mucous membranes have been exposed to the virus.
Twenty-seven health-care workers received PEP as part of the investigations in Indiana and California. Previous reports have described administration of PEP to contacts of humans with rabies.5 The indications for PEP among health-care workers who care for patients with rabies include exposure of mucous membranes or open wounds to infectious body fluids or tissue (e.g., saliva, tears, cerebrospinal fluid, or neurologic tissue) from the patient. Adherence to standard infection-control precautions minimizes the risk for health-care workers' exposure to rabies.4
During 2004, the first documented survival of a patient who had not received preexposure vaccination or rabies PEP occurred in a patient from Wisconsin. Treatment for this patient included an intensive protocol that included drug-induced coma and administration of antiviral drugs.1 However, the benefits of any particular experimental regimen have not been determined, and no single specific course of therapy for rabies in humans has been demonstrated effective after clinical signs manifest.6
Rabies is usually a fatal illness in humans. To consider use of the Wisconsin rabies treatment protocol, the disease must be diagnosed as early in the course as possible, which requires enhanced clinical awareness of the disease among health-care providers. Rabies should be included in the differential diagnosis of any unexplained acute, rapidly progressive viral encephalitis. Although initial signs and symptoms of rabies are nonspecific, a history of an animal bite or travel to a rabies-indigenous country, combined with clinical signs such as paresthesia, hypersalivation, dysphagia, hydrophobia or aerophobia, behavioral changes, or sudden autonomic instability, should lead to a strong suspicion of rabies. Rapid diagnosis of rabies virus infection can be beneficial to the treatment of the patient and can facilitate appropriate prophylaxis for exposed persons. In addition to current measures, other national and international interventions are needed to raise awareness, improve health education, expand diagnostic testing, and improve rabies prevention, control, and treatment.
This report is based, in part, on contributions by I Kuzmin, MD, M Niezgoda, MS, L Orciari, MS, A Velasco, MS, and P Yager, CDC.
*The patient recovered completely.
†Additional information on emergency use of investigational new drugs is available at http://www.fda.gov/cder/about/smallbiz/faq.htm.
‡Medical College of Wisconsin. Rabies protocol.
Available at http://www.mcw.edu/rabies.
Human Rabies—Indiana and California, 2006. JAMA. 2007;297(21):2340-2343. doi:10.1001/jama.297.21.2340