**3.1 Demographic Information**

The median age of patients was 29 years and ranged from 2 to 82 years. Forty cases (37%) were less than 20 years of age and 20 (19%) were 60 years or older. Eighty-three cases (77%) were male and 25 (33%) were female. Cases were reported from 35 states and territories (Figure 1); California and Texas were the only states with more than 5 cases and reported 20 cases (19%) and 19 cases (18%) respectively. Of the states not represented in this series, all had reported a case of human rabies within the last century with the exception of Hawaii, North Dakota, Wyoming and Vermont. Illness onset occurred in all months but was most likely to occur during the fall (August, September, and October) when compared to other seasons (P-value = 0.001). Similarly, definite exposures occurring within the United States were more frequent in fall and summer months though this observation was not statistically significant (P-value = 0.27).

Fig. 1. Geographic Distribution of Human Rabies Cases in the United States, 1960–2010

#### **3.2 Source of infection**

254 Non-Flavivirus Encephalitis

A total of 108 cases of human rabies were reported in the United States from 1960 through 2010 (Table 1). One hundred and four cases (96%) died and 4 (4%) survived. Fifty-three cases were diagnosed antemortem while 55 cases were diagnosed postmortem. Fifty-one of the cases diagnosed antemortem had positive laboratory diagnostic test results with the remaining 2 cases diagnosed based on the signs, symptoms, and clinical history alone

The median age of patients was 29 years and ranged from 2 to 82 years. Forty cases (37%) were less than 20 years of age and 20 (19%) were 60 years or older. Eighty-three cases (77%) were male and 25 (33%) were female. Cases were reported from 35 states and territories (Figure 1); California and Texas were the only states with more than 5 cases and reported 20 cases (19%) and 19 cases (18%) respectively. Of the states not represented in this series, all had reported a case of human rabies within the last century with the exception of Hawaii, North Dakota, Wyoming and Vermont. Illness onset occurred in all months but was most likely to occur during the fall (August, September, and October) when compared to other seasons (P-value = 0.001). Similarly, definite exposures occurring within the United States were more frequent in fall and summer months though this observation was not statistically

Fig. 1. Geographic Distribution of Human Rabies Cases in the United States, 1960–2010

**3. Results** 

**3.1 Demographic Information** 

significant (P-value = 0.27).

The majority of human rabies cases (78/108; 72%) resulted from exposures that occurred within the United States and its territories. However, exposures in 12 foreign countries were responsible for a total of 30 imported cases. Exposures in Mexico were the most common and accounted for 13 imported cases. Animals were linked epidemiologically to 98 cases (91%) either by exposure history or RABV variant typing. Transplantation of infected organs or tissue was responsible for 5 cases (5%) and exposure to a laboratory RABV was implicated in two cases (2%). In 3 cases (3%) there was no history of exposure and no RABV variant identified to suggest a likely source of infection. The animal species linked epidemiologically to human rabies were bats (48/98; 49%), dogs (37/98; 38%), skunks (5/98; 5%), foxes (2/98; 2%), a cat (1/98; 1%), a bobcat (1/98; 1%), and a raccoon (1/98; 1%). The species responsible for infection was unclear in three cases in which the RABV variant identified did not match the exposure history; patient 24 reported a bite from a stray cat while a RABV variant associated with skunks was identified, patient 55 reported contact with a sick cow that later died while a RABV variant associated with bats was identified, and patient 101 had a history of exposure to a dog and a fox while the RABV variant identified was most closely related to viruses found in bats. These appear to have been spillover infection from a primary reservoir species to another animal. In all other cases where an animal exposure was reported the RABV variant identified matched the species of the exposing animal.

#### **3.3 Exposure history**

A definite history of exposure to RABV was reported in 54 cases; 47 reported an animal bite, 2 involved exposures to laboratory RABV, and 5 had undergone transplantation of infected organs or tissue. A probable exposure was reported in 26 cases, 21 of which involved a probable exposure to a bat and 5 described a probable exposure to a dog. Of the 28 cases with an unknown exposure, a bat RABV variant was identified in 16 cases, a dog RABV variant was identified in 8 cases, and RABV variants associated with raccoons and skunks were identified in 1 case each. Among the definite exposures involving animal bites, 27 cases were indigenous cases involving bites from 12 bats, 7 dogs, 4 skunks, 2 foxes, 1 cat, and 1 bobcat. Only 2 of the 20 imported cases involved a RABV variant not associated with a domestic animal species. Only 9 animals involved in exposures were available for diagnostic testing, though all tested positive for RABV antigen. With respect to seasonality, rabies cases associated with bats (either by exposure or by identification of a bat RABV variant) were more likely to have onset of illness during the fall when compared to all other cases excluding those acquired through transplantation (OR 3.30; 95% CI 1.45–7.54). Exposures to bats were also more likely to occur during fall months when compared to exposures to other animals (OR 16.50; 95% CI 1.83–148.61).

#### **3.4 Prophylaxis**

Sixteen patients received PEP prior to the onset of symptoms. All of these cases occurred prior to 1980 and before the introduction of modern cell culture vaccines. Only two cases completed the PEP regimen according to recommended guidelines and can be considered true failures. The failure of PEP to prevent disease in the remaining 14 patients was attributed to either a delay in administration (i.e. administered greater than 72 hours after exposure), receipt of too few doses of vaccine, or failure to administer rabies immune globulin. Two patients had received rabies vaccine prior to their exposure: patient 22

Human Rabies Epidemiology and Diagnosis 257

**n (%)** 

instability 29 (26.9%) 47 (32.6%) 0.76 (0.44–1.31) 0.34

combativeness 55 (50.9%) 86 (59.7%) 0.70 (0.42–1.16) 0.20 Nausea or vomiting 38 (35.2%) 66 (45.8%) 0.64 (0.38–1.07) 0.09 Ataxia 20 (18.5%) 38 (26.4%) 0.63 (0.34–1.17) 0.17 Anorexia 19 (17.6%) 37 (25.7%) 0.62 (0.33–1.15) 0.17 Insomnia 11 (10.2%) 27 (18.8%) 0.49 (0.23–1.04) 0.08

delirium 67 (62.0%) 123 (85.4%) 0.28 (0.15–0.51) <0.001† Seizures 27 (25.0%) 79 (54.9%) 0.27 (0.16–0.47) <0.001† Malaise or fatigue 39 (36.1%) 101 (70.1%) 0.24 (0.14–0.41) <0.001† Headache 29 (26.9%) 90 (62.5%) 0.22 (0.13–0.38) <0.001†

Table 2. Signs and Symptoms Among Cases Testing Positive and Negative for Rabies

reporting a definite animal exposure was a limiting factor in this analysis.

The onset of illness of rabies cases was more likely to occur during summer or fall months (May through October) when compared to non-rabies cases (OR 1.77; 95% CI 1.04–3.01). While no significant difference was observed between indigenous rabies cases and nonrabies cases in the seasonal pattern of exposures to animals, the limited number of cases

The presenting signs and symptoms of human rabies in the United States were often nonspecific such as fever, malaise, headache, weakness, fatigue, sore throat, and anorexia. The most commonly reported signs and symptoms reported among rabies cases during the course of illness were fever (83%), confusion or delirium (62%), agitation or combativeness (51%), paresthesia or localized pain (50%), and dysphagia (49%). In contrast, the most common signs and symptoms reported among non-rabies cases were confusion or delirium (85%), fever (78%), malaise or fatigue (70%), headache (63%), and agitation or combativeness (60%). When comparing these two groups, aerophobia, hydrophobia, and paresthesia or localized pain were more likely to be reported among rabies cases than non-rabies cases with an OR of 14.59, 7.50, and 5.86 respectively (P-values ≤ 0.001). Dysphagia and localized weakness were also more likely to occur among rabies cases with ORs of 2.73 and 2.26 (Pvalues ≤ 0.004). Priapism or spontaneous ejaculation was reported more commonly among rabies cases (OR 2.73) but this finding did not reach statistical significance. Among nonrabies cases, headache, malaise or fatigue, seizures, and confusion or delirium were more frequent than in rabies cases with ORs of 0.22–0.28 (P-values <0.001). Although not statistically significant, insomnia was also seen more often in non-rabies cases. Fever, muscle spasm, hypersalivation, anxiety, hallucinations, autonomic instability, agitation or

Hypersalivation 28 (25.9%) 38 (26.4%) 0.98 (0.55–1.72) 1 Anxiety 33 (30.6%) 51 (35.4%) 0.80 (0.47–1.37) 0.50 Hallucinations 26 (24.1%) 42 (19.2%) 0.77 (0.44–1.36) 0.39

**Negative, N=144** 

**Odds Ratio** 

**(95% CI\*) P-value** 

**Variable Positive, N=108** 

Autonomic

Agitation or

Confusion or

**3.6.2 Seasonality** 

**3.6.3 Signs and symptoms** 

**n (%)** 

\* CI = Confidence interval, † Statistically significant

received four doses of an experimental rabies vaccine 13 years prior to illness onset, however subsequent serologic testing failed to detect antibodies in serum; patient 29 received pre-exposure prophylaxis and regular booster doses with duck embryo vaccine and had a positive RABV antibody titer 6 months before exposure. Patient 22 died of his illness, while patient 29 survived.

### **3.5 Clinical course**

Excluding cases acquired through laboratory exposures and tissue and organ transplantation, specific dates of definite and probable exposures reported for 28 cases were used to calculate a median incubation period of 41.5 days (range 8–701 days). Fifty patients had sought healthcare prior to admission with a median of one day (range 0–6 days) between illness onset and presentation for medical evaluation. Hospital admission dates were reported in 94 cases with a median of 4 days (range 0–16 days) from the onset of illness to hospital admission. Median time from onset of illness to admission to an intensive care unit (ICU) was also 4 days (range 1–13 days) in 14 cases reporting ICU admission date. Similarly, in 65 cases reporting the date of intubation the median time from onset of illness to intubation was 5 days (range 1–19 days). Time from onset of illness to development of fever was reported in 30 cases giving a median of 2 days (range 0–11 days); time from onset of illness to development of coma was reported in 33 cases giving a median of 7 days (range 1–28 days). The median length of illness (defined as days from illness onset until death) was 13.5 days (range 1–133 days).

#### **3.6 Comparison of rabies cases and non-rabies cases**

#### **3.6.1 Demographics**

Data available from 108 confirmed human rabies cases and 144 encephalitis cases where rabies was ruled out by laboratory diagnostic testing were compared (Table 2). Rabies cases were older on average than non-rabies cases with a mean of 34.4 years (range 2–82 years) versus 30.7 years (range <1–78 years). However, this result was not statistically significant. Male gender was nearly 2 times more likely among rabies cases than non-rabies cases (Pvalue 0.019).


received four doses of an experimental rabies vaccine 13 years prior to illness onset, however subsequent serologic testing failed to detect antibodies in serum; patient 29 received pre-exposure prophylaxis and regular booster doses with duck embryo vaccine and had a positive RABV antibody titer 6 months before exposure. Patient 22 died of his

Excluding cases acquired through laboratory exposures and tissue and organ transplantation, specific dates of definite and probable exposures reported for 28 cases were used to calculate a median incubation period of 41.5 days (range 8–701 days). Fifty patients had sought healthcare prior to admission with a median of one day (range 0–6 days) between illness onset and presentation for medical evaluation. Hospital admission dates were reported in 94 cases with a median of 4 days (range 0–16 days) from the onset of illness to hospital admission. Median time from onset of illness to admission to an intensive care unit (ICU) was also 4 days (range 1–13 days) in 14 cases reporting ICU admission date. Similarly, in 65 cases reporting the date of intubation the median time from onset of illness to intubation was 5 days (range 1–19 days). Time from onset of illness to development of fever was reported in 30 cases giving a median of 2 days (range 0–11 days); time from onset of illness to development of coma was reported in 33 cases giving a median of 7 days (range 1–28 days). The median length of illness (defined as days from illness onset until death) was

Data available from 108 confirmed human rabies cases and 144 encephalitis cases where rabies was ruled out by laboratory diagnostic testing were compared (Table 2). Rabies cases were older on average than non-rabies cases with a mean of 34.4 years (range 2–82 years) versus 30.7 years (range <1–78 years). However, this result was not statistically significant. Male gender was nearly 2 times more likely among rabies cases than non-rabies cases (P-

**Negative, N=144** 

4 (3.7%) 2 (1.4%) 2.73 (0.50–15.19) 0.41

**Odds Ratio** 

**(95% CI\*) P-value** 

**n (%)** 

Mean age (range) 34.35 (2–82) 30.72 (<1–78) 0.173 Aerophobia 10 (9.3%) 1 (0.7%) 14.59 (1.84–115.83) 0.001† Hydrophobia 36 (33.3%) 9 (6.3%) 7.50 (3.42–16.43) <0.001†

localized pain 54 (50.0%) 21 (14.6%) 5.86 (3.22–10.64) <0.001†

Dysphagia 53 (49.1%) 43 (29.9%) 2.26 (1.35–3.80) 0.003† Localized weakness 44 (40.7%) 34 (23.6%) 2.22 (1.29–3.83) 0.004† Male gender 83 (76.9%) 87 (62.6%) 1.98 (1.13–3.49) 0.019† Fever 90 (83.3%) 113 (78.5%) 1.37 (0.72–2.61) 0.422 Muscle spasm 45 (41.7%) 59 (41.0%) 1.03 (0.62–1.71) 1

illness, while patient 29 survived.

13.5 days (range 1–133 days).

**Variable Positive, N=108** 

**n (%)** 

**3.6.1 Demographics** 

value 0.019).

Paresthesia or

Priapism or spontaneous ejaculation

**3.6 Comparison of rabies cases and non-rabies cases** 

**3.5 Clinical course** 


Table 2. Signs and Symptoms Among Cases Testing Positive and Negative for Rabies

### **3.6.2 Seasonality**

The onset of illness of rabies cases was more likely to occur during summer or fall months (May through October) when compared to non-rabies cases (OR 1.77; 95% CI 1.04–3.01). While no significant difference was observed between indigenous rabies cases and nonrabies cases in the seasonal pattern of exposures to animals, the limited number of cases reporting a definite animal exposure was a limiting factor in this analysis.

#### **3.6.3 Signs and symptoms**

The presenting signs and symptoms of human rabies in the United States were often nonspecific such as fever, malaise, headache, weakness, fatigue, sore throat, and anorexia. The most commonly reported signs and symptoms reported among rabies cases during the course of illness were fever (83%), confusion or delirium (62%), agitation or combativeness (51%), paresthesia or localized pain (50%), and dysphagia (49%). In contrast, the most common signs and symptoms reported among non-rabies cases were confusion or delirium (85%), fever (78%), malaise or fatigue (70%), headache (63%), and agitation or combativeness (60%). When comparing these two groups, aerophobia, hydrophobia, and paresthesia or localized pain were more likely to be reported among rabies cases than non-rabies cases with an OR of 14.59, 7.50, and 5.86 respectively (P-values ≤ 0.001). Dysphagia and localized weakness were also more likely to occur among rabies cases with ORs of 2.73 and 2.26 (Pvalues ≤ 0.004). Priapism or spontaneous ejaculation was reported more commonly among rabies cases (OR 2.73) but this finding did not reach statistical significance. Among nonrabies cases, headache, malaise or fatigue, seizures, and confusion or delirium were more frequent than in rabies cases with ORs of 0.22–0.28 (P-values <0.001). Although not statistically significant, insomnia was also seen more often in non-rabies cases. Fever, muscle spasm, hypersalivation, anxiety, hallucinations, autonomic instability, agitation or

Human Rabies Epidemiology and Diagnosis 259

2007). The last survivor in this series (patient 103) showed clinical signs and symptoms of encephalitis 2 months after exposure to bats and RABV antibodies were detected in both serum and CSF prior to the administration of a single dose of rabies vaccine and rabies immune globulin. This patient made a full recovery without the need for intensive care. No

Excluding the surviving cases, a total of 31 cases reported treatment after the onset of symptoms consisting of either administration of rabies immune globulin, immunization with rabies vaccine, induction of coma via the Milwaukee Protocol, or receipt of one or more antiviral medications including acyclovir, ganciclovir, amantadine, ribavirin, interferon, cytarabine, or adenine arabinoside. Thirteen patients were administered rabies immune globulin, 8 patients were immunized with rabies vaccine, 23 patients received antiviral therapy, and 5 patients underwent treatment with the Milwaukee Protocol during the course of illness. The median length of illness among all of those treated was longer when compared to those who did not receive any of the specified treatments (23.0 days vs. 13.0 days, P-value 0.029 by Log-rank test). It may be that those who received treatment were also more likely to receive supportive care which has previously been seen to increase the length of illness (Anderson, 1984). When treatments were compared individually, the median length of illness was longer for patients receiving rabies vaccine (14.0 day versus 13.5 days), the Milwaukee Protocol (28.0 days versus 13.0 days), or antiviral therapy (17.0 day versus 13.0 days) but shorter for patients receiving rabies immune globulin (13.0 days versus 14.0 days). However, none of the differences for individual treatments reached statistical significance (P-value 0.198-0.537 by Log-rank test). Overall, the median length of illness was

Data on administration of PEP was available in 71 cases (Table 1). A minimum of 3,359 individuals received PEP due to contact with a human rabies patient with a median of 39 individuals per case (range 1–180). The average number of individuals receiving PEP was greater in cases diagnosed postmortem compared to cases diagnosed antemortem (51 vs. 54) though this result was not statistically significant (P-value 0.453). The number of total risk assessments performed was reported for 22 cases. Typically 33% of individuals evaluated

Results of RABV laboratory diagnostic testing were collected and summarized for 62 patients where specimens were collected prior to death (Table 3). When available, the number of days after illness onset the sample was collected (i.e. the sample collection date – illness onset date) is reported in parentheses following the test result. The cutaneous nerves of nuchal skin biopsies were tested for RABV antigen in 41 cases. A positive result was obtained in 24 cases (59%) with the first positive sample occurring a median of 7 days (range 2-12 days) after illness onset. RABV was detected in corneal impressions in 9 of 19 cases (47%) with 3 (16%) reporting inconclusive results and 7 (37%) reporting negative results. RABV antigen was first detected in corneal impressions a median of 11 days (range 2-13 days) after onset of rabies. Samples of brain were tested in 6 cases; 1 was inconclusive, 1 was negative, and 4 were first found positive a median of 12 days (range 7-21 days)

RABV was identified from any of the surviving patients.

**3.8 Postexposure prophylaxis of contacts of cases** 

for RABV exposure received PEP per case (range 3–69%).

13.5 days (range 1–133 days).

**3.9 Rabies virus diagnostic testing 3.9.1 Antemortem test results** 

combativeness, nausea or vomiting, and ataxia all appeared to occur with equal likelihood in rabies and non-rabies cases.

#### **3.6.4 Laboratory values**

Laboratory values reported for the first collected samples of serum and CSF from rabies and non-rabies cases were compared. The serum white blood cells of rabies cases were elevated higher than non-rabies cases with mean values of 14.8 x 103 cells/µL (range 7.0–46.6 x 103 cells/µL) and 12.1 x 103 cells/µL (range 2.9–29.4 x 103 cells/µL) for rabies and non-rabies cases respectively (P-value 0.009). Over 95% of both rabies and non-rabies cases reporting CSF values had an abnormal CSF white blood cell count, red blood cell count, protein, or glucose. The percentage of abnormal values for each of these tests was similar between rabies and non-rabies cases. The white blood cell count in CSF was elevated in both rabies cases and non-rabies cases with mean values of 61.3 cells/µL (range 0–1000 cells/µL) and 89.0 cells/µL (range 0–980 cells/µL) respectively though this difference was not statistically significant (P-value 0.213). A lymphocytic predominance in CSF was seen in the majority of rabies cases. Rabies cases also demonstrated higher percentages of lymphocytes with a mean of 79% lymphocytes (range 31–100%) compared to a mean of 66% lymphocytes (range 0–100%) in non-rabies cases (P-value 0.034). Segmented neutrophils in CSF were also found to be higher in rabies cases with a mean of 30% (range 1–99%) compared to a mean of 15% (range 0–97%) in non-rabies cases (P-value 0.013). Protein in CSF was elevated in both groups with mean values of 72.8 mg/dL (range 15.0–178.0 mg/dL) in rabies cases and 86.2 mg/dL (range 4.0–1140.0) in non-rabies cases. This difference was not statistically significant (P-value 0.460). The mean value of glucose in CSF was elevated in rabies cases (85.8 mg/dL, range 24.0–211.0 mg/dL) while the mean of non-rabies cases was within the normal range (73.2 mg/dL, range 14.0–157.0 mg/dL) (P-value 0.004). No significant differences between rabies and non-rabies cases were found in CSF red blood cell counts, serum lymphocyte counts, serum segmented neutrophil counts, or serum glucose (P-values 0.193-0.781).

#### **3.7 Treatment**

Treatment of human rabies was successful in 4 cases. Of these, 2 had received rabies vaccine before the onset of illness. The first survivor (patient 18) became ill 2 days after completing a 14-day course of duck embryo vaccine following a bat bite. The diagnosis of rabies was based on CSF RABV antibody detection and serum antibody titers most consistent with clinical infection in combination with compatible epidemiologic and clinical histories. This patient required intensive supportive care but recovered fully within 6 months after onset. The second survivor (patient 29) had onset of symptoms ~2 weeks after spraying suspensions of a modified live RABV strain while performing research as a laboratory technician. Diagnosis was confirmed by detection of CSF antibodies and rising RABV serum antibody titers. The patient had received pre-exposure prophylaxis with annual boosters and had demonstrated RABV antibodies in response to these vaccinations. It is hypothesized that the strain he was exposed to may have developed increased infectivity following passage through animal and cell culture systems (Gibbons, 2002). Following intensive medical care, the patient survived but was left with severe neurologic sequelae. The third survivor (patient 94) developed rabies after being bitten by a bat. RABV antibodies were found in both serum and CSF. She was treated with an experimental treatment protocol (later termed the Milwaukee Protocol) involving induction of coma and administration of antiviral agents and recovered with only minor residual deficits (Hu,

combativeness, nausea or vomiting, and ataxia all appeared to occur with equal likelihood

Laboratory values reported for the first collected samples of serum and CSF from rabies and non-rabies cases were compared. The serum white blood cells of rabies cases were elevated higher than non-rabies cases with mean values of 14.8 x 103 cells/µL (range 7.0–46.6 x 103 cells/µL) and 12.1 x 103 cells/µL (range 2.9–29.4 x 103 cells/µL) for rabies and non-rabies cases respectively (P-value 0.009). Over 95% of both rabies and non-rabies cases reporting CSF values had an abnormal CSF white blood cell count, red blood cell count, protein, or glucose. The percentage of abnormal values for each of these tests was similar between rabies and non-rabies cases. The white blood cell count in CSF was elevated in both rabies cases and non-rabies cases with mean values of 61.3 cells/µL (range 0–1000 cells/µL) and 89.0 cells/µL (range 0–980 cells/µL) respectively though this difference was not statistically significant (P-value 0.213). A lymphocytic predominance in CSF was seen in the majority of rabies cases. Rabies cases also demonstrated higher percentages of lymphocytes with a mean of 79% lymphocytes (range 31–100%) compared to a mean of 66% lymphocytes (range 0–100%) in non-rabies cases (P-value 0.034). Segmented neutrophils in CSF were also found to be higher in rabies cases with a mean of 30% (range 1–99%) compared to a mean of 15% (range 0–97%) in non-rabies cases (P-value 0.013). Protein in CSF was elevated in both groups with mean values of 72.8 mg/dL (range 15.0–178.0 mg/dL) in rabies cases and 86.2 mg/dL (range 4.0–1140.0) in non-rabies cases. This difference was not statistically significant (P-value 0.460). The mean value of glucose in CSF was elevated in rabies cases (85.8 mg/dL, range 24.0–211.0 mg/dL) while the mean of non-rabies cases was within the normal range (73.2 mg/dL, range 14.0–157.0 mg/dL) (P-value 0.004). No significant differences between rabies and non-rabies cases were found in CSF red blood cell counts, serum lymphocyte

counts, serum segmented neutrophil counts, or serum glucose (P-values 0.193-0.781).

Treatment of human rabies was successful in 4 cases. Of these, 2 had received rabies vaccine before the onset of illness. The first survivor (patient 18) became ill 2 days after completing a 14-day course of duck embryo vaccine following a bat bite. The diagnosis of rabies was based on CSF RABV antibody detection and serum antibody titers most consistent with clinical infection in combination with compatible epidemiologic and clinical histories. This patient required intensive supportive care but recovered fully within 6 months after onset. The second survivor (patient 29) had onset of symptoms ~2 weeks after spraying suspensions of a modified live RABV strain while performing research as a laboratory technician. Diagnosis was confirmed by detection of CSF antibodies and rising RABV serum antibody titers. The patient had received pre-exposure prophylaxis with annual boosters and had demonstrated RABV antibodies in response to these vaccinations. It is hypothesized that the strain he was exposed to may have developed increased infectivity following passage through animal and cell culture systems (Gibbons, 2002). Following intensive medical care, the patient survived but was left with severe neurologic sequelae. The third survivor (patient 94) developed rabies after being bitten by a bat. RABV antibodies were found in both serum and CSF. She was treated with an experimental treatment protocol (later termed the Milwaukee Protocol) involving induction of coma and administration of antiviral agents and recovered with only minor residual deficits (Hu,

in rabies and non-rabies cases.

**3.6.4 Laboratory values** 

**3.7 Treatment** 

2007). The last survivor in this series (patient 103) showed clinical signs and symptoms of encephalitis 2 months after exposure to bats and RABV antibodies were detected in both serum and CSF prior to the administration of a single dose of rabies vaccine and rabies immune globulin. This patient made a full recovery without the need for intensive care. No RABV was identified from any of the surviving patients.

Excluding the surviving cases, a total of 31 cases reported treatment after the onset of symptoms consisting of either administration of rabies immune globulin, immunization with rabies vaccine, induction of coma via the Milwaukee Protocol, or receipt of one or more antiviral medications including acyclovir, ganciclovir, amantadine, ribavirin, interferon, cytarabine, or adenine arabinoside. Thirteen patients were administered rabies immune globulin, 8 patients were immunized with rabies vaccine, 23 patients received antiviral therapy, and 5 patients underwent treatment with the Milwaukee Protocol during the course of illness. The median length of illness among all of those treated was longer when compared to those who did not receive any of the specified treatments (23.0 days vs. 13.0 days, P-value 0.029 by Log-rank test). It may be that those who received treatment were also more likely to receive supportive care which has previously been seen to increase the length of illness (Anderson, 1984). When treatments were compared individually, the median length of illness was longer for patients receiving rabies vaccine (14.0 day versus 13.5 days), the Milwaukee Protocol (28.0 days versus 13.0 days), or antiviral therapy (17.0 day versus 13.0 days) but shorter for patients receiving rabies immune globulin (13.0 days versus 14.0 days). However, none of the differences for individual treatments reached statistical significance (P-value 0.198-0.537 by Log-rank test). Overall, the median length of illness was 13.5 days (range 1–133 days).

#### **3.8 Postexposure prophylaxis of contacts of cases**

Data on administration of PEP was available in 71 cases (Table 1). A minimum of 3,359 individuals received PEP due to contact with a human rabies patient with a median of 39 individuals per case (range 1–180). The average number of individuals receiving PEP was greater in cases diagnosed postmortem compared to cases diagnosed antemortem (51 vs. 54) though this result was not statistically significant (P-value 0.453). The number of total risk assessments performed was reported for 22 cases. Typically 33% of individuals evaluated for RABV exposure received PEP per case (range 3–69%).
