Risk factors for systemic hypersensitivity reactions after booster vaccinations with human diploid cell rabies vaccine: A nationwide prospective study

Risk factors for systemic hypersensitivity reactions after booster vaccinations with human diploid cell rabies vaccine: A nationwide prospective study Daniel B. Fishbein *°°, Kristine M. Yenne *t, David W. Dreesen t, Carol F. Teplis *t, Neha Mehta* and Deborah J. Briggs ~ To determine the incidence of and risk factors for adverse reactions following the boosters, we conducted a nationwide prospective study of persons receiving pre-exposure booster vaccination with human diploid cell rabies vaccine ( HDCV). Persons who had previously received three pre-exposure doses of H D C V and whose rabies neutralizing antibody titres were <%1:5 were enrolled in the study if they stated that they intended to receive a booster. O f the 98 persons enrolled in the study, 40 (41%) were in risk groups for whom boosters are not recommended. Three (3%) of 98 developed generalized urticaria or wheezing within 1 day of receiving boosters and three others (3%) developed urticaria 6 to 14 days after the booster. No differences were found between individuals with reactions (either type) and those with no adverse reaction according to age, gender, occupation, history of previous allergies, or time since or route of primary vaccination. Reactions were somewhat more common among persons who received primary vaccinations by the intramuscular route ( i.m. ) and booster vaccinations by the intradermal route (i.d.) (3/15, 20%) or primary vaccinations i.d. and booster vaccinations i.m. (2/10, 20%), and somewhat less common among persons who received both these vaccinations i.d. (1/52, 2%) or i.m. (0/7). The number of persons who develop allergic reactions may be minimized by administering vaccinations only when vaccination is strictly indicated. The influence of the route of primary and booster vaccinations on the development of reactions deserves further study. Keywords:Human diploid cell rabies vaccine; systemic reaction; booster

Human diploid cell rabies vaccine (HDCV, Imovax R) was licensed for use in the United States in 1980, and shortly thereafter became the most widely used rabies * Viral and Rickettsial Zoonoses Branch, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases and International Branch, Division of Field Epidemiology, Epidemiology Program Office, Centers for Disease Control, Public Health Service, US Department of Health and Human Service, Atlanta, GA 30333, USA. t College of Veterinary Medicine, University of Georgia, Athens, GA 30602, USA. Department of Veterinary Diagnosis, College of Veterinary Medicine, Veterinary Medical Center, Kansas State University, Manhattan, KS 66506-5600, USA. °°To whom correspondence should be addressed at: International Branch, Division of Field Epidemiology, Epidemiology Program Office, Mailstop C-08, Centers for Disease Control, 1600 Clifton Road, Atlanta, GA 30333, USA. (Received 4 November 1992; revised 2 February 1993; accepted 2 February 1993) Use of trade names is for identification only and does not imply endorsement by the Public Health Service or the US Department of Health and Human Services 0264-410x/93/14/1390-05 © 1993 Butterworth-Heinemann Ltd

1390 Vaccine, Vol. 11, Issue 14, 1993

vaccine in the country 1. The vaccine, produced by Pasteur-M6rieux Serum et Vaccins in France, is manufactured by growing the Pitman Moore strain of rabies virus in the MRC5 human diploid cell line, concentrating the final product and then inactivating the virus with fl-propiolactone2. Compared with rabies vaccines available in the United States before 1980, HDCV has been shown to have excellent immunogenicity and efficacy, and a lower number of postvaccinal reactions. The only relatively common side-effect associated with HDCV is a hypersensitivity reaction that follows HDCV boosters and is characterized by generalized pruritus, hives and subcutaneous oedema 3. These reactions were first recognized in early studies of HDCV in the 1970s. Concern increased when reaction rates between 10 and 21% were noted in three cohorts that had received both primary and booster injections with HDCV 4'5. The reaction usually developed 8 to 11 (but sometimes as early as 3 and as late as 13 ) days after booster vaccination5'6. One unusual aspect of these reactions has been their apparent resemblance to

Systemic reactions after HDCV boosters: D.B. Fishbein et al.

delayed-type hypersensitivity. Although immediate-type reactions have been reported following primary vaccination with HDCV, they have been extremely rare in comparison with delayed-type reactions4. Most of the patients' clinical signs improved spontaneously within 2-3 days or when treated with antihistamines, but a few patients required epinephrine or corticosteroid therapy to reduce the discomfort of the reaction. The frequency of delayed-type hypersensitivity reactions prompted the Advisory Committee on Immunization Practices (ACIP) to discontinue recommending routine biennial HDCV boosters for persons who receive pre-exposure vaccination but have only infrequent exposures to the disease ('infrequent risk group' )7. The current ACIP guidelines recommend that persons at risk of frequent exposures to rabies obtain a virus-neutralizing antibody titre (VNA) every 2 years and receive boosters only when their titre falls to <1:58. However, this recommendation complicates the pre-exposure regimen, since it necessitates the inconvenience and expense of drawing blood for titre determination every two years. Several retrospective studies of the epidemiological features of these immune complex-like reactions have shown that the reaction develops in individuals after administration of both intradermal (i.d.) and intramuscular (i.m.) booster doses4'5'9. No significant differences were found in the prebooster VNA titres of persons who had a hypersensitivity reaction compared with those who did not, but two studies have suggested that persons who have reactions have higher VNA responses after booster 1°. In one study, the rate ratio was 3 to 1 (males:females); another study reported a similar gender difference 5'6. Apart from the aforementioned studies, little is known about factors that predispose an individual to HDCVmediated, systemic hypersensitivity reactions. Certain host- and/or vaccine-associated risk factors may not have been thoroughly studied because most vaccinees in previous studies were of similar age and occupation, and had received their primary vaccinations by the same dose and route, and their booster vaccinations by a single dose and route and with the same lot of vaccine. The purpose of this prospective study was to determine the incidence and possible risk factors associated with the hypersensitivity reactions observed in persons administered routine HDCV booster vaccinations. We attempted to select a diverse cohort to determine: (1) if particular host or vaccine characteristics predispose individuals to systemic hypersensitivity reactions and to determine (2) if the relatively high reaction incidence reported by previous investigations is representative of the general population or is due to some other confounding factor.

MATERIALS AND METHODS Subjects were selected from persons whose sera were submitted to the Department of Veterinary Diagnosis, College of Veterinary Medicine, Veterinary Medical Center, Kansas State University (KSU) for rabies VNA determination. The serological specimens were originally obtained or requested by physicians and health departments throughout the United States, and referred to KSU because it is one of the few commercial laboratories performing rabies VNA tests.

Between May 1990 and November 1993, the names of all persons whose serum specimens were tested at KSU and who were found to have rabies VNA < 1 : 5 were sent to the investigators at the Centers for Disease Control in Atlanta. Individuals with titres <1:5 were selected because this group is the only one for whom booster vaccinations are indicated according to the ACIP guidelines in effect at the time of the study 11. We subsequently attempted to contact these persons by telephone through their physicians or the original laboratory requesting the test. Participants who could be contacted were read a brief description of the study (explaining that the study was part of postmarketing surveillance); the investigators did not attempt to influence the participant's decision to receive a booster vaccination and did not mention hypersensitivity reactions at the time of enrolment.

Study population Individuals were included in the study if they had previously received a primary three-dose HDCV series (either i.m. or i.d.), had a rapid fluorescent focus inhibition test virus-neutralizing antibody (RFFIT) titre of < 1:5 (complete virus neutralization), planned to have an HDCV booster, and agreed to participate before (or within 2 days of) receiving their booster. Epidemiological and clinical information was obtained from each study participant over the telephone by using a standardized questionnaire. Information collected included gender, age and race, specific information concerning the individual's occupation, rabies exposure risk classification, rabies vaccination history, current medications and allergies. Participants were classified by rabies risk category using the categories developed by the ACIP a. Each participant was asked to note the vaccine name, lot number and route and dose of their booster. We attempted to contact participants twice, once within 7 days of administration of the booster and once at 21 to 28 days after administration of the booster, although all persons originally included in the study (as defined in the preceding paragraph) were retained if they were contacted by telephone once, at least 21 days after their vaccination. Participants were read a specific list of reactions and were asked whether they had experienced any of these reactions. During the final call, each individual was asked to assess the degree to which the vaccine reaction interfered with their daily lives, and whether they visited a physician as a result of a possible vaccine-associated reaction.

Case definition We defined an immune-mediated, systemic hypersensitivity reaction as a reaction in an individual who reported generalized urticaria or pruritic rash, angiooedema, wheezing and dyspnoea, or generalized arthralgia or arthritis within 2 weeks of an HDCV booster injection. Individuals who reported any of these signs were considered to have a 'systemic hypersensitivity reaction'; those who had none of these symptoms or signs were considered to have 'no systemic hypersensitivity reaction'. Persons reporting a systemic hypersensitivity reaction were further subdivided by onset of the reaction (i.e. immediate- or delayed-type) on the basis of the date that clinical signs initially developed4. As in previous studies, reactions developing in minutes to hours

Vaccine, Vol. 11, Issue 14, 1993 1391

Systemic reactions after HDCV boosters: D.B. Fishbein et al. Table I Epidemiological and clinical characteristics of groups, by type of systemic hypersensitivity reaction

Characteristic group (n=92) Age (years) Median Range Sex Male Female Occupation Clinical veterinary medicine Other occupations History of allergies Missed work/school following booster Saw physician following booster

No systemic reaction (n=89)

Any systemic reaction (n=6)

Immediate systemic reaction (n=3)

Delayed systemic reaction (n=3)

37.0 20-71

34.0 21-64

47.0 27--64

34.0 21-50

47 (53) a 45 (47)

1 (17)* 5 (84)

2 (33)" 2 (67)

0 (0) ° 3 (100)

35 (38)

3 (50)

1 (33)

2 (67)

57 (62) 49 (53) 3 (3)

3 (50) 4 (67) 3 (50) b

2 (67) 1 (33) 2 (67)*

1 (33) 3 (100) 3 (33)

1 (1)

4 (67)=

1 (33)

3 (100)C

"Number with characteristic in reaction group ( % ) bp < 0.01, Fisher's exact test °p < 0.001, Fisher's exact test

(maximum 24 hours) were defined as immediate reactions and those developing in 2 to 21 days were defined as delayed reaction. Statistical analysis of data Epidemiological and clinical data were analysed by using non-parametric univariate methods (Wilcoxon's rank sum test for continuous variables and Z2 or Fisher's exact test for discrete variables) to determine which variables were associated with immediate and delayed hypersensitivity reactions at a statistically significant level.

reactions were more likely to develop headaches, and delayed reactions were usually associated with generalized myalgia (Table 2). None of the six subjects who had reported a systemic hypersensitivity reaction was hospitalized. Four of the six sought medical attention, compared with one (1%) of those not experiencing a reaction (p < 0.01 ). Three (50%) missed work, compared with only three (3%) of those persons not experiencing a reaction (p < 0.01). Risk factors associated with reactions No significant differences were found among persons who did and did not report reactions according to age, gender, occupation, history of previous allergies, or time interval between primary and booster vaccination (Table 1 ). Reactions were somewhat more common among persons who received primary vaccinations (i.m.) and booster vaccinations (i.d.) (3/15, 20%) and those who received primary vaccinations i.d. and booster vaccinations i.m. (2/10, 20% ). None of the seven persons who received both their primary and booster vaccinations i.m., and only one (2%) of the 52 participants who received their primary and booster vaccinations i.d. developed a reaction (Table 3). Immediate reactions were somewhat more common (2/10, 20%) among persons who received their primary vaccination i.m. and their boosters i.d. (Table 4) and delayed reactions were most common among persons who received primary vaccination i.d. and their booster vaccinations i.m. (2/15, 13 % )

(Table 5 ). The lot number of vaccine that study participants received was obtained from 93 of the 98 study participants. The 93 participants received booster

Table 2 booster

Symptoms and signs reported by study participants following

RESULTS Characteristics of vaccinees Between May 1990 and December 1991,98 individuals from 25 states agreed to participate in the study and satisfactorily completed their postbooster interviews (Table 1 ). The mean age of the study group was 38.6 (range 20- 71 ) years; 50 (51% ) were female and 48 (49 % ) were male; 38 (39%) worked in clinical veterinary medicine. Most (58, 59%) were in the ACIP's 'frequent risk' group, 35 (36%) were in the 'infrequent risk' category and five (5%) were in the 'rare risk' group. Clinical features of reactions After their vaccination, six (6%) of the vaccinees reported hypersensitivity reactions. Three (3 % ) reported immediate reactions, all of which developed on the day the subjects received their booster vaccination. One of these three participants also developed angio-oedema, and another dyspnoea and wheezing. Three other (3%) study participants reported delayed reactions; all began 7 to 14 days after booster. Local reactions (redness, tenderness, swelling or itching at the injection site) were reported by approximately equal proportions of persons in all groups (data not shown). In addition to the systemic hypersensitivity symptoms that were part of the case definition, persons who reported immediate

1392

Vaccine, Vol. 11, Issue 14, 1993

Characteristic

No systemic reaction (n = 92)

Any systemic reactions (n = 6)

Symptoms/signs in case definition Urticaria or general- NA" 6 (100) b ized pruritic rash Wheezing and NA 1 (17) dyspnoea Angio-oedema NA 1 (17) Arthralgia or NA 1 (17) arthritis (diffuse) Other symptoms or signs Fever 0 0 Generalized pruritus 0 1 (17) (no rash) Generalized myalgia 3 (3) 3 (50) Dizziness 3 (3) 1 (17) Lymphadenopathy 2 (2) 1 (33) Headaches 4 (4) 3 (50)~ Abdominal pain 2 (2) 0 Nausea 5 (6) 1 (17) Vomiting 1 (1) 0 Diarrhoea 4 (4) 0

Immediate systemic reaction (n = 3)

Delayed systemic reaction (n = 3)

3 (100)

3 (100)

1 (33)

0

1 (33) 0

0 1 (33)

0 1 (33)

0 1 (33)

0 1 (33) 0 2 (67)c 0 1 (33) 0 0

3 (100)" 0 1 (33) 1 (33) 0 0 0 0

"NA: Not applicable since members of this group could not have this reaction according to case definition bNumber with characteristic (%); in some cases denominator (not shown) is less than the total number in group because information was unknown or not available ~p < 0.05 ~p < 0.01 ~p < 0.0001

Systemic reactions after HDCV boosters: D.B. Fishbein et al. Table 3 Proportion of participants reporting immediate or delayed systemic hypersensitivity reaction by route of primary and booster vaccination • Route of booster vaccination

i.m. i.d. Total

Route of primary vaccination

i.m.

i.d.

Total

0/7 (0) ~ 2/10 (20) 2/17 (12)

3/15 (20) 1/52 (2) 4/67 (6)

3/22(14) 3/62 (5) 6/84 (7)

"Excluding 16 participants for whom no information was available on route of primary vaccination

~Number with allergic reaction/number in group (percentage allergic reactions)

Table 4 Proportion of participants reporting immediate systemic hypersensitivity reaction by route of primary and booster vaccination° Route of booster vaccination

i.m.

i.d.

Total

i.m. i.d. Total

0/7 (0) b 2/10 (20) 2/17 (12)

1/15 (7) 0/52 (2) 1/67 (2)

1/22 (5) 2/62 (3) 3/84 (4)

Route of primary vaccination

"Excluding 16 participants for whom no information was available on route of primary vaccination bNumber with allergic reaction/number in group (percentage allergic reactions)

vaccinations from 28 different lots of vaccine. Although the large number of individual lots prevented a statistical comparison, reactions did not appear associated with any individual lot of vaccine. Three (50%) of those experiencing reactions were members of the infrequent risk category. DISCUSSION Immediate or delayed reactions were reported by 6 (6%) of the persons in this study. Half of these reactions occurred within 24 hours of vaccination and were classified as immediate. Previous epidemiological studies, almost all retrospective, reported that immediate reactions after HDCV boosters were rare, about 0.01% ( 1/ 10 000)4. The explanation for the mechanism of these immediate reactions is obscure, and the small numbers of immediate reactions prevent definitive epidemiological analysis of risk factors for this type of reaction, but immediate reactions do appear to be more common among recipients of HDCV boosters than previously reported. A number of other variables potentially related to the development of hypersensitivity reactions were analysed and none except route of immunization approached statistical significance. Hypersensitivity reactions were more common in persons whose routes of primary and booster immunization differed. Although no study participants who received both primary and booster vaccinations i.m. developed reactions, the size of this group (seven participants) was very small. If the rate of reactions among all persons who received at least one i.m. vaccination (5/32) is compared with that for people who received no i.m. vaccination (1/52), the differences are statistically significant. Another possible explanation that must be considered is that i.m. vaccination is more sensitizing than other routes of systemic vaccination ~2.

Finally, all these differences may represent chance variation. Previous investigators have speculated about the relation between route of primary injection and the occurrence of systemic hypersensitivity reactions. Cox and Schneider believed that the reaction was related to sensitization by i.d. immunization and suggested that this route should not be used for primary vaccination3, but later studies suggested that the reactions were also common among persons who received their primary vaccination i.m. 9. Reanalysis of four previously published studies suggests that reactions were somewhat more common (28/152, 18%) in persons who received both primary and booster immunizations i.m. than in those who received both immunizations i.d. (27/266, 10% )4,s,9,1o. Routes of primary and booster immunization were not mixed in any of these studies. No study has randomized persons to receive boosters i.m. or i.d. We did not conduct immunological studies of the vaccinees in this study. Previous reports of vaccinees who have developed a delayed-hypersensitivity reaction revealed a specific IgE response to non-viral component(s ) of the vaccine. This observation is surprising because the immunological mediator of these reactions, IgE, is usually associated with Coombs' type I (anaphylactoid) reactions, while IgG-antigen complexes cause Coombs' type III (delayed-type arthus) reactions. A leucocytoclastic vasculitis has' also been noted 6. The principal antigen implicated in the IgE-mediated response is a modified protein component of the vaccine, a fl-propiolactone-human serum albumin (BPL-HSA) complex formed during the preparation of the vaccine6'9'x3, but the relation of IgE to delayed-hypersensitivity reactions (which are usually IgG-mediated) remained obscure. A later investigation revealed a dual reaction in certain individuals; both IgG and IgE responses to viral antigens (including BPL-HSA and fetal calf serum) were noted, as were positive tests to vaccine and mock vaccine 6. Our results suggest that both types of reactions may be associated with booster vaccination with HDCV, and make IgE a more plausible immunological mediator. However, none of these assays can be used to predict who might be at increased risk of a reaction. Although none of the reactions reported by study participants was life-threatening and none of the subjects who had reactions was hospitalized, it would be desirable to prevent these reactions if possible. Our study revealed two ways that the frequency of adverse reactions may be decreased. First, boosters should be administered only to members of risk groups for whom they are indicated a4. Second, combining results of this study and previous studies, it seems possible that the use of the i.d. Table 5 Proportion of participants reporting delayed systemic hypersensitivity reaction by route of primary and booster vaccinationa Route of booster

Route of primary vaccination

vaccination

i.m.

i.d.

Total

i.m. i.d. Total

0/7 (0) ~ 0/10 (0) 0/17 (0)

2/15 (13) 1/52 (2) 3/67 (6)

2/22 (9) 1/62 (2) 3/84 (4)

"Excluding 16 participants for whom no information was available on route of primary vaccination

~Number with allergic reaction/number in group (percentage allergic reactions)

Vaccine, Vol. 11, Issue 14, 1993

1393

S y s t e m i c r e a c t i o n s after H D C V boosters: D.B. F i s h b e i n et al.

dose/route for both primary and booster vaccinations may result in lower rates of reaction. Although primary i.d. vaccination is somewhat less immunogenic zs, the peak antibody response after booster does not differ by route of primary immunization a6. Booster vaccination i.d. is also more economical; cost reduction is particularly pertinent since a number of studies suggest that pre-exposure vaccination is overused and is cost-effective only in risk groups with very frequent exposures to rabies 17A8.

6

7

8 9

ACKNOWLEDGEMENTS

10

The authors are grateful to Brian Woo-Ming, Carole Lapinski and Brenda Carter for assistance with data collection. K.M.Y. was supported by an educational grant from Connaught Laboratories Inc. to the University of Georgia. © US Government.

11 12 13

REFERENCES 1 Anon. Recommendation of the Immunization Practices Advisory Committee (ACIP). Supplementary statement on rabies vaccine and serologic testing. MMWR Morb. Mortal. Wkly Rep. 1981,30, 535-536 2 Wiktor, T.J., Plotkin, S.A. and Koprowski, H. Development and clinical trials of the new human rabies vaccine of tissue culture (human diploid cell) origin. Dev. Biol. Stand. 1978, 40, 3-9 3 Cox, J.H. and Schneider, LG. Prophylactic immunization of humans against rabies by intradermal inoculation of human diploid cell culture vaccine. J. Clin. Microbiol. 1976, 3, 98-101 4 Anon. Systemic allergic reactions following immunization with human diploid cell rabies vaccine. MMWR Morb. Mortal. Wkly Rep. 1984, 33, 185-187 5 Dreesen, D.W., Bernard, K.W., Parker, R.A., Deutsch, A.J. and Brown, J. Immune complex-like disease in 23 persons following a

1394

Vaccine, VoI. 11, Issue 14, 1993

14 15 16

17 18

booster dose of rabies human diploid cell vaccine. Vaccine 1986, 4, 45-49 Warrington, R.J., Martens, C.J., Rubin, M., Rutherford, W.J. and Aoki, F.Y. Immunologic studies in subjects with a serum sickness-like illness after immunization with human diploid cell rabies vaccine. J. Allergy Clin. Immunol. 1987, 79, 605-610 Anon. Rabies prevention--United States, 1984. Centers for Disease Control, Atlanta. Conn. Med. 1984, 48, 813-819 Centers for Disease Control. Rabies prevention~United States, 1991: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb. Mortal. Wkly Rep. 1991,40, 1-19 Swanson, M.C., Rosanoff, E., Gurwith, M., Deitch, M., Schnurrenberger, P. and Reed, C.E. IgE and IgG antibodies to beta-propiolactone and human serum albumin associated with urticarial reactions to rabies vaccine. J. Infect. Dis. 1987, 155, 909-913 Fishbein, D.B., Dreesen, D.W., Holmes, D.F. et al. Human diploid cell rabies vaccine purified by zonal centrifugation: a controlled study of antibody response and side effects following primary and booster pre-exposure immunization. Vaccine 1990, 7, 437-442 Anon. Rabies prevention~United States, 1984. MMWR Morb. Mortal. Wkly Rep. 1984, 33, 393-402, 407-408 Anderson, J.A. and Adkinson, N.F. Allergic responses to drugs and biologic agents. JAMA 1987, 258, 2891-2899 Anderson, M.C., Baer, H., Frazier, D.J. and Quinnan, G.V. The role of specific IgE and beta-propiolactone in reactions resulting from booster doses of human diploid cell rabies vaccine. J. Allergy Clin. Immunol. 1987, 80, 981-868 Anon. Rabies p r e v e n t i o ~ n i t e d States, 1991. Recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb. Mortal. Wkly Rep. 1991, 40, 1-19 Briggs, D.J. and Schwenke, J.R. Longevity of rabies antibody titre in recipients of human diploid cell rabies vaccine. Vaccine 1992, 10, 125-129 Turner, G.S., Nicholson, K.G., Tyrrell, D.A. and Aoki, F.Y. Evaluation of a human diploid cell strain rabies vaccine: final report of a three year study of pre-exposure immunization. J. Hyg. (London) 1982, 89, 101-110 Mann, J. and Burkhart, M.J. Rabies preexposure prophylaxis. West. J. Med. 1983, 139, 386-387 Bernard, K.W. and Fishbein, D.B. Pre-exposure rabies prophylaxis for travellers: are the benefits worth the cost? Vaccine 1991, 9, 833-836