SHORT COMMUNICATION
INT J TUBERC LUNG DIS 7(4):399–402 © 2003 IUATLD
Rates of adverse reactions to first and second doses of BCG vaccination: results of a large community trial in Brazilian schoolchildren I. Dourado,* M. H. Rios,† S. M. M. Pereira,* S. S. Cunha,*‡ M. Y. Ichihara,*§ J. C. L. Goes,* L. C. Rodrigues,‡ A. L. Bierrenbach,‡ M. L. Barreto* * Instituto de Saúde Coletiva/Universidade Federal da Bahia, † Bahia State Department of Health, Salvador, Bahia, Brazil; ‡ Infectious Disease Epidemiology Unit, London School of Hygiene & Tropical Medicine, London, UK; § Coordenação Regional da Fundação Nacional de Saúde, Salvador, Bahia, Brazil SUMMARY OBJECTIVE:
To evaluate the incidence of adverse reactions to first and second bacille Calmette-Guérin (BCG) vaccination in schoolchildren. S E T T I N G A N D D E S I G N : Enhanced surveillance in a Brazilian trial. Suspected reactions were reported to a nurse who visited cases and completed a standard form. R E S U L T S : Among 71 341 schoolchildren studied, 33 reactions were reported. Of these, 25 fulfilled the criteria, resulting in a rate of one per 2854 vaccinations, with
no deaths or BCG-osis. Reactions to second doses were more common than to first BCG vaccinations, but this difference was not statistically significant. C O N C L U S I O N S : Adverse reactions to a second dose of BCG may be more frequent than reactions to a first dose, but they are still rare events. K E Y W O R D S : BCG vaccine; adverse reactions; BCGREVAC
BACILLE CALMETTE-GUÉRIN (BCG) vaccination has been used in many countries of the world as a neonatal vaccination against severe forms of tuberculosis.1,2 In Brazil, intradermal BCG is routinely given during the neonatal period. The strain of BCG vaccine used in Brazil has also been shown to provide protection against tuberculous meningitis and leprosy.3,4 In 1994, the Ministry of Health recommended routine application of a second dose of BCG in schoolchildren.5 Although revaccination is routine in several countries,2 a joint statement from the World Health Organization (WHO) Global Programmes on Tuberculosis and on Vaccines in 1995 opined that there was no scientific evidence for protection (nor for lack of protection), and therefore did not recommend repeat BCG schemes.6 As a result, the BCG-REVAC vaccine trial was implemented in Brazil in 1996 to assess the protective effect of BCG revaccination. Lotte et al., defined a standard classification of adverse reactions to BCG,7 and, after literature review, reported that rates varied substantially but were essentially rare. For example, the rates of local abscess in children under 20 years of age caused by vaccination varied from 1 in 800 000 vaccinations in Germany to
1 in 50 vaccinations in Benin, with a median of one event in 1300 vaccinations in the five studies reviewed. Variations in rates of adverse reactions have been attributed to variations in the application techniques, the BCG strain used, the quantity of vaccine given, and the methods of detection. A study of adverse reactions to the first BCG dose in Brazil reported a rate of one in 2304 applied doses.8 There are no published estimates of the frequency of adverse reactions to a second dose. This paper reports the frequency of adverse reactions associated with BCG vaccination in children from one site of the BCG-REVAC trial with and without a presumed history of neonatal BCG.
METHODS The study population consisted of the vaccinated arm of children residing in one of the BCG-REVAC trial sites in Salvador, Bahia, Brazil. The methods of the trial are described elsewhere.9 In summary, and of relevance to this paper, the BCG-REVAC is an ongoing randomised controlled trial with no placebo conducted in children aged 7–14 years enrolled in state schools. Nearly 90% of this population had
Correspondence to: Professor Inês Dourado, Instituto de Saúde Coletiva/Universidade Federal da Bahia, Rua Padre Feijó, 29, 4º andar, Canela, Salvador, Bahia, 40.110.170 Brazil. Tel: (557) 1-245-0544. Fax: (557) 1-237-5856. e-mail:
[email protected] Article submitted 12 November 2001. Final version accepted 24 October 2002.
400
The International Journal of Tuberculosis and Lung Disease
neonatal BCG, so the vaccination received constituted a second dose. Parents/guardians of children allocated to vaccination received an information pack and a form to sign if they wished to withdraw their children from the study. Trained nurses inspected the right upper arms of all children. The site of neonatal BCG vaccination was examined in order to classify BCG scar as present (one or more), absent or doubtful. Children were not vaccinated if they or their parents/guardians refused vaccination, or if they were determined to have two scars or unclear scar readings. Trained nurses applied 0.1 ml of BCG vaccine by intradermal injection in those without scar or those with only one scar. Vaccination was performed between September 1996 and September 1997. The BCG vaccine used in the trial (and in routine neonatal vaccination in Brazil) was produced at the Ataulpho de Paiva Foundation, a nongovernmental laboratory, and was composed of the Moreau-Rio de Janeiro substrain at a concentration of approximately 100 000 attenuated viable bovine bacilli per 0.1 ml. Vaccines were stored at the State Immunisation Programme facilities, and were kept in appropriate conditions throughout the fieldwork. The ethics committees at both the University Hospital (Universidade Federal da Bahia, Brazil) and the London School of Hygiene & Tropical Medicine approved the trial. Adverse reaction was ascertained by enhanced routine passive surveillance. Leaflets containing information about the aims and the methodology of the trial and describing potential adverse reactions were sent to local health facilities, school principals, teachers and parents. These leaflets requested that parents, teachers and health professionals report any suspected adverse reactions to the project nurse by phone. The project nurse examined all students with reported adverse reactions; data were collected on a standardised form used by the National Immunisation Programme for surveillance of adverse reactions. Demographic, clinical and vaccination status data were obtained and linked to those from the trial database. When appropriate, students were referred to the state tuberculosis reference hospital. Cases of adverse reactions were defined according to two classifications: Lotte et al.’s and that of the Brazilian National Immunisation Program system;10 the latter is slightly more sensitive, as it includes nonsuppurating lymphadenitis. Local dissemination of the lesion was defined as those lesions disseminated to the deltoid area. Tuberculous-like cutaneous lesions were those around the BCG vaccine. Because this was an unblinded trial without baseline comparison for general manifestations such as fever or headache, which can occur among unvaccinated children, only vaccinated children were included in the follow-up. The risk of adverse reactions is reported according to presence or absence of a neonatal BCG scar, assumed to indicate
previous neonatal BCG vaccination.11 There was no bacteriological confirmation of adverse reactions.
RESULTS The REVAC-BCG databank contains information on 71 718 students who were vaccinated during the project. Of these, 174 (0.2%) were outside the cohort age range and 197 (0.3%) had no information on birth date. This analysis is thus based on 71 347 schoolchildren aged 7–14 years for whom information on presence of previous BCG scar was available. Of these, 11 981 (16.8%) had no BCG scar, 59 360 (83.2%) had one BCG scar and six (0.0%) had doubtful scar. Thirty-three students reported adverse reactions, of which 24 fulfilled both study criteria; one student (Case 14) had a non-suppurating lymphadenitis and thus fulfilled only the Brazilian National Immunisation Programme criteria (Table). Eight students were excluded: seven had non-specific reactions, and one had rubella. No deaths, permanent injuries or disseminated infections (BCG-osis) were notified. Information pertaining to the immunological status of the children was not collected; human immunodeficiency virus (HIV) prevalence in this age group is extremely low in Salvador, so HIV infection in the study population is unlikely. As indicated in the Table, most reactions (60%) were local cutaneous lesions (e.g., ulcers 1 cm), and were equally distributed among vaccinators and schools. Figure 1 shows a 12-year-old girl (Case 21) who presented an ulcer after receiving BCG. Seven students (28%) had axillary lymph node enlargement without suppuration. Other reactions were less frequent: local dissemination of lesion (four cases); suppurating lymphadenitis (two cases); tuberculosis-like cutaneous lesions (two cases) and dissemination of the lesion to the left arm and left leg (one case). Four children had acute cutaneous eruptions with no other cause, also known as Lotte’s ‘post-BCG syndrome’ (possible a hypersensitivity phenomena, Figure 2, Case 25). In another three cases, this symptom appeared in conjunction with non-suppurating lymphadenitis (Case 2), erythema (Case 18) and fever (Case 19). Five cases required hospitalisation for less than 48 hours, and none reported further complications. All events occurred within 1 to 70 days after vaccination. The median time to onset was 26 days. Considering the 25 adverse reactions, we estimated a total incidence of 35/100 000 (25/71 341), corresponding to one case in 2854 applied doses. Two adverse reactions occurred in students with no scar (1 in 5990, 95% confidence interval [CI] 1/1485– 1/34 482), and 23 occurred in those with one scar (1 in 2580 applied doses, 95%CI 1/1692–1/3984). The relative risk of adverse reactions in students with scar compared to those without scar was 2.3 (95%CI 0.69–7.80). The incidence was higher for females
Adverse BCG reactions
Table
401
Distribution of adverse reactions among children aged 7–14 years after BCG vaccine, Salvador, Brazil, 1996–1997
Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Previous scar
Main reaction/secondary reactions
Onset day for main reaction
Lotte’s criteria
PNI
1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Ulcer 1cm/headache Axillary lymphadenitis NS/acute cutaneous eruptions Ulcer 1cm/local dissemination of the lesion/axillary lymphadenitis NS Ulcer 1cm Ulcer 1cm Ulcer 1cm/axillary lymphadenitis/fever Ulcer 1cm Local dissemination of the lesion Ulcer 1cm/local hyperaemia Dissemination of the lesion to left arm and left leg Ulcer 1cm/local hyperaemia/axillary lymphadenitis NS Ulcer 1cm/cervical lymphadenitis/axillary lymphadenitis Ulcer 1cm/fever/headache Axillary lymphadenitis NS/fever Ulcer 1cm/oedema of the right arm/axillary lymphadenitis S Ulcer 1cm/cheloid at the BCG site/axillary lymphadenitis NS Ulcer 1cm/local dissemination of the erythema/cutaneous lesions Acute cutaneous eruptions/erythema Acute cutaneous eruptions/fever Ulcer 1cm Ulcer 1cm/ rash Acute cutaneous eruptions Local dissemination of the lesion Tuberculous-like cutaneous lesions Acute cutaneous eruptions
4 38 48 37 15 15 24 49 70 60 NA 31 8 36 6 NA 61 2 3 8 2 1 21 30 2
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
S suppurating; NS non-suppurating; NA not available; PNI Brazilian National Immunisation Programme.
(4.06 104) than for males (2.04 104); however, this difference was not statistically significant. There was no difference by age (RR 1; 95%CI 0.5–1.9) when comparing the 11–14 and 7–10 year age groups (data not shown).
and one event in 2580 vaccinations for students with a prior scar. The rate in students receiving their first dose is similar to that found previously in Brazil.8
DISCUSSION The rate of adverse reactions was one event in 5990 vaccinations for students receiving their first dose,
Figure 1
Case 21, a 12-year-old female, ulcer 1 cm (4.6 cm).
Figure 2 Case 25, an 11-year-old male, acute cutaneous eruptions.
402
The International Journal of Tuberculosis and Lung Disease
Because no control group was used, the misclassification of general symptoms such as fever and headache as adverse reactions was avoided by discarding all events not corresponding to Lotte’s classification (Table). The timing of events is as expected, although marginally shorter than previously described. de Souza et al. reported onset times varying from less than 2 weeks to 9 weeks for complications such as abscess, lymph node enlargement, ulcer and pedunculated lesions.8 In most of our study cases, the time spans were similar, except for one case of ulcer (Case 1) only 4 days after vaccination. For reactions such as acute cutaneous eruption, onset time was between 1 and 3 days. One large retrospective study that collected data through review of literature and inquiry by the International Union Against Tuberculosis documented that most acute cutaneous eruptions were reported less than 3 months after vaccination.7 In conclusion, the rate of adverse reactions associated with a second dose of BCG vaccination was found to be roughly twice the rate associated with a first dose, but this difference was not statistically significant. As a result, we conclude that adverse reactions to a second dose of BCG are rare, and that these adverse reactions do not represent an important hindrance to the policy of revaccination. Acknowledgements The authors are grateful to Dr Miguel Aiub Hijjar (then head of the National Tuberculosis Programme), Dr Maria de Lourdes de Sousa Maia (director of the National Immunisation Programme), the Bahia Health Secretariat, the BCG-REVAC trial team, and Dr Helena Sato, for suggestions on the manuscript. This research was supported by Fundação Nacional the Saúde (Brazilian Ministry of Health), and the Department for International Development (DFID,UK).
References 1 Fine P E M, Carneiro I A M, Milstein J B, Clemens C J. Issues relating to the use of BCG in immunizations programmes: A discussion document. Geneva: World Health Organization Department of Vaccine and Biologicals, 1999. 2 Trnka L, Dankova D, Zitova J, et al. Survey of BCG vaccination policy in Europe: 1994–1996. Bull World Health Organ 1998: 76: 85–91. 3 Rodrigues L C, Diwan V K, Wheeler J G. Protective effect of BCG against tuberculous meningitis and miliary tuberculosis: a meta-analysis. Int J Epidemiol 1993; 22: 1154–1158. 4 Lombardi C, Pedrazzani E S, Pedrazzani J C, Ferreira Filho P, Zicker F. The protective efficacy of BCG against leprosy in Sao Paulo, Brazil. Bol Oficina Sanit Panam 1995; 119: 415– 421. 5 Brasil, Ministério da Saúde. Segundo Informe Técnico Sobre a Vacinação/Revacinação BCG (Technical report on vaccination/ revaccination with BCG). Brasília: Ministério da Saúde, 1994. 6 World Health Organization. Global Tuberculosis Programme and Global Programme on vaccines. Weekly Epidemiol Rec 1995; 70: 229–236. 7 Lotte A, Wasz-Hockert O, Poisson N, Dumitrescu N, Verron M, Couvet E. BCG complications: estimates of the risk among vaccinated subjects and statistical analysis of their main characteristics. Adv Tuberc Res 1984; 21: 107–193 8 de Souza G R, Sant’Ana C C, Lapa e Silva J R, Mano D B, Bethlem N M. Intradermal BCG vaccination complications: analysis of 51 cases. Tubercle 1983; 64: 23–27. 9 Barreto M L, Rodrigues L C, Cunha S S, et al. Design of the Brazilian BCG-REVAC trial against tuberculosis: a large, simple randomized community trial to evaluate the impact on tuberculosis of BCG revaccination at school age. Control Clin Trials 2002; 23: 540–553. 10 Ministério da Saúde. Manual de vigilância epidemiológica de eventos adversos após vacinação (Guidelines on epidemiological surveillance of adverse events after vaccination). Brasília, Brazil: Ministério da Saúde, 1998: pp 11–16. Available on http://www.funasa.gov.br/pub/pdfs/manu_eadpv.pdf. 11 Pereira S M, Dourado I, Barreto M L, et al. Sensitivity and specificity of BCG scar reading in Brazil. Int J Tuberc Lung Dis 2001; 5: 1067–1070.
RÉSUMÉ O B J E C T I F : Etudier l’incidence de réactions défavorables lors d’une première et d’une seconde vaccination par le bacille de Calmette-Guérin chez les enfants des écoles. C O N T E X T E / S C H É M A : Surveillance accrue lors d’un essai au Brésil. Les réactions suspectes sont signalées à une infirmière qui rend visite aux cas et complète un formulaire standard. R É S U L T A T S : Sur 71.341 écoliers, on a signalé 33 réac-
tions, dont 25 répondaient aux critères, soit un taux de un pour 2.854 vaccinations, sans aucun décès ni aucune BCG-ite. Les réactions étaient plus courantes lors de la seconde que lors de la première vaccination, mais cette donnée n’est pas statistiquement significative. C O N C L U S I O N S : Les réactions défavorables à une seconde dose de BCG peuvent être plus fréquentes que lors d’une première dose, mais restent des incidents rares. RESUMEN
O B J E T I V O : Estudiar la incidencia de las reacciones adversas a la primera y segunda vacunación con el bacilo de Calmette y Guerin (BCG) en escolares. M A R C O D E R E F E R E N C I A Y D I S E Ñ O : Vigilancia estrecha en un estudio en Brasil. Las reacciones sospechosas fueron informadas a una enfermera que visitaba a los casos y completaba un formulario estándar. R E S U L T A D O S : Sobre un total de 71.341 escolares se infor-
maron 33 reacciones, de las cuales 25 respondían a los criterios, es decir una tasa de un por 2.854 vacunaciones, sin muertes ni BCGítis. Las reacciones fueron más frecuentes en la segunda vacunación BCG que en la primera, pero esta diferencia no fue estadísticamente significativa. C O N C L U S I Ó N : Las reacciones adversas a una segunda dosis de vacunación BCG pueden ser más frecuentes que a la primera, pero son de escasa frecuencia.
