gms | German Medical Science

The TOKEN study was deigned to assess a possible association between hexavalent vaccination (HV) and unexplained sudden unexpected death (uSUD) in children between their 2nd and 24th month of life.

Methods: For the epidemiological study part of the study, reports of sudden deaths were collected through 400 Local Health Authorities (LHAs) across Germany between July 2005 and July 2008. Additional case reports were received from the forensic institutes participating in the pathological study part. Cases classified as R95-99 (ICD-10) were included in the epidemiological analyses, after parental informed consent was given. Data on vaccination and other factors were obtained by parental and doctors' questionnaire. Temporal associations (relative risks, RR) of uSUD and HV were examined by self-controlled case series analyses (SCCS). Potentially causal associations were assessed by case-control analyses, using prospectively recruited controls, and multivariate odds ratios (OR) were estimated. Any P values <0.05 were considered statistically significant.

Results: Out of the 676 uSUD cases reported, 254 (37.6%) could be included with parental consent, whereas parental consent could not be obtained for 62.4%. The main study analysis showed no increased risk of sudden death within one week after HV (RR 0.59; 95%CI 0.26-1.33). The case-control analysis (aOR 0.53; 95%CI 0.20-1.37) was in accordance with this finding. The SCCS analysis revealed a RR of 1.54 (0.67-3.54) during the first three days after hexavalent vaccination, followed by a RR of 0.27 (0.06-1.12) on days 4-7. The multivariate case-control analysis (aOR 1.11; 0.36-3.43) neither suggested a risk increase within the first three days after vaccination. Explorative analyses suggested that the well-known risk factors for Sudden Infant Death Syndrome (SIDS) may play a major role: 90% of children who had died within 3 days after HV had been exposed to prone sleeping, maternal smoking, or overheating.

Discussion: Despite all efforts, participation of bereaved parents was low. Data protection concerns which mandated that the initial contact with parents was to be established indirectly by the LHAs proved to impede recruitment considerably. There was also strong evidence of a preferential self-selection of parents whose child had died shortly after HV. Due to this bias, any true association is likely to be over-estimated in this study. Thus, many of the study questions posed cannot be answered with certainty. Despite all study limitations, it is concluded that the risk of sudden unexplained death within one week after HV is not increased and that the established risk factors for SIDS should be strictly avoided.