No effect of oral polio vaccine administered at birth on mortality and immune response to BCG. A natural experiment.

BACKGROUND

WHO recommends oral polio vaccine at birth (OPV0) in polio endemic countries. During a period without OPV in Guinea-Bissau in 2004, we observed that not receiving OPV0 was associated with significantly decreased mortality in boys and better immune response to BCG vaccination. In 2007, whilst conducting a trial of BCG and vitamin A supplementation (VAS) at birth to low birthweight (LBW) children, OPV was again lacking for a short period. We used this natural experiment to test the previous observations.

METHODS

In the trial LBW infants were randomised to early or delayed BCG and VAS or placebo at birth. We noted whether the children received OPV0 or not. We compared children who received No OPV0 with those who received OPV0 in the 2 months before and the 2 months after the period without OPV. Mortality was compared in Cox regression models providing adjusted hazard ratios (aHR); the immune response to BCG was assessed in Poisson models providing adjusted prevalence ratios (aPR).

RESULTS

Ninety-nine children received No OPV0 and were compared with 243 children who received OPV0. No OPV0 was associated with insignificantly higher mortality during the first year of life, the aHR being 1.83 (95% CI: 0.93-3.61). The effect was similar in boys and girls. Overall, there was no significant association between No OPV0 and having a positive PPD response (aPR=1.33 (0.64-2.78)) or a scar (aPR=1.02 (0.93-1.11)) after BCG vaccination, though No OPV0 boys were more likely to develop a scar (aPR: 1.10 (1.01-1.20)).

CONCLUSIONS

The findings did not support our previous observation that not receiving OPV0 was associated with reduced mortality in boys. The findings weakly supported that OPV0 leads to a dampened response to simultaneously administered BCG vaccine in boys.