The effect of Haemophilus influenzae type B and pneumococcal conjugate vaccines on childhood meningitis mortality: a systematic review

BACKGROUND Two of the most prevalent causes of severe bacterial meningitis in children, Haemophilus influenzae type B (Hib) and Streptococcus pneumoniae, are preventable by existing vaccines increasingly available in developing countries. Our objective was to estimate the dose-specific effect of Hib and pneumococcal conjugate vaccines (PCV) on childhood meningitis mortality in low-income countries for use in the Lives Saved Tool (LiST). METHODS We systematically searched and reviewed published vaccine efficacy trials and observational studies reporting the effect of Hib or PCV vaccines on organism-specific meningitis, bacterial meningitis and all-cause meningitis incidence and mortality among children less than five years old in low- and middle-income countries. Data collection and quality assessments were performed using standardized guidelines. For outcomes available across multiple studies (≥ 2) and approximating meningitis mortality, we pooled estimates reporting dose-specific effects using random effects meta-analytic methods, then combined these with meningitis etiology data to determine the preventable fraction of childhood meningitis mortality for inclusion in LiST. RESULTS We identified 18 studies of Hib conjugate vaccines reporting relevant meningitis morbidity and mortality outcomes (2 randomized controlled trials [RCTs], 16 observational studies) but few provided dose-specific effects. A meta-analysis of four case-control studies examined the dose-specific effect of Hib conjugate vaccines on Hib meningitis morbidity (1 dose: RR=0.64, 95% CI 0.38-1.06; 2 doses: RR=0.09, 95% CI 0.03-0.27; 3 doses: RR=0.06, 95% CI 0.02-0.22), consistent with results from single RCTs. Pooled estimates of two RCTs provided evidence for the effect of three doses of PCV on vaccine-serotype meningitis morbidity (RR=0.16, 95% CI 0.02-1.20). We considered these outcomes of severe disease as proxy estimates for meningitis mortality and combined the estimates of protective effects with meningitis etiology data to provide an estimate of the preventable fraction of childhood meningitis mortality with three doses of Hib (38-43%) and pneumococcal conjugate vaccines (28-35%) for use in LiST. CONCLUSIONS Few RCTs or vaccine effectiveness studies evaluated the dose-specific impact of Hib and PCV vaccines on childhood meningitis mortality, necessitating use of proxy measures to estimate population impact in LiST. Our analysis indicates that approximately three-quarters of meningitis deaths are preventable with existing Hib and PCV vaccines.