Session: Poster Session: Vaccines: Adults II
Monday, October 27, 2008: 12:00 AM
Room: Hall C
Background: Limited and conflicting data exist on the association of pneumococcal serotypes with mortality from invasive pneumococcal disease (IPD). Methods: Nation-wide population-based cohort study based on the linkage of IPD-laboratory surveillance data between 1977 and 2007 to several Danish national data sources. Isolates were serotyped by Quellung reaction. Associations between pneumococcal serotypes and mortality within 30 days following hospitalization were examined in a multivariate logistic regression analysis, controlling for potential confounders. Serotype 1 was the reference group. We ranked serotypes according to their mortality-ORs multiplied by the frequency in the study population. Alternative vaccine formulations were suggested. Results: A total of 18.858 IPD patients were included (90% adults, 15% meningitis). The overall 30-day-mortality was 18% and only 3% in children <5 years. Increasing age, male sex, meningitis, comorbidity level, alcoholism, and early decade of diagnosis were significantly associated with mortality. IPD patients with pneumococcal serotypes 31, 11A, 35F, 17F, 3, 16F, 19F, 10A, and 15B had substantially and statistically significantly increased mortality after controlling for confounders. In children <5 years, associations between serotypes and 30-days-mortality were not statistically significant. We found that in our study population, alternative 7- and 10-valent vaccine formulations would target a significantly larger number of cases and deaths than the current vaccines (p< 0.001). No significant differences were observed for formulations containing ≥11 serotypes. Conclusions: Our results support that pneumococcal serotypes strongly and independently affect 30-day IPD-mortality. These findings may be important for vaccine development in the future.