Effectiveness and safety of pneumococcal vaccines used alone or combined with influenza vaccination in dialysis patients: A systematic review and meta-analysis

Background: A lower conversion vaccination rate and a more rapid decline in antibody titers over time in dialysis patients raise concerns about the effectiveness of pneumococcal vaccination (PV) in this population, which has not been systematically reviewed. Methods: We searched PubMed, Cochrane Library, Embase and three Chinese databases from inception until February 29th, 2020 for interventional, cohort and case–control studies evaluating PV alone or combined with influenza vaccination (IV) on outcomes (all-cause mortality, pneumonia, cardiovascular events, antibody response and safety). Independent reviewers completed citation screening, data extraction, risk assessment, meta-analysis, and GRADE rating of the quality of evidence. Results: Five cohort studies and one quasirandomized control trial enrolling 394,299 dialysis patients with high to moderate quality were included. Compared with unvaccinated individuals, those receiving PV had lower risk of all-cause mortality [Adjusted relative risk (RR) 0.73, 95% CI 0.67–0.79, I2 = 31.1%, GRADE low certainty] and cardiovascular events (adjusted RR 0.80, 95% CI 0.69–0.93, I2 = 47.2%, GRADE low certainty) without serious adverse effect reported. Compared with no vaccination, lower all-cause mortality was observed in those receiving PV combined with IV (Adjusted RR 0.71, 95%CI 0.67–0.75, I2 = 63.3%), PV alone (Adjusted RR 0.86, 95% CI 0.78–0.94,I2 = 0%], and IV alone (Adjusted RR 0.76, 95% CI 0.73–0.79, I2 = 0%]. There was no difference between pneumococcal vaccinated patients vs non-vaccinated patients with respect to pneumonia. Immune response to pneumococcal conjugate vaccine-13 was weaker in polysaccharide pneumococcal vaccine-23-pre-vaccinated compared with vaccine-naive patients. Conclusions: The use of pneumococcal vaccine especially combined with influenza vaccination is associated with lower risks of all-cause mortality but may be affected by residual confounding/healthy vaccinee bias.