The risk of pneumococcal infections after immunisation with pneumococcal conjugate vaccine compared to non-vaccinated inflammatory arthritis patients

Background/Purpose: The purpose was to examine the risk of putative pneumococcal infections between adult arthritis patients on different anti-rheumatic drugs immunised with heptavalent pneumococcal conjugate vaccine (Prevenar7; PCV7) and non-vaccinated individually matched arthritis patients.

Methods: A consecutive cohort of 505 patients with rheumatoid arthritis (RA) and spondylarthropathy (SpA) including psoriatic arthritis, regularly followed at the outpatient Department of Rheumatology Skåne University Hospital in Lund and Malmö, was immunised with a single dose of PCV7 (exposed group). Of these, 497 patients (RA=248; SpA=249) were included. At vaccination, RA patients were treated with methotrexate (MTX) (n=84), anti-TNF+MTX (n=87) or anti-TNF as monotherapy (n=77). SpA patients were treated with anti-TNF as monotherapy (n=81), anti-TNF+MTX (n=82) or NSAIDs/analgesics (n=86). For each vaccinated patient, we identified four reference subjects (n=1988) from the same geographic area (Skåne county, Sweden), individually matched for age, gender, diagnosis and date. These were considered unexposed to conjugate pneumococcal vaccination. The Skåne Health Care Register containing all data on inpatient and outpatient health care was searched for all individuals seeking health care for putative pneumococcal infections using ICD-10 diagnostic codes. All events occurring 4 years before vaccination and up to 4.5 years after vaccination were divided into serious infections; (pneumonia, other lower respiratory infections, meningitis, sepsis, septic arthritis) and non-serious infections; (upper respiratory infections, otitis media and sinusitis). Relative risks (RR) of infections were calculated as events after/events before vaccination. Ratios of relative risk (RRR) were calculated between exposed and unexposed groups of patients. Generalized Estimating Equation was used to handle correlated data for several events in the same individual, PROC GENMOD in the SAS System 9.3.

Results: Although non-significant, the point estimate of RRR suggested a reduced risk of serious infections in vaccinated patients compared to the unexposed group (Table). However, time to first event of all infections before and after vaccination did not differ between the groups. Mean time (CI 95%) to first serious infection was significantly longer for immunised patients compared to the unexposed group [48.5 months (47.9-49.0) and 47.0 (46.7-47.3)], respectively.

Conclusion: Vaccination with heptavalent pneumococcal conjugate vaccine tended to reduce the risk of putative serious pneumococcal infections with about 45% compared to unvaccinated patients in this observational cohort study.