OP-26A - Evaluation of the Potential Confounding Effect of Shingles on the Risk of Guillain-barré Syndrome After Vaccination with the Recombinant Zoster Vaccine
Background: A ‘statistical signal’ for Guillain-Barré syndrome (GBS) after recombinant zoster vaccine (RZV) from the vaccine safety datalink rapid cycle analyses was presented in Feb 2019 Advisory Committee on Immunization Practices (ACIP) meeting. Confounding may generate such association if RZV vaccination tends to occur close to a shingles episode (SE) (which was suggested anecdotally) and if GBS may result from a recent SE (as observed by Kang et al). Our aim was to explore this possibility in existing real-world data.
Objectives: Measure the association between a SE and 1) GBS 2) RZV vaccination.
Methods: The associations were both investigated in two US claims databases (Truven Health Analytics’ Marketscan commercial ( < 65 years old) and Medicare supplemental (>= 65 years old)). Incident SE and GBS were identified as the first record of any relevant ICD9/10 codes. RZV vaccinations were detected with the specific CPT or NDC codes. The risk of GBS within 42 days after a SE was assessed using a self-controlled case series (SCCS) analysis. The evaluation of the association between a SE and the probability of RZV was based on a matched case-control analysis. Each subject vaccinated with RZV was matched with subjects on gender, age class and having a visit within the same 2-weeks calendar period as the RZV date (=index date) of the matched vaccinated subject but with no recorded RZV vaccination. Exposure was defined as a SE within 6 weeks before the index date.
Results: 106 GBS cases were included in the SCCS analysis of the 50 - 64 years old. The relative incidence (RI) and 95% confidence interval (CI) of GBS between the first 42 days after SE and the next 42-days period was 2.65 (1.69 - 4.17). 85 GBS cases with preceding SE were detected in >= 65 years old. The RI was 1.81 (1.12 - 2.92). The odds ratio (OR) (95%CI) of SE before RZV in subjects <65 was 3.18 (2.69 - 3.75) for the period -41 to -15 days before RZV, and 0.80 (0.67 - 0.95) for the period -14 to -1 days. In >= 65 years subjects, the ORs were 1.65 (1.28 - 2.12) and 0.75 (0.56 - 1.01) respectively.
Conclusions: Analyses suggested a two-fold increased risk of GBS within 42 days after a SE and an increased probability of RZV vaccination within 15 to 42 days after a SE. Despite limitations including the absence of medical confirmation of the events and no consideration of other covariates, the occurrence of these two associations confirmed that a confounding is possible in the association between RZV vaccination and the risk of GBS. The authors suggest that studies on this association should be designed to address this possibility.