Risk of cutaneous herpes zoster in patients with spondyloarthritis treated with conventional and biologic disease-modifying antirheumatic drugs

Objectives To investigate the risk of cutaneous herpes zoster (HZ) in spondyloarthritis (SpA) compared with that in rheumatoid arthritis (RA), and in disease-modifying antirheumatic drugs (DMARDs) used in SpA. Method A total of 727 patients with an expert diagnosis of SpA were identified retrospectively from four rheumatology centers in Hong Kong. Electronic medical records from 1995 to 2018 were reviewed for incidence of cutaneous HZ and demographic data including age, sex, comorbidities, smoking and drinking status. DMARDs used included sulphasalazine, methotrexate, leflunomide, steroids, etanercept, infliximab, adalimumab, golimumab, secukinumab and ustekinumab. Cox regression models were used to evaluate hazard ratios (HRs) of different DMARDs in patients with SpA. Propensity score was used for matching and comparison with 857 patients with RA. Results There were 23 cases of cutaneous HZ in patients with SpA and 59 cases in patients with RA. Among patients with SpA, 7 cases of cutaneous HZ may be attributed to sulfasalazine treatment, 7 to methotrexate, 2 to leflunomide, 2 to infliximab, 1 to etanercept, 2 to adalimumab, and 1 to secukinumab. Risks of cutaneous HZ were the same in SpA (stratified HR 0.97; 95% CI 0.58; 1.61; P = .89) and RA. Methotrexate (adjusted HR 3.47; 95% CI 1.25; 9.63; P = .02) and infliximab (adjusted HR 10.67; 95% CI 1.37; 82.88; P = .02) were found to be associated with HZ after adjustments for traditional risk factors. Conclusion Risk of cutaneous HZ in SpA was not lower than in RA. Methotrexate and infliximab were associated with cutaneous HZ in SpA.