Herpes zoster infections in systemic lupus erythematosus: Risk factors and outcome

Objective. To determine factors that influence the frequency and outcome of herpes zoster infection in patients with systemic lupus erythematosus (SLE). Methods. In this case-central retrospective study, patients with a history of zoster infection were identified from our computerized database of 348 patients with SLE. Medical records were reviewed to establish activity of SLE at the time of zoster, as well as complications of the zoster infection. Results. Fifty-five episodes of zoster occurred among 47 (13.5%) patients, at a rate of 16 episodes/1000 patient-years of followup. Dissemination occurred in 6 episodes (11%), and was more frequent during immunosuppressive therapy [odds ratio (OR) = 4.0]. Bacterial superinfection occurred in 5 (9%), resulting in one death from sepsis, and was increased among patients receiving prednisone ≥ 60 mg daily (OR = 4.1). Compared to those without zoster, patients with zoster were significantly more likely to have previously had serious disease manifestations including nephritis, thrombocytopenia or hemolytic anemia, and to have received treatment with cyclophosphamide (all p ≤ 0.05). However, 65% of zoster episodes occurred during mild or inactive SLE, when the majority of patients were receiving less than 20 mg prednisone daily and no immunosuppressive therapy. Conclusion. Herpes zoster infections occur at increased frequency among patients with SLE compared to the general population, and carry significant morbidity. Patients who have had severe manifestations of lupus are at greatest risk of zoster, though not necessarily at the time of disease flare or immunosuppressive therapy. If disease activity allows, a reduction in prednisone dosage may reduce the risk of bacterial superinfection during zoster episodes.