Background and aims Herpes-zoster (HZ) has higher occurrence in systemic lupus erythematosus (SLE) compared to the general population. This study aimed to identify risk factors associated with HZ infections in SLE.

Methods We included patients from Lupus Database of University of Santo Tomas (UST) Hospital, Manila, Philippines, who were diagnosed with HZ infection. Controls included SLE patients without HZ matched for age, sex and disease duration. SLE disease activity, corticosteroid, immunosuppressives, and hydroxychloroquine were compared between groups.

Results In a review of 626 patient records, 65 SLE patients (61, 93.8% females) developed HZ, with incidence 10.4/1000 person-years. Mean age was 36.5 years+11 (range 19–60), mean SLE disease duration to HZ 6.1 years+3.3 (range 2–17). HZ lesions were localised in 63 (97%), disseminated in 2 (3%) patients. Four patients had >2 HZ episodes. All responded favourably to anti-viral therapy with minimal sequelae. Compared with controls of 130 SLE patients without HZ, cases were significantly more likely to have received high-dose prednisone 65/65 (OR 16.41, p=0.0066) with mean prednisone 18.5+12 mg/day and cyclophosphamide (Cyc) 19/65 (OR 7.05, p<0.0001). IV Cyc with mycophenolate mofetil (MMF) conferred greatest risk for HZ infection. There was no association of disease activity with HZ risk, whereas hydroxychloroquine was a negative risk factor for HZ infection (OR 0.26, p= 0.0005).

Conclusions Immunosuppressive agents and corticosteroids are risk factors associated with development of HZ in SLE. On the other hand, hydroxychloroquine appeared to have a protective role against HZ.