OBJECTIVE: It is generally agreed that there is a seasonal variation in the prevalence of cutaneous manifestations of systemic lupus erythematosus (SLE). We investigated whether there is seasonal variation in the incidence of noncutaneous lupus flare in Hong Kong. METHODS: We reviewed all noncutaneous lupus flare in 222 consecutive patients with SLE followed in our clinic from 1995 to 2005. Specific organ involvement of each flare was reviewed. The variation in the prevalence of lupus flare by calendar month and the relation with climatic factors were determined. RESULTS: The total followup was 18,412 patient-months. In total, there were 313 episodes of noncutaneous flare recorded in 129 patients. There were more lupus flares in December and January [2.31 episodes, vs 1.58 episodes per 100 patient-months for other calendar months; relative risk (RR) 1.46, 95% CI 1.12-1.90, p = 0.004], and more flares of lupus nephritis in December and January (1.14 episodes, vs 0.60 episodes per 100 patient-months for other calendar months; RR 1.90, 95% CI 1.29-2.80, p = 0.001). There were more cases of membranous nephropathy in December and January (0.46 episode, vs 0.18 episode per 100 patient-months for other calendar months; RR 2.59, 95% CI 1.36-4.93, p = 0.0027), while the variation in prevalence of proliferative lupus nephritis was not statistically significant. There was also a significant U-shape correlation between the rate of lupus flare and the monthly average environmental temperature (r = 0.802, p = 0.0096), with higher flare rate at extremes of temperature. CONCLUSION: We found substantial seasonal variation in the incidence of noncutaneous flare in our SLE patients, with peak incidence in December and January. There was a U-shaped relation between environmental temperature and the prevalence of noncutaneous flare. Keeping a warm living environment and avoiding exposure to extremes of temperature may help to reduce flare for SLE patients in subtropical countries.