Review : Photosensitivity in systemic lupus erythematosus: laboratory testing of ARA/ACR definition

Abstract

Objective. The aim of our study was to evaluate the prevalence of photosensitivity in SLE as defined by either clinical or laboratory assessment, the concordance of findings obtained by two methods, and the relationship between photosensitivity and clinical and immunological parameters.

Methods. Forty-four SLE patients and 31 healthy subjects were included. Patients and con trols underwent a standard questionnaire testing and the minimal erythemal dose (MED) measurement performed by Dermalight-Blue Point. The standard questionnaire was designed in order to meet, as near as possible, the definition of photosensitivity included in the ARA/ACR criteria for classification of SLE.

Results. The prevalence of photosensitivity was (patients vs controls): 57% vs 45% accord ing to questionnaire; 79.5% vs 51.6% (P = 0.02) according to MED. The agreement between questionnaire and phototest was absent in SLE (kappa 0.01) and poor in controls (kappa 0.36). Discoid rash was significantly associated with questionnaire positive (P = 0.01) and renal involvement with questionnaire negative results (P = 0.02), serositis with MED abnormality (P = 0.03), malar rash and anti-Sm antibody with MED normal values (P = 0.03 and P = 0.01, respectively). Moreover, by multivariate analysis, malar rash and anti-Sm antibody significantly predicted MED-defined photosensitivity, with probability ranging from 42% (presence of both) to 92% (lack of both).

Conclusions. Photosensitivity is frequently observed in SLE patients as well as in healthy subjects. Its prevalence is significantly higher in SLE than in controls only when it is detected using the laboratory method. However, due to the difficulty in objectively defining such manifestation, the disagreement between questionnaire and MED results was high and its clinical meaning appears ambiguous. Thus, the use of photosensitivity as a classification criterion for SLE remains questionable, at least when it is assessed according to the ARA/ACR definition.