Recurrent Episodes of Stevens Johnson Syndrome (SJS): Clinical and Epidemiologic Characteristics
Methods: Children with an ICD-9 diagnosis of SJS and meeting clinical criteria for SJS between March 2013 and March 2014 were included in the study. SJS clinical criteria were defined as an episode of acute illness with inflammation of 2 or more mucous membranes and consistent skin lesions. Incomplete SJS was defined as an episode with consistent rash and involvement of one mucus membrane, or involvement of 2 mucus membranes and no rash. Recurrent SJS was defined as an episode of SJS or incomplete SJS separated from any previous episodes by at least 1 month of symptom resolution.
Results: Seven children had 21 SJS or incomplete SJS episodes, with a median of 3 per child (range 2 to 4). All 7 children were male, 6 (86%) were Caucasian, and 4 (57%) had history of asthma. Nine oral mucositis only episodes were excluded from our analysis. Fourteen episodes (67%) met the clinical definition for SJS and 7 (33%) met the definition of incomplete SJS. Age during episodes ranged from 4 to 15 years. The median time between episodes was 15 months (range 48 days to 5.6 years). Two episodes were preceded by antibiotics and none by anticonvulsants. Eleven (52%) episodes had documented preceding upper respiratory symptoms and 4 (19%) had radiographic pneumonia. Mycoplasma pneumoniae was identified by throat PCR in 3 (50%) of 6 individuals tested and 4 (25%) of 16 episodes tested. Ten episodes were tested for HSV by throat/oral PCR and all were negative. No family history of SJS was reported. Steroids were given for 4 episodes and IVIG was given for 3 episodes. Three children (5 episodes) had ocular disease requiring temporary amniotic membrane grafts.
Conclusion: We report the largest case series of recurrent SJS. Recurrent SJS does not appear to be related to repeat use of offending medications. The clinical presentation of children with recurrent SJS varies between episodes and individuals. Possible hypotheses for recurrent SJS include a genetic or immunologic predisposition and recurrent Mycoplasma pneumoniae infections.
A. Demirjian, None
X. Lin, None
C. C. Robinson, None
M. Glode, None
P. Kutty, None
S. R. Dominguez, None