Physician's Perspective

Medication-Induced Stevens-Johnson Syndrome in Elders

Steven R. Gambert, MD, AGSF, MACP Editor-in-Chief, Clinical Geriatrics®

I recently had the opportunity to consult on the case of an 88-year-old man who was transferred to our facility from his local hospital with what was presumed to be Stevens-Johnson syndrome resulting from allopurinol, which had been prescribed by his family physician 1 to 2 weeks earlier. Although the patient had a number of chronic medical conditions, including the recent onset of gout, the patient had remained independent in his activities of daily living and instrumental activities of daily living. Besides the newly prescribed allopurinol, no other changes had been made to his medication regimen, and he had no history of recent infections. While Stevens-Johnson syndrome is not commonly encountered in geriatric practice, it remains a sobering example of how quickly someone can develop a life-threatening condition even when they are otherwise well and functional.

When I first saw the patient, more than 60% of his body was affected, including his peripheral skin, conjunctiva, and mucous membranes. Besides topical treatment with silver sulfadiazine, the patient required significant pain management and rehydration for fluid and electrolyte loss through his denuded skin. The patient remained delirious until his pain could be systemically managed. Almost from the beginning, his blood pressure became increasingly difficult to maintain and his airway was a concern. He had a clear advance directive that expressed his wish to remain “do-not-resuscitate or intubate.” It also stated that he did not want to be kept alive by artificial means if he had a “terminal illness.” A family meeting took place, and given the situation and potential outcome, the family asked that only comfort care be provided. The patient passed away shortly thereafter.

Stevens-Johnson syndrome is a skin condition that results in cell death and separation of the epidermis from the dermis. It is thought to be a hypersensitivity reaction that affects the skin and mucous membranes and may result from a medication, infection, or rarely, cancer. It is most commonly associated with use of an antibiotic or antifungal, such as sulfonamides, penicillins, levofloxacin, azithromycin, fluconazole, and nystatin. Other commonly associated medications include allopurinol, phenytoin, isotretinoin, carbamazepine, valproic acid, and lamotrigine. More rarely, nonsteroidal anti-inflammatory medications may cause this disorder, although a higher risk of this has been noted in older patients. In cases of drug-induced Stevens-Johnson syndrome, symptoms typically start within 1 week of initiating treatment, as was the case with our patient. Infections that have been associated with Stevens-Johnson syndrome include herpes simplex, influenza, mumps, histoplasmosis, and mycoplasma pneumonia, among others.

Treatment of Stevens-Johnson syndrome is similar to that for patients suffering from thermal burns, with supportive care and symptom control being key. Proper intravenous (IV) fluid management, nutrition, and pain relief are essential, and any suspected causative medications must be discontinued immediately. Attention to the airway is important, as respiratory failure can result. Although many think that use of corticosteroids play a role in shortening the disease’s course, their use remains somewhat controversial and no randomized trials have made a conclusive argument either way. Intravenous immunoglobulin has been used to reduce the length of the hypersensitivity reaction and to improve symptoms. Early ophthalmology consultation is another essential step because corneal scarring and blindness may occur.

In general, when <10% of the skin is affected, the mortality rate is approximately 5% in younger persons and somewhat higher in elders. As the percentage of skin involvement increases, so does the mortality rate, with elders being especially vulnerable. In geriatric patients, skin involvement of as little as 20% of body surface area has been associated with a mortality rate as high as 50%, whereas younger individuals generally require 50% skin involvement to reach this mortality rate.

More recent data have linked Stevens-Johnson syndrome to human leukocyte antigen (HLA) typing. In the case of allopurinol, one study reported that 61% of patients with allopurinol-induced illness carried the HLA-B58 allele ( remains to be determined whether new advances in genomic/personalized medicine can help physicians better predict success and adverse outcomes from particular medications, and whether this knowledge could be applied to prevent Stevens-Johnson syndrome.

Although we have come a long way since Stevens-Johnson syndrome was first reported in 1922, it remains a life-threatening illness. Early recognition and prompt treatment are crucial, especially in elders.



Dr. Gambert is Professor of Medicine and Associate Chair for Clinical Program Development, Co-Director, Division of Gerontology and Geriatric Medicine, Department of Medicine, University of Maryland School of Medicine; Director, Geriatric Medicine, University of Maryland Medical Center and R Adams Cowley Shock Trauma Center; and Professor of Medicine, Division of Gerontology and Geriatric Medicine, Johns Hopkins University School of Medicine, Baltimore, MD.

Send your comments and questions for Dr. Gambert to: