Conference Coverage

Expert proposes rethinking the classification of SJS/TEN



In the opinion of Neil H. Shear, MD, a stepwise approach is the best way to diagnose possible drug-induced skin disease and determine the root cause.

Dr. Neil H. Shear professor emeritus of dermatology, clinical pharmacology & toxicology, and medicine at the University of Toronto.

Dr. Neil H. Shear

“Often, we need to think of more than one cause,” he said during the annual meeting of the Society for Pediatric Dermatology. “It could be drug X. It could be drug Y. It could be contrast media. We must think broadly and pay special attention to skin of color, overlapping syndromes, and the changing diagnostic assessment over time.”

His suggested diagnostic triangle includes appearance of the rash or lesion(s), systemic impact, and histology. “The first is the appearance,” said Dr. Shear, professor emeritus of dermatology, clinical pharmacology and toxicology, and medicine at the University of Toronto. “Is it exanthem? Is it blistering? Don’t just say drug ‘rash.’ That doesn’t work. You need to know if there are systemic features, and sometimes histologic information can change your approach or diagnosis, but not as often as one might think,” he said, noting that, in his view, the two main factors are appearance and systemic impact.

The presence of fever is a hallmark of systemic problems, he continued, “so if you see fever, you know you’re probably going to be dealing with a complex reaction, so we need to know the morphology.” Consider whether it is simple exanthem (a mild, uncomplicated rash) or complex exanthem (drug rash with eosinophilia and systemic symptoms or fever, malaise, and adenopathy).

As for other morphologies, urticarial lesions could be urticaria or a serum sickness-like reaction, pustular lesions could be acneiform or acute generalized exanthematous pustulosis, while blistering lesions could suggest a fixed drug response or Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN).

Dr. Shear considers SJS/TEN as a spectrum of blistering disease, “because there’s not a single diagnosis,” he said. “There’s a spectrum, if you will, depending on how advanced people are in their disease.” He coauthored a 1991 report describing eight cases of mycoplasma and Stevens-Johnson syndrome. “I was surprised at how long that stood up as about the only paper in that area,” he said. “But there’s much more happening now with a proliferation of terms,” he added, referring to MIRM (Mycoplasma pneumonia–induced rash and mucositis), RIME (reactive infectious mucocutaneous eruption); and Fuchs syndrome, or SJS without skin lesions.

What was not appreciated in the early classification of SJS, he continued, was a “side basket” of bullous erythema multiforme. “We didn’t know what to call it,” he said. “At one point we called it bullous erythema multiforme. At another point we called it erythema multiforme major. We just didn’t know what it was.”

The appearance and systemic effects of SJS comprise what he termed SJS type 2 – or the early stages of TEN. Taken together, he refers to these two conditions as TEN Spectrum, or TENS. “One of the traps is that TENS can look like varicella, and vice versa, especially in very dark brown or black skin,” Dr. Shear said. “You have to be careful. A biopsy might be worthwhile. Acute lupus has the pathology of TENS but the patients are not as systemically ill as true TENS.”

In 2011, Japanese researchers reported on 38 cases of SJS associated with M. pneumoniae, and 78 cases of drug-induced SJS. They found that 66% of adult patients with M. pneumoniae–associated SJS developed mucocutaneous lesions and fever/respiratory symptoms on the same day, mostly shortness of breath and cough. In contrast, most of the patients aged under 20 years developed fever/respiratory symptoms before mucocutaneous involvement.

“The big clinical differentiator between drug-induced SJS and mycoplasma-induced SJS was respiratory disorder,” said Dr. Shear, who was not affiliated with the study. “That means you’re probably looking at something that’s mycoplasma related [when respiratory problems are present]. Even if you can’t prove it’s mycoplasma related, that probably needs to be the target of your therapy. The idea ... is to make sure it’s clear at the end. One, so they get better, and two, so that we’re not giving drugs needlessly when it was really mycoplasma.”

Noting that HLA-B*15:02 is a marker for carbamazepine-induced SJS and TEN, he said, “a positive HLA test can support the diagnosis, confirm the suspected offending drug, and is valuable for familial genetic counseling.”

As for treatment of SJS, TEN, and other cytotoxic T-lymphocyte–mediated severe cutaneous adverse reactions, a randomized Japanese clinical trial evaluating prednisolone 1-1.5 mg/kg/day IV versus etanercept 25-50 mg subcutaneously twice per week in 96 patients with SJS-TEN found that etanercept decreased the mortality rate by 8.3%. In addition, etanercept reduced skin healing time, when compared with prednisolone (a median of 14 vs. 19 days, respectively; P = .010), and was associated with a lower incidence of GI hemorrhage (2.6% vs. 18.2%, respectively; P = .03).

Dr. Shear said that he would like to see better therapeutics for severe, complex patients. “After leaving the hospital, people with SJS or people with TEN need to have ongoing care, consultation, and explanation so they and their families know what drugs are safe in the future.”

Dr. Shear disclosed that he has been a consultant to AbbVie, Amgen, Bausch Medicine, Novartis, Sanofi-Genzyme, UCB, LEO Pharma, Otsuka, Janssen, Alpha Laboratories, Lilly, ChemoCentryx, Vivoryon, Galderma, Innovaderm, Chromocell, and Kyowa Kirin.

Next Article: