Atypical Stevens-Johnson Syndrome in an Adolescent Treated with Duloxetine

To the Editor:

T he serotonin and norepinephrine reuptake inhibitor, duloxetine, has been approved by the U.S. Food and Drug Administration for the treatment of major depressive disorder, generalized anxiety disorder, fibromyalgia, and diabetic peripheral neropathic pain in adults (Eli Lilly, 2009). Recent, preliminary reports suggest that duloxetine may also be effective for major depression in adolescents (Delgado et al. 2007; March et al. 2009); multisite, double-blind, placebo-controlled trials are underway to test these preliminary findings. The extant literature suggests that duloxetine is generally well tolerated in children and adolescents (March et al. 2009) and we are unaware of any reports of serious, life-threatening interactions or adverse events associated with duloxetine in youth. Here, we report a case of an atypical Stevens-Johnson syndrome (SJS) in a 15-year-old adolescent boy with major depressive disorder who was treated with duloxetine monotherapy.

B is a 15-year-old White boy with a history of major depressive disorder without psychotic features, diagnosed using the Mini International Neuropsychiatric Interview for Children and Adolescents (MINI-KID) (Sheehan et al. 2010). At initial, outpatient presentation, B's Childhood Depression Rating Scale-Revised (CDRS-R) (Poznanski et al. 1979) score was 40 and his Clinical Global Impressions-Severity (Guy 1976) score was 5 and he was taking no medications. Duloxetine was started at 30 mg daily. After 2 weeks, B's CDRS-R score had decreased to 29 and duloxetine was increased to 60 mg daily. After 8 weeks of duloxetine monotherapy, depressive symptoms remained and duloxetine was increased to 90 mg daily, which was continued for 2 weeks at which point B's CDRS score was 27. At that time, duloxetine was titrated to 120 mg daily. Twelve weeks after beginning duloxetine and following 1 week of taking 120 mg daily, B complained of “sore throat” and cough. He was initially seen by his primary care physician and several days later in the emergency department of a community hospital. He was discharged home with supportive care (e.g., antitussive, albuterol inhaler). On the following day, B noted ∼4 vesicles in his oropharynx, which developed into 10–15 vesicles and bullae over the subsequent 24 hours and were accompanied by swelling of his lips, progressively worsening odynophagia for solid foods, and bilateral conjunctivitis. He denied having a headache, stiff neck, confusion, cough, rhinorrhea, abdominal pain, or joint aches. Upon presentation to the emergency department, his temperature was 38.1°C (100.5°F), heart rate was 112 beats per minute, blood pressure was 105/73 mm Hg, and respiratory rate was 16 breaths per minute. Multiple vessicles and bullae involved the tongue, lips, and oropharynx and there were slightly enlarged lymph nodes in the anterior cervical region and erythema of the urethral meatus. In addition, B was noted to have bilateral conjunctivitis, although there was no evidence of uveitis. The emergency department physician was initially concerned about herpetic stomatitis and possible atypical pneumonia because of radiographic evidence of bilateral perihilar infiltrates, which were radiographically consistent with pneumonia versus atelectasis. B was admitted to the inpatient medicine service for aggressive volume repletion with intravenous dextrose (5%) and half-normal saline commenced and he was treated with intravenous valacyclovir (300 mg every 8 hours) and oral azithromycin. At that time, duloxetine was stopped and laboratory studies were remarkable for a white count of 14.6 (78% neutrophils, 8% lymphocytes, 10% monocytes, and 4% eosinophils). Serum electrolytes were within normal limits as were blood urea, nitrogen, and creatnine, and herpes simplex virus polymerase chain reaction was requested. Over the next 12 hours, B's temperature increased to 38.7°C (101.6°F) and antibiotic coverage was expanded to include intravenous ceftriaxone (2 g daily). Intravenous ketorolac and morphine were also added for analgesia and the bullae and vessicles on his lips and tongue as well as throughout his oropharynx began to coalesce and appear ulcerated. On hospital day 3, B continued to require intravenous hydration secondary to his odynophagia; however, he was transitioned to oral acyclovir and cefdinir. By hospital day 4, B was able to tolerate clear liquids and the herpes simplex virus polymerase chain reaction test was negative; at that time, all antiviral medications were discontinued and SJS was the primary diagnosis. B was discharged, following a 4-day hospitalization, in fair condition to complete a course of cefdinir and azithromycin. Within 3 weeks of discontinuing duloxetine, B began experiencing worsening depressive symptoms and episodes of crying. Fluoxetine was started at 10 mg daily and slowly titrated to 40 mg daily over ∼4 months and remission of depressive symptoms was again achieved. Fluoxetine has been continued for 9 months, with sustained remission of his depressive symptoms.

This case highlights the atypical presentation of an adolescent with SJS, a potential serious side effect of duloxetine, which has heretofore only been reported in adults (Lacy et al. 2007; Lilly Research Laboratories 2010). To this end, the absence of “skin” involvement has been recently, but controversially, noted in other cases of SJS in youth (Ravin et al. 2007). Nonetheless, the course of symptom progression in our patient is generally consistent with “classic” SJS, in that there was a period of prodromal febrile illness, which can be easily mistaken for a nonspecific viral syndrome, and this initial phase progressed to a febrile illness accompanied by the development of multiple mucous membrane lesions (oropharynx, conjunctiva, and urethra).

Current data suggest that in patients treated with or exposed to duloxetine, SJS occurs very rarely (i.e., in 0.01% of these individuals) (Lilly Research Laboratories 2010). However, postmarketing surveillance to more precisely monitor for SJS associated with duloxetine treatment in youth is needed. In the meantime, clinicians should remain vigilant for mucosal membrane lesions with or without a serious rash in patients treated with duloxetine (Lacy et al. 2007).