Stevens-Johnson Syndrome in a Patient on Concomitant Treatment with Levetiracetam and Trimethoprim/Sulfamethoxazole

Background

Trimethoprim/sulfamethoxazole (TMP-SMX), also known as cotrimoxazole, first received U.S. Food and Drug Administration (FDA) approval in 1973, later becoming one of the World Health Organization’s (WHO) essential medicines because of its affordability and wide range of use as an antibiotic for treatment and prophylaxis of many types of illnesses [1,2]. Levetiracetam is another widely prescribed drug after approval by the FDA in 2000, as adjunctive epileptic medication for focal seizures, myoclonic seizures, and primary generalized seizures [3]. It is also used off-label for anxiety disorder treatment [4]. In most cases, these medications have been safe and effective, but have sometimes been associated with major life-threatening cutaneous adverse drug reactions (cADRs) such as Sevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). In their systematic review, Lee et al [5] found that 28% (95% CI, 24–33%) of worldwide reported cases of SJS/TEN were associated with antibiotics, with sulfonamides having the strongest association 32% (95% CI, 22–44%). However, in their FDA adverse event reporting system-based pharmacovigilance study, Li et al [6] found that antiepileptics accounted for more than 16% of all reported drugs related to severe cADRs. These cADRs are mechanistically linked to delayed T cell- and NK cell-mediated hypersensitivity reactions. These reactions are characterized by a painful erythematous rash followed by blistering and skin detachment appearing on the face and trunk and spreading to the extremities [7–9]. Although it is rare, SJS is associated with a substantial morbidity and mortality, estimated at 40% if the offending agent is not discontinued [10].

At present, there is no up-to-date known drug–drug interaction between these 2 medications and no previously reported SJS cases for patients on both medications. In this paper, we present the rare case of a patient with SJS after receiving concomitant treatment with TMP-SMX and levetiracetam.

Case Report

This case report describes a severely ill 51-year-old White man with history of large-bowel perforation status after right laparotomy hemicolectomy with ileostomy in early 2022, split-thickness skin graft to the open abdominal surgical wound later that year, ileostomy take-down and incisional hernia repair in early 2023, and exploratory laparotomy repair of anastomotic leak and wound vac apposition to interval entero-cutaneous fistula and the non-healing latest surgical incision later the same year. The patient presented to the Emergency Department several months later with severe abdominal pain and a blistering skin rash that involved his entire face. The rash included mucosal erosions and crusts on the lips, neck, and the dorsum of the hands, with positive Nicolsky’s sign (Figure 1). The patient’s body surface area involvement was approximately 10%. The patient presented from a long-term acute care facility and has been on total parenteral nutrition for 5 weeks, had an ostomy bag collecting drainage from the enterocutaneous fistula, an indwelling Foley catheter, had received a 10-day course of TMP-SMX to treat a UTI within the last 2 weeks prior to hospitalization, no urine cultures and sensitivity available, and had been on levetiracetam for the last 2 months before presenting to the ED. However, his history of seizures was unclear and the only records we found showed that the medication was prescribed by a telemedicine primary care provider for off-label use for anxiety in a severely ill patient, but no documentation of formal diagnosis of seizures or evaluation by a neurologist or psychiatrist was found. The patient and his family denied any prior history of seizures. Upon admission to our hospital, he was hypotensive with a blood pressure of 85/52, temperature of 37.4°C, heart rate of 106 bpm, respiratory rate of 14 per/min, and O2 saturation 96% on room air. He was started on sepsis protocol, received appropriate fluid resuscitation, and was started on ceftriaxone 1 g IV daily and metronidazole 500 mg IV q12h. A urine culture revealed a pseudomonas-positive urinary tract infection (UTI) sensitive to ciprofloxacin and the blood culture showed candidemia. He was then started on micafungin, ciprofloxacin, and acyclovir. He was admitted to the Intensive Care Unit for closer monitoring of his respiratory status in the context of oral mucosa blistering.

Further studies included a Tzank smear that was negative for herpes simplex virus. A punch biopsy of skin, studied with hematoxylin and eosin (H&E) stain at a power field 200× showed necrotic keratinocytes, full-thickness epidermal necrosis, epidermo-dermal junction separation accompanied by neutrophils, and lymphocytes dermal infiltrate indicative of SJS (Figure 2). Levetiracetam was immediately stopped and acyclovir was discontinued, and he received symptomatic and supportive management only with appropriate wound care. Topical analgesics were applied along with antimicrobial therapy with micafungin and ciprofloxacin for his candidemia and UTI, respectively. On hospital day 17, the skin rash and lesions had completely healed, with some residual and faint erythema.

Discussion

Our patient, a White man who presented with SJS, was on 2 drugs known to be related to induction of SJS. It was, however, impossible to identify which one was the culprit. There are, unfortunately, no diagnostic tests useful for cADRs or SJS, and in vivo tests might be ethically questionable. No steroids, cyclosporin, immunoglobulin, or tumor necrosis factor (TNF)-alpha inhibitors were used for the treatment since there has been no strong evidence supporting their use in the treatment and reduction of mortality related to SJS [11].

It is now well established that the etiology of cADRs is multifactorial and encompasses genetic susceptibilities [10,12]. These include ethnicity-specific association and human leukocyte antigens (HLA) profiles, drug metabolism, and comorbidities, especially hematologic malignancy, kidney and liver disease, or some viral infections such as HIV and EBV [13,14]. The precise immune-histopathology of SJS is still being investigated, but this specific cADRs type is described as a type IV hypersensitivity reaction associated with a patient-specific set of HLA and the activation of CD4 and CD8 lymphocytes that occurs within 7 days to several weeks after drug exposure. The immune response is a result of T cell receptor activation following the presentation of the drug or its modified peptide sequence bound to the surface of HLA to T cell lymphocytes [13]. Hence, CD8 T cells and NK cells are considered to be the major inducer of keratinocyte apoptosis and epidermal necrosis [12,14].

Many papers in the literature provide evidence of the high potential of TMP-SMX as the triggering agent for SJS [5,13]. A multicenter case-control study from Taiwan, Thailand, and Malaysia showed a strong association between TMP-SMX-induced cADRs and the HLA-B*13: 01 allele, and further stratification of data revealed HLA-B*15: 02, HLA-C*06: 02, and HLA-C*08: 01 were more specific to SJS in the Thai population [15]. Similar findings were also reported by Kongpan et al [16]. However, in

Europeans, HLA-B*38 was the only allele found to be associated to TMP-SMX-induced SJS [17].

In the U.S., studies on N-acetyl transferase 2 (NAT2) phenotypic variation and TMP-SMX-induced severe cADRs revealed a strong association between slow acetylation phenotype and cADRs [14,15]. It was also estimated that in the U.S. population 56% of White patients, 44% of Black patients, and 5–20% of Asian patients are NAT2 slow acetylators. Thus, the U.S. FDA issued a warning about the association between NAT2 acetylator status and adverse drug reactions (ADRs), but no formal TMP-SMX drug labeling exists for the potential preventive role of NAT2 genetic testing [18,19].

Levetiracetam was also consistently found to induce severe cADRs, specifically SJS [20–24], with more than 90% of SJS developing within the first 2–4 months of antiepileptic drug use [19,25]. HLA-B*15: 02 and HLA-A*31: 01 and carbamazepine associations were found in Han Chinese and Thai populations and Europeans, respectively but no HLA allele susceptibility has been reported to date [15].

There are, however, very few published reports on the occurrence of cADRs in patients simultaneously taking 2 or more of the potentially high-risk drugs associated with SJS or TEN. Pérez-Calderón et al [7] reported a case of a patient who developed TEN subsequently to the concomitant use of 2 high-risk agents, TMP-SMX and temozolomide. In their investigation, controlled administration of other drugs revealed that these agents were simultaneous culprits, which suggested mutual potentiation of their respective cADR associations.

Noguchi et al [26] used the Japanese Adverse Drug Event Report database to quantify the risk of SJS or TEN associated with combinations of 2 antiepileptic drugs. They found that some drug combinations such as lacosamide-levetiracetam were associated with signals for SJS or TEN, but also found that in any combination, the signal intensity for a drug combination was more intense than that observed for a single agent. These and other data suggest a potentiation phenomenon that a combination of 2 high-risk agents could exert. Sawal et al [27] also reported a case of SJS after concomitant use of carbamaze-pine-levetiracetam and suggested the potentiating effect of combined levetiracetam and carbamazepine as a trigger for SJS. Finally, Zou et al [28] reported a case of levetiracetam-potentiated SJS in a patient who was on oxcarbazepine. These authors also suggested the possibility of cross-reactivity between levetiracetam and sulfa-type drugs that could lead to SJS.

Based on our review of the literature, we speculate that our patient might have suffered a complex interaction between TMP-SMX (a sulfa-containing drug known for its high risk for SJS/TEN association) and levetiracetam (another drug found to be associated with SJS and to potentiate other agents’ SJS association) amplifying their respective abilities to trigger SJS.

Conclusions

We presented a rare case of SJS that might have been induced by a combination of TMP-SMX and levetiracetam to help increase awareness among physicians about the potential risk of simultaneously prescribing these 2 agents.

Pharmacogenomics has been recently adopted in the U.S. to prevent ADRs related to carbamazepine; however, there is no HLA allele susceptibility yet identified for either levetiracetam or TMP-SMX. The U.S. FDA has issued a warning about the association between NAT2 acetylator status and ADRs related to TMP-SMX use. Thus, it seems appropriate that the FDA TMPSMX drug label emphasized the potentially preventive role of NAT2 genetic testing, which could help clinicians make an informed decision about their plan of care for patients. We also believe that advances in pharmacogenomics and their availability in different healthcare settings will help prevent potentially life-threatening ADRs, thus enhancing patient care and safety and dramatically reducing health care costs.