Dapsone-induced DRESS after infliximab-induced vasculitis: a case of cerebral infarction in the context of multiple drug reactions

  1. Michael Roth 1,
  2. Jeffery T Gray 1,
  3. Paul Gabriel Peterson 2 and
  4. Paul A Clark 3
  1. 1 Department of Medicine, Walter Reed National Military Medical Center, Bethesda, Maryland, USA
  2. 2 Department of Radiology, Walter Reed National Military Medical Center, Bethesda, Maryland, USA
  3. 3 Department of Medicine, Inova Fairfax Hospital, Falls Church, Virginia, USA
  1. Correspondence to Dr Michael Roth; roth0470@umn.edu

Publication history

Accepted:05 Nov 2020
First published:17 Dec 2020
Online issue publication:17 Dec 2020

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

Drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome is a life-threatening condition characterised by peripheral eosinophilia, rash and multi-organ failure arising several weeks after exposure to the culprit medication. Although rare, DRESS syndrome triggered by specific agents has been associated with specific genetic polymorphisms more prevalent in different ethnic groups, including an association between dapsone-induced DRESS and Human Leukocyte Antigen (HLA)-B:13*01, a single nucleotide polymorphism more prevalent in those of Asian descent. DRESS and drug-related vasculitis may affect any organ system including the central nervous system (CNS), usually manifesting as encephalitis, meningitis or embolic cerebrovascular accidents related to eosinophilic cardiac disease and thrombosis. CNS vasculitis is a much rarer complication of drug reactions that may manifest as multifocal ischemia on neuroimaging. In circumstances of drug-related vasculitides, treatment with high-dose corticosteroids may lead to rapid improvement and, ultimately, resolution of associated focal neurologic deficits.

Background

Drug reaction with eosinophilia and systemic symptoms (DRESS) is a rare but potentially severe clinical entity occurring in between 1 in 1000 and 1 in 10 000 drug exposures but carrying mortality rates as high as 10%.1 2 Symptoms usually arise between 2 and 6 weeks after initial exposure but persist despite withdrawal of the culprit agent. Many drugs have been associated with DRESS including anticonvulsants such as phenytoin or carbamazepine, allopurinol and sulfur-containing drugs including sulfa agents and dapsone.1 2 Although the pathophysiology behind DRESS has not been fully elucidated, there appears to be an association between certain medications, polymorphisms in HLA genes and severe drug reactions including DRESS. In particular, HLA polymorphisms more common in certain ethnic populations may place those populations at higher risk for DRESS triggered by specific medications.1–3 Herpesvirus infection and/or reactivation, particularly with human herpesvirus 6 (HHV-6), has also been associated with DRESS.1 2 4

The typical clinical features of DRESS include a morbilliform rash, leucocytosis with profound eosinophilia or atypical lymphocytosis, fever, lymphadenopathy and the laboratory and clinical manifestations of multi-organ damage, most frequently hepatic injury.1 Although any organ may be involved, DRESS-associated central nervous system (CNS) vasculitis is an exceedingly rare phenomenon, with only a handful of cases reported in the literature.5 Despite the paucity of cases, early recognition of CNS vasculitis is crucial because of the potential for focal motor deficits and lasting neurological sequelae. We present a case of a patient presenting with DRESS and focal neurological deficits who was subsequently found to have multiple ischaemic foci on neuroimaging suggesting a CNS vasculitis.

Case presentation

A 31-year-old woman of Vietnamese descent with a medical history remarkable for Crohn’s disease controlled on infliximab and a documented allergy to trimethoprim–sulfamethoxazole was brought to the emergency department by family with hypersomnolence and a diffuse morbilliform rash. Initial history revealed that she had been started on dapsone 4 weeks earlier for biopsy-proven leucocytoclastic vasculitis thought to be secondary to her infliximab despite a prior reaction to sulfa medications. Two weeks after starting dapsone, she developed malaise and a diffuse, pruritic, morbilliform rash prompting an emergency department visit at an outside hospital and discontinuation of dapsone. Her symptoms persisted prompting admission to the same outside hospital 1 week later at which point she was diagnosed with DRESS, started on intravenous corticosteroids, and discharged 3 days later with an 8-day prednisone taper. Two days after discharge, she became increasingly somnolent with worsening rash prompting presentation to our institution.

On initial vital signs, she was afebrile but had a blood pressure of 92/63 mm Hg and a heart rate of 122 beats per minute. On physical examination, the patient was lethargic and disoriented to place and situation. Skin examination was notable for erythematous facial patches and a morbilliform rash covering the torso, buttocks, arms, legs, palms and soles without evidence of mucosal involvement (figure 1). Anterior cervical, axillary and inguinal lymphadenopathy were also noted. Neurological examination demonstrated bilateral patellar hyperreflexia and multiple beats of ankle clonus, right greater than left. Initial laboratory studies included a complete blood count remarkable for a leucocytosis to 54×109 cells /L with absolute eosinophilia (8640/µL) and anaemia with a haemoglobin of 93 g/L. A liver function panel was remarkable for elevated liver enzymes with an alanine aminotransferase (ALT) of 108 U/L and an aspartate aminotransferase (AST) of 76 U/L. An ECG demonstrated T wave inversions, and subsequent laboratory studies disclosed a troponin T of 0.81 ng/mL. The patient was admitted to the medical intensive care unit for further evaluation and initiation of corticosteroid therapy for presumed drug reaction.

Figure 1

Characteristic morbilliform rash present on patient’s hand (A) and arm (B). This rash was present over the patient’s torso, legs and soles.

Investigations

A brain MRI revealed multiple foci of restricted diffusion in the bilateral frontal, parietal and occipital lobes largely, but not entirely, in vascular watershed areas, indicative of acute infarction (figure 2). Also observed were areas of pathological enhancement in a similar vascular distribution. Accompanying MR angiogram demonstrated vessel irregularity in the distal middle cerebral artery and anterior cerebral artery distributions but no evidence of frank vascular thrombosis. The imaging findings were read as suggestive of a vasculitic process. A transcranial doppler study demonstrated no evidence for a cardioembolic source. Subsequent vasculitis work-up was remarkable for low C3 (83.4 mg/dL) and C4 (5.4 mg/dL) but negative for anti-neutrophil cytoplasmic antibodies (ANCA), anti-nuclear antibodies (ANA) and anti-double stranded DNA (ds-DNA) antibodies, a negative rapid plasma reagin and negative serologies and PCR for Epstein-Barr virus, cytomegalovirus, HIV and chronic viral hepatitides. Flow cytometry of peripheral blood was also performed and negative for a lymphoproliferative process, and total IgG values were within normal limits.

Figure 2

T2 fluid-attenuated inversion recovery (FLAIR) (A), diffusion-weighted Imaging (DWI) (B) and apparent diffusion coefficient (ADC) (C) MRI brain sequences demonstrating cortical infarcts in the watershed areas between the middle cerebral arteries and the anterior cerebral arteries bilaterally.

The patient’s troponin T initially downtrended to 0.40 ng/mL on hospital day 2, but then remained persistently elevated at that level raising concern for eosinophilic myocarditis. Subsequent cardiac MRI demonstrated numerous foci of subendocardial late gadolinium enhancement (figure 3) suggesting subendocardial fibrosis with no evidence of intracardiac thrombus and a normal left ventricular ejection fraction of 56%. Eosinophilic endomyocarditis secondary to DRESS was strongly suspected based on these findings.

Figure 3

Delayed enhancement inversion recovery image from cardiac MRI in the two-chamber plane demonstrating foci of abnormal late gadolinium enhancement consistent with endomyocardial fibrosis, suggestive of eosinophilic myocarditis.

Treatment

The patient was started on intravenous methylprednisolone at a dose of 1000 mg every 24 hours. The patient’s rash and eosinophilia improved rapidly following initiation of steroid therapy, with a decrease in absolute eosinophil count to 283/µL 2 days after admission. Her methylprednisolone dose was tapered starting on hospital day 6, and she was successfully transitioned to oral prednisone at a dose of 80 mg two times per day starting on hospital day 13. HHV-6 DNA testing ordered on admission returned positive on hospital week 2 further suggesting a diagnosis of DRESS. Her anaemia initially worsened to a nadir of 8.1 g/dL with an elevated lactate dehydrogenase to 896 U/L and a low haptoglobin to <10 mg/dL suggestive of haemolysis, presumably secondary to prior dapsone use. Her haemoglobin subsequently recovered gradually without intervention to 9.8 g/dL on discharge. Her transaminitis increased to an ALT and AST of 2727 and 1457 U/L, respectively, on hospital day 7 with an associated hyperbilirubinemia to 2.5 mg/dL (direct 1.6 mg/dL) before improving prior to discharge. Her mental status improved throughout her hospital stay, but she continued to have brisk reflexes and mild generalised weakness. She was discharged to an acute rehab facility on hospital day 15 for continued rehabilitation after cerebrovascular accident.

Outcome and follow-up

High-dose steroids were continued on discharge with 80 mg of oral prednisone daily and tapered gradually over time before complete discontinuation 3 months after admission. Laboratory monitoring at outpatient follow-up demonstrated improvement in haemoglobin to 131 g/L and in her ALT and AST to 55 and 27 U/L, respectively, with resolution of hyperbilirubinemia. Repeat cardiac MRI 3 months after her admission showed near-complete resolution of late gadolinium enhancement. At clinical follow-up 5 months after discharge, her neurological deficits had resolved with the exception of subtle right-hand weakness, and she had returned to full-time office work.

Discussion

We present a severe case of DRESS occurring after dapsone treatment for infliximab-triggered vasculitis and associated with hepatic injury, eosinophilic myocarditis and multifocal cerebral infarction. Despite the severity of our patient’s presentation, she exhibited rapid recovery after initiation of high-dose corticosteroids.

Though the pathogenic mechanisms of DRESS have not been fully determined, it is thought to result from an interplay between several factors, including genetic variations in drug metabolism and immune function and the re-activation of HHV infections, particularly HHV-6.1 2 4 Indeed, the association between HHV-6 and DRESS is so well established in previous studies that the presence of HHV-6 DNA has been suggested as a diagnostic and prognostic marker for the disease.2 4 6 Certain HLA phenotypes have also been associated with an increased risk of DRESS reactions. It is thought that interactions between drug molecules and HLA proteins can lead to hapten formation, altering antigen presentation and inducing a systemic autoimmune reaction. Well known examples include the association between abacavir-induced hypersensitivity and HLA-B*5701 and that between allopurinol and HLA-B*5801.1 2 Of particular relevance to our case is the association between HLA-B*13:01, a specific single nucleotide polymorphism common in individuals of East and Southeast Asian descent, and dapsone-induced DRESS, with up to 20% prevalence in Chinese populations and 4% prevalence in those of Southeast Asian descent compared with a near absence in Africans and Europeans.3 Higher incidences of dapsone-induced DRESS in these populations suggest that providers should exercise significant caution and consider alternative agents prior to prescribing dapsone in these patient populations.

Neurological manifestations in DRESS are varied but uncommon. Patients may develop an encephalitis or meningitis associated with HHV-6 re-activation that can present as headaches, seizures and even coma.1 Patients with eosinophilic myocarditis may develop cerebral infarctions secondary to cardiac emboli, with thromboembolic events occurring in up to 29% of patients with idiopathic eosinophilia.7 8

Cerebral infarctions secondary to vasculitis are not uncommon in hypereosinophilic syndrome and eosinophilic granulomatosis with polyangiitis, but are quite rare in DRESS, with only a handful of cases being reported to date.5 8–10 High levels of circulating eosinophils are thought to induce endothelial cell damage via degranulation and release of cationic proteins such as major basic protein, eosinophil cationic protein and eosinophil peroxidase.7 8 Upregulation of proinflammatory factors including tumour necrosis factor (TNF) alpha and cachectin further promotes eosinophilic degranulation, exacerbating vascular endothelial damage and leading to ischemia and infarction.8 11

As demonstrated in previous case studies, the diagnosis of cerebral vasculitis may be suggested in the proper clinical context by a constellation of MRI findings including multifocal ischaemic lesions with small vessel perivascular enhancement on contrasted series in the absence of an obvious embolic source.5 8 9 Other diagnostic considerations for these imaging findings include reversible cerebral vasoconstriction syndrome (RCVS), intracranial atherosclerosis and autoimmune and infectious encephalopathies.5 Of these possibilities, atherosclerosis was deemed unlikely given the patient’s relative youth and multifocal lesions. While embolic phenomena are a possible explanation given the patient’s concomitant myocarditis, the lack of visualised cardiac thrombus and negative transcranial doppler studies likewise make this diagnosis less likely. Similarly, RCVS would be an unusual explanation in our case given its usual association with a ‘thunderclap’ headache and sudden onset hypertension.12 A diagnosis of cerebral vasculitis could have been further suggested by cerebral angiography and definitively confirmed with a brain biopsy, though given our patient’s rapid improvement with the provision of steroids, these interventions were deferred as it was felt that they would not ultimately change management.

Cerebral vasculitis may be caused by a variety of autoimmune, infectious, neoplastic or drug-induced pathologies.12 13 Given our largely negative work-up for common infectious and autoimmune processes (with the exception of HHV-6, a finding highly correlated with DRESS), the patient’s negative flow cytometry, the coincidence of the imaging findings with biopsy-proven leukocytoclastic vasculitis (LCV) of the skin and a case of DRESS, and our patient’s rapid neurological improvement with steroid treatment, the diagnosis in this case was presumed to be most likely a drug-induced vasculitis. While infliximab-induced vasculitis has been demonstrated to commonly affect the peripheral nervous system, cerebral involvement has also been reported in previous studies of anti-TNF therapy.14–16 DRESS has also been reported as a likely cause of cerebral vasculitis in several case series and reports, though similar to our case, diagnostic biopsy was reportedly not pursued in these instances due to clinical improvement with empiric corticosteroid treatment.5 10 While the timing of symptoms in our case may argue more strongly for DRESS as the underlying aetiology, the possibility of infliximab-induced vasculitis cannot be discounted as this may occur months to years after initiation of therapy. Regardless of the culprit agent, this case demonstrates that a high index of suspicion is required in potential cases of drug-induced and/or eosinophilic vasculitis as treatment will centre around withdrawal of offending agents and prompt administration of corticosteroids rather than on antiplatelet agents traditionally used for cerebrovascular accidents.5 10 14–16

Patient’s perspective

This circumstance changed my life in multiple ways. I’m more inclined to inquire and do my own research when a new medication is prescribed. These events may not have led to the severity that they did, if doctors would have listened to the information I was providing them. Additionally, if my chart was properly reviewed then my sulfa allergy would not have been overlooked. Every provider was quick to dismiss me and say it was the leukocytoclastic vasculitis, when in fact I knew it wasn’t from my experience. My case is a perfect depiction of how modern day medicine tends to be a systematic approach because of how electronic medical record systems are set up to follow certain protocols. Doctors fail the patient by not taking a moment to think outside the box and on the patient’s presenting symptoms. My hope with all of this, is that medical professionals pay close attention to a patient’s history.

Learning points

  • HLA B*13:01 is strongly associated with an increased risk of drug reaction with eosinophilia and systemic symptoms (DRESS) induced by dapsone and is a common genetic polymorphism in those of East and Southeast Asian descent.

  • Given the risk of DRESS, consideration should be given to selecting alternative agents to dapsone in patients of East and Southeast Asian descent.

  • Central nervous system (CNS) vasculitis is a rare but serious complication of drug reactions and should be considered in the differential diagnosis for patients with new extensive rash, focal neurological deficits and multiple ischaemic lesions on brain MRI in the absence of magnetic resonance angiographic lesions occurring in the context of a culprit medication.

  • Treatment for CNS lesions due to presumed drug induced vasculitis focuses on extended courses of highdose corticosteroids.

Acknowledgments

A special thank you to Dr K James DeMarco in the Department of Radiology at Walter Reed National Military Medical Center for your assistance with obtaining and interpreting representative neuroimaging findings.

Footnotes

  • Contributors This manuscript was primarily drafted by MJR, MD and was edited and revised by JTG, MD and PC, DO, all three of whom were primarily responsible for the medical care of this patient. Figures were chosen, assembled and captioned with the direction and assistance of PGP. All of the four aforementioned authors reviewed and approved the final draft of the manuscript and figures.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Disclaimer The views expressed in this abstract are those of the authors and do not reflect the official policy of the Department of the Navy, Department of Defense, or US Government.

  • Competing interests None declared.

  • Patient consent for publication Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer