Drug Rash with Eosinophilia and Systemic Symptoms (DRESS, hypersensitivity syndrome [HSS]; drug-induced hypersensitivity syndrome [DIHS); drug-induced delayed multiorgan hypersensitivity syndrome [DIDMOHS])

Are You Confident of the Diagnosis?

  • What you should be alert for in the history

  • Characteristic findings on physical examination

The term drug rash with eosinophilia and systemic symptoms (DRESS) was introduced in the mid-1990s to refer to a specific drug-induced hypersensitivity syndrome with well-defined characteristics, namely a skin eruption, hematologic abnormalities and internal organ involvement. The syndrome is heralded by a fever that ranges from 38 to 40°C. The onset of rash coincides with the fever, or it occurs shortly thereafter. A widespread exanthematous, or occasionally papulopustular or exfoliative eruption is seen (Figure 1).

Figure 1.

Facial edema is a common concomitant. Diffuse lymphadenopathy and pharyngitis accompany these findings. Liver abnormalities are seen in up to 50% of patients as evidenced by a transaminitis, elevated alkaline phosphatase and bilirubin. In some patients, hepatitis is severe and jaundice may result. Fulminant liver failure necessitating liver transplantation has also been described. Pneumonitis, nephritis, myocarditis, pancreatitis and encephalitis or aseptic meningitis occur less frequently. Rare findings include a thyroiditis with subsequent hypothyroidism, a hemophagocytic syndrome, adrenal insufficiency and colitis.

Laboratory findings include a leukocytosis with an eosinophilia in the majority of cases. An atypical lymphocytosis may also be seen in the early stages. Because eosinophilia is not universal, a Japanese group has proposed that the term drug-induced hypersensitivity syndrome (DIHS) replace DRESS.

Anticonvulsants (phenytoin, carbamazepine, phenobarbital), sulphonamide antimicrobials, minocycline, dapsone, allopurinol, and abacavir are among the most frequent offenders. DRESS from nevirapine, azathioprine and lamotrigine is also well described. Telepravir for hepatitis C is a newer offender. The onset of DRESS occurs 2 to 6 weeks after initiation of the offending medication.

This syndrome is classified as a severe cutaneous adverse reaction (SCAR) and mortality of up to 10% has been reported; 14 of 15 patients received intensive care unit treatment in one study. Convalescence is prolonged and both the rash and the hepatitis may take weeks, or sometimes months, to resolve. Relapses while tapering corticosteroid therapy can occur and these may be serious.

  • Expected results of diagnostic studies

A high index of suspicion is needed to make the diagnosis of DRESS, which is clinched by the presence of the triad outlined above. Laboratory testing should include a complete blood count with differential count, liver and renal function tests, as well as a urinalysis. Further testing is based on the clinical picture.

A skin biopsy is therefore not required to make the diagnosis in clear-cut cases. However, a skin biopsy may be helpful in cases of erythroderma, for example, to help clarify the diagnosis and rule out other causes of erythroderma such as Sezary syndrome and psoriasis, among others. When commented on in the literature, skin biopsies have manifested a broad range of findings. In eight patients in one study, lichenoid dermatitis was observed in three, changes of erythema multiforme in four, leukocytoclastic vasculitis in one and pseudolymphomatous changes in the last (Figure 2).

Figure 2.

Patch testing and the lymphocyte transformation assay (LTA) have been used to pin down the causative drug in cases of doubt. However, the overall sensitivity and specificity of patch testing is unknown and patch testing has yielded variable results in DRESS. It should be performed 2 to 6 months after the resolution of DRESS for best results. The positive predictive value appears to be highest with carbamazepine and lowest with phenobarbital. The LTA yields more reliable results. However, the sensitivity, specificity and predictive values seem to vary according to the drug.

  • Diagnosis confirmation

Viral syndromes have many of the same features of DRESS. The HIV seroconversion syndrome or acute Epstein-Barr virus infection may need to be considered.

Finally, skin and lymph nodes biopsies may manifest pseudolymphomatous features. These findings along with the peripheral atypical lymphocytosis make the spectrum of lymphoma, including angioimmunoblastic lymphadenopathy, a consideration.

The differential diagnosis of DRESS is broad. When the skin eruption is exanthematous, a morbilliform drug eruption may be suspected if attention is not given to the concomitant systemic and hematologic findings. DRESS may display pustules, which may make the distinction from acute generalized exanthematous pustulosis (AGEP) difficult. Again, the systemic and laboratory findings clinch the diagnosis of DRESS. A recent study reports that the vast majority of patients presenting with Stevens-Johnson syndrome or toxic epidermal necrolysis (TEN) from anticonvulsants displayed systemic and hematologic abnormalities. This suggests that targetoid lesions and epidermal necrosis be included in the morphologic list of DRESS. If the dermatologic presentation is exfoliative and erythrodermic, mycosis fungoides, psoriasis or other causes of erythroderma need to be ruled out.

Who is at Risk for Developing this Disease?

The incidence of DRESS to phenytoin, carbamazepine, and phenobarbital has been estimated to be 1 reaction per 5000 to 10,000 exposures. For sulphonamides, it is estimated to be 1 per 1000 to 10,000.

A genetic predisposition to DRESS has recently been identified. HLAB*5801 has been shown to be a genetic marker for SCAR, including DRESS, from allopurinol in the Han Chinese population. Similarly, the presence of HLAB*5701 has been strongly linked to the abacavir hypersensitivity syndrome in many populations, and nevirapine sensitivity to the presence of HLA-DRB1*0101.

Furthermore, the anticonvulsant hypersensitivity syndrome is thought to be a result of a genetic predisposition. Metabolites of phenytoin, carbamazepine, and phenobarbital are detoxified by epoxide hydroxylases. Arene oxide metabolism is genetically determined. Toxic metabolites accumulate when detoxification is defective and are thought to initiate an immune response and cell damage. Similarly, slow acetylators or those with glutathione deficiency are at risk for the sulphonamide hypersensitivity syndrome.

Evidence is mounting that reactivation of HHV-6 (or other viruses such as cytomegalovirus and Epstein-Barr virus) and the immune response to the virus seen plays an important role in the pathogenesis of disease. Specifically, a Japanese group has recently shown that when HHV-6 is reactivated, a more severe and prolonged course results. Whether the inciting drug bears any influence on the severity of DRESS has not been comprehensively studied. However, DRESS from allopurinol was found to be more severe compared with other drugs in one study.

What is the Cause of the Disease?

  • Etiology

Major advances in the understanding of the etiopathogenesis of DRESS have been made in the past decade, including the awareness of the genetic predisposition as mentioned above.

Human herpesvirus 6 (HHV-6) has been demonstrated to subserve an important role in the pathogenesis of DRESS. Reactivation of HHV-6 has been demonstrated in many patients with DRESS at 2 to 3 weeks after the onset of rash, as evidenced by a rise in HHV-6 IgG titres and HHV-6 DNA levels; 62% of 100 patients were found to display HHV-6 reactivation in a recent series. Pathogenetically, it is postulated that the offending drug induces immunosuppression that favors HHV-6 reactivation. Also, sodium valproate has been shown to directly stimulate HHV-6 replication in vitro.

The role that this reactivation, and the consequent immune response, has in disease pathogenesis is becoming clearer. As stated above, it has been shown recently that those patients with viral reactivation display more severe disease and have a more prolonged course. Also, all 5 of the 100 patients mentioned in the above study who died manifested evidence of HHV-6 reactivation. Active virus has been demonstrated in the liver and serum of a patient with fulminant hepatitis, and in the cerebrospinal fluid in a patient with encephalitis. While prognostic criteria, such as the SCORTEN criteria used to prognosticate in TEN have not yet been defined for DRESS, HHV-6 reactivation is emerging as a portender of a poor prognosis based on the above evidence.

An association between flaring of DRESS and HHV-6 reactivation has been shown. It is postulated that decreasing the corticosteroid dose incites an augmentation of the antiviral immune response that leads to flares.

  • Pathophysiology

In the anticonvulsant hypersensitivity syndrome metabolites of phenytoin, carbamazepine, and phenobarbital are detoxified by epoxide hydroxylases. Toxic metabolites accumulate when detoxification is defective and are thought to initiate an immune response and cell damage. In the sulphonamide metabolism, two pathways are operational. The first involves acetylation, which renders a non-toxic metabolite that is subsequently excreted. Slow acetylators are therefore at risk for toxicity. The second pathway, which involves cytochrome enzymes, requires glutathione to detoxify reactive metabolites.

Glutathione deficiency may also predispose to the syndrome. Similarly, reactive metabolites have been hypothesized to be responsible for DRESS from dapsone and minocycline.

On an immunologic level, drug-specific CD4 and CD8 cells release interleukin-5, which activates eosinophils and subsequent inflammation.

Systemic Implications and Complications

DRESS will have systemic involvement, by definition. The liver is affected more frequently than other internal organs. This is marked by a transaminitis and a high alkaline phosphatase and bilirubin. Fulminant liver failure may rarely occur. The kidney may be affected by interstitial nephritis or vasculitis, the central nervous system by encephalitis or aseptic meningitis, and the lungs by interstitial pneumonitis, the respiratory distress syndrome, or vasculitis. Pancreatitis, myocarditis, pericarditis, thyroiditis and colitis have also been reported.

In severe cases, a systemic immune response-type syndrome with shock and multiorgan failure may be seen. Pancytopenia, the syndrome of inappropriate ADH secretion (SIADH), adrenal insufficiency and a hemophagocytic syndrome may also rarely be seen.

All patients should undergo a complete blood count with differential, liver and renal function tests, electrolytes and a urinalysis. Further testing is based on the clinical picture and may include chest imaging studies, electrocardiogram, echocardiogram, thyroid function tests and pancreatic or cardiac enzymes.

Late complications of DRESS have been reported to include autoimmune diseases, such as type 1 diabetes mellitus, autoimmune thyroid disease, systemic lupus erythematosus and sclerodermoid skin lesions. These occur months to years after DRESS. The pathogenesis is unknown, but functional defect of T regulatory cells has been proposed as a cause for autoimmunity in these patients.

Treatment Options

Table I

Table I.
Medical treatment
Stop offending drug
Systemic corticosteroids
Intravenous immunoglobulin (IVIG)

Optimal Therapeutic Approach for this Disease

An in-depth history of drug exposure may be required to identify the responsible drug. It is imperative that this drug be discontinued promptly.

The presence of internal organ involvement is an indication for systemic corticosteroids, such as prednisone 1mg/kg/day. For severe cases, corticosteroids may be administered intravenously initially. Systemic steroids need to be tapered slowly, given the real risk of flaring on tapering.

New recommendations on the management of DRESS were recently proposed by the cutaneous adverse events working group of the French Society of Dermatologists. In this algorithm, for life-threatening illness, such as a hemophagocytic syndrome, adrenal insufficiency, encephalitis, or renal or respiratory failure, IVIG 2g/kg/day for 5 days is recommended in addition to corticosteroids. Furthermore, in severe cases with signs of HHV-6 viral reactivation, ganciclovir and/or IVIG in addition to corticosteroids along with close monitoring is recommended.

Cyclosporine has been employed in the literature as a steroid-sparing agent in one case, and when steroids were ineffective for an interstitial nephritis in another case. Both patients did well. This therapy may be considered in severe cases in addition to corticosteroids.

Patient Management

All possible drug culprits should be discontinued. A discussion with the patient and his or her family regarding the outlook of DRESS should be held. Based on the severity of the condition, the following therapeutic options may need to be discussed with the patient and his or her family members:

The patient should be monitored initially in the hospital. Patients with worsening organ function, pancytopenia, the hemophagocytic syndrome, or SIADH need to be monitored in the intensive care unit setting with consultation from the appropriate subspecialty teams. If organ dysfunction is worsening, oral corticosteroid therapy may be switched to the intravenous route, and IVIG may need to be considered, based on the algorithm outlined above. Fulminant liver failure necessitates an evaluation by the liver transplant team.

In severe cases especially, reactivation of viruses, including HHV-6, CMV and EBV, should be looked for. Ganciclovir may be considered in very ill patients with HHV-6 viral reactivation, as per the algorithm described in the section on the optimal therapeutic approach for this disease.

Stable patients with minor organ involvement may be managed in the outpatient setting. For all patients who are improving, the tapering of systemic steroids should occur slowly over weeks to months. The speed of tapering should be based on the speed of resolution of the underlying organ involvement, as evidenced by the appropriate laboratory tests. For patients with significant organ involvement that is improving, follow-up with respective specialists should also be arranged.

The likelihood of flaring of tapering of steroids needs to be discussed with the patient early on. If the patient experiences a flare of DRESS on tapering of steroids, this should be evaluated fully clinically and with laboratory tests. These flares may not be trivial. Flares may signal late onset viral reactivation and may need to be managed in the hospital. Flares may need antiviral therapy and/or IVIG based on whether HHV-6 viral reactivation is found, and the overall status of the patient.

Finally, the uncommon possibility of the late-onset autoimmune sequelae of DRESS should be explained and discussed.

Unusual Clinical Scenarios to Consider in Patient Management

While DRESS is resolving, or years after resolution, autoimmune complications may uncommonly eventuate. Typically, type 1 diabetes and thyroiditis have been reported to occur in the first year following DRESS. These may occur while the patient is still on their steroid taper. Fasting glucose and thyroid function tests should be actively monitored during this time frame.

What is the Evidence?

Ahluwalia, J1, Abuabara, K, Perman, MJ, Yan, AC. “Human herpesvirus 6 involvement in paediatric drug hypersensitivity syndrome”. Br J Dermatol. vol. 172. 2015 Apr. pp. 1090-5. (A series of pediatric DRESS cases confirming more severe disease is seen in the presence of HHV6 reactivation).

Bocquet, H, Bagot, M, Roujeau, JC. “Drug-induced pseudolymphoma and drug hypersensitivity syndrome (Drug Rash With Eosinophilia and Systemic Symptoms: DRESS)”. Semin Cutan Med Surg. vol. 15. 1996. pp. 250-7. (The original diagnostic criteria for DRESS were set forth in this article.)

Chen, ST, Wu, PA. “Severe cutaneous eruptions on telaprevir”. J Hepatol. vol. 57. 2012 Aug. pp. 470-2. (A case series detailing telaprevir-induced DRESS).

Descamps, V, Ben Saïd, B, Sassolas, B, Truchetet, F, Avenel-Audran, M, Girardin, P. “Management of drug reaction with eosinophilia and systemic symptoms (DRESS)”. Ann Dermatol Venereol. vol. 137. 2010. pp. 703-8. (This article in French provides an outstanding current review of DRESS and details management recommendations by a panel of experts.)

Elzagallaai, AA, Knowles, SR, Rieder, MJ, Bend, JR, Shear, NH, Koren, G. “Patch testing for the diagnosis of anticonvulsant hypersensitivity syndrome: a systematic review”. Drug Saf. vol. 32. 2009. pp. 391-408. (A systematic review of the usefulness of patch testing in the diagnosis of DRESS.)

Eshki, M, Allanore, L, Musette, P, Milpied, B, Grange, A, Guillaume, JC. “Twelve-year analysis of severe cases of drug reaction with eosinophilia and systemic symptoms: a cause of unpredictable multiorgan failure”. Arch Dermatol. vol. 145. 2009. pp. 67-72. (An in-depth clinical study of 15 patients with severe DRESS.)

Kano, Y, Ishida, T, Hirahara, K, Shiohara, T. “Visceral involvements and long-term sequelae in drug-induced hypersensitivity syndrome”. Med Clin North Am. vol. 94. 2010. pp. 743-59. (A comprehensive review of reported visceral involvement and long-term complications of DRESS.)

Knowles, SR, Shear, NH. “Recognition and management of severe cutaneous drug reactions”. Dermatol Clin. vol. 25. 2007. pp. 245-53. (A classic review article on DRESS Stevens-Johnson syndrome/TEN.)

Shiohara, T, Iijima, M, Ikezawa, Z, Hashimoto, K. “The diagnosis of a DRESS syndrome has been sufficiently established on the basis of typical clinical features and viral reactivations”. Br J Dermatol. vol. 156. 2007. pp. 1083-4. (A summary of the features of DRESS/DIHS, including the role of reactivation of viruses.)

Teraki, Y, Shibuya, M, Izaki, S. “Stevens-Johnson syndrome and toxic epidermal necrolysis due to anticonvulsants share certain clinical and laboratory features with drug-induced hypersensitivity syndrome, despite differences in cutaneous presentations”. Clin Exp Dermatol. vol. 35. 2010. pp. 723-8. (An article that describes seven patients with features of SJS/TEN and DRESS from anticonvulsants.)

Tohyama, M, Hashimoto, K, Yasukawa, M, Kimura, H, Horikawa, T, Nakajima, K. “Association of human herpesvirus 6 reactivation with the flaring and severity of drug-induced hypersensitivity syndrome”. Br J Dermatol. vol. 157. 2007. pp. 934-40. (This rigorous clinical study provides evidence of the pathogenetic role of HHV-6 in DRESS.)

Walsh, SA, Creamer, D. “Drug reaction with eosinophilia and systemic symptoms (DRESS): a clinical update and review of current thinking”. Clin Exp Dermatol. vol. 36. 2011. pp. 6-11. (Another recent current review of DRESS that emphasizes the spectrum of disease and the need for identification of prognostic factors.)