Antituberculosis-Drugs Induced DRESS: A Multidrug Hypersensitivity or Drug Hypersensitivity Relapse? A Case Report

Abstract

DRESS related to first-line antituberculosis drugs (ATD) is a challenging diagnosis. With a long-lasting combined treatment of 4-concomitantly administrated drugs, identification of the culprit drug remains difficult and may expose patients to treatment interruption and affect their outcome. A 42-year-old female, treated with isoniazid, rifampicin, pyrazinamide and ethambutol for multifocal tuberculosis, developed, 40 days later, hyperthermia, facial edema, cervical lymphadenopathy and generalized exanthema. Biological test results revealed eosinophilia, atypical lymphocytes, and liver injury. DRESS was suspected, and ATD were withdrawn. As patch tests for the 4 ATD showed negative results, we decided to reintroduce pyrazinamide, ethambutol and rifampicin separately with a 3-day interval. Pyrazinamide and rifampicin were tolerated. However, after receiving ethambutol, she developed fever and generalized rash, with no biological abnormalities. Since ethambutol was claimed to be the culprit drug, isoniazid was added, and 10 hours later, the patient developed fever, facial edema, generalized rash, eosinophilia and liver injury. This clinical and biological pattern resolved 2 weeks later. This report suggests a hypersensitivity relapse to ethambutol after isoniazid-induced DRESS.

Introduction

Drug-induced hypersensitivity syndrome, also known as DRESS (Drug Rash Eosinophilia and Systemic Symptoms), is a severe, systemic, T cell mediated drug reaction with combinations of cutaneous presentation and internal visceral involvement. Initially associated with the use of anticonvulsants, the onset of the syndrome is now associated with other drugs, such as allopurinol, and anti-infective drugs including mainly sulfonamides and betalactams. ¹ First-line antituberculosis drugs (ATD) including isoniazid, rifampicin, pyrazinamide, and either streptomycin or ethambutol are highly efficacious treatments achieving cure rates of 90% to 95%. ² The most common adverse drug reactions (ADRs) of ATD are nausea/vomiting, hepatotoxicity and patterns of cutaneous drug reactions ranging from local pruritus to severe cutaneous adverse reactions (SCARs). ³ ATD-induced DRESS is an uncommon ADR, reported in only 0.7% ⁴ of all patients with tuberculosis. Note that, over the past few years, ATD has attracted attention as one of the potential triggers of DRESS, as an increasing number of case reports/series have been documented. ⁵ In the majority of cases, more than one drug were responsible for DRESS objected by either skin tests or rechallenge test suggesting either a multiple drug hypersensitivity syndrome (MDHS) or drug-related relapse. ⁶ Such cases illustrate the challenge faced by clinicians in identifying urgently and accurately which ATD have caused DRESS syndrome as they are prescribed in combination. Management of ATD induced DRESS is difficult as withdrawal of drugs can lead to drug resistance and switching to a less-effective second-line treatment. Here, we report a presumed case of isoniazid-induced DRESS with hypersensitivity relapse induced by ethambutol.

Case Report

A 42-year-old female with no relevant medical history was treated with isoniazid, rifampicin, pyrazinamide and ethambutol for multifocal tuberculosis. Forty days later, the patient developed hyperthermia, cervical lymphadenopathy, facial edema and generalized exfoliated exanthema (Figure 1).

Figure 1.

Clinical features during DRESS Syndrome: (A) facial edema and (B) A generalized morbliform Exanthema.

Blood tests revealed eosinophilia (1560), atypical lymphocytes. Liver tests showed cholestatic liver injury: Aspartate aminotransferase (AST) = 133(3.3 ULN), alanine aminotransferase (ALT) = 144 IU L⁻¹(3.6 ULN), alkaline phosphatase (ALP) = 462 IU L⁻¹ (6.1 ULN), gamma-glutamyl transferase (GGT) = 726 IU L⁻¹, (4.8 ULN). Prothrombin time ratio and renal function were normal. Based on these findings, DRESS was suspected and 4 ATD were withdrawn. The patient began to improve gradually with the disappearance of the skin rash, 2 weeks later, with use of cetirizine therapy and topical corticosteroids. Serological tests were negative for viral hepatitis, Epstein-Barr virus, cytomegalovirus and parvovirus B 19. A RegiSCAR score of 8 points indicates that the case presented here was definitely of DRESS ⁷ (Table 1).

Table 1.

RegiSCAR Scoring System for Classifying DRESS Cases, Published by Kardaun et al and Applied to Our Case.

Score	–1	0	1	2	Case score
Fever ≥38.5°C	No/U	Yes			1
Enlarged lymph nodes		No/U	Yes		1
Eosinophilia		No/U	0.7-1.499 × 109 L−1	≥1.5 × 109 L−1	1
Eosinophils, if leukocytes <4.0 × 109 L−1			10%-19.9%	≥20%
Atypical lymphocytes		No/U	Yes		1
Skin involvement
Skin rash extent (% body surface area)		No/U	>50%		1
Skin rash suggesting DRESS	No	U	Yes		1
Biopsy suggesting DRESS	No	Yes/U			0
Organ involvement*
Liver		No/U	Yes		1
Kidney		No/U	Yes		0
Muscle/heart		No/U	Yes		0
Pancreas		No/U	Yes		0
Other organ		No/U	Yes
Resolution ≥15 days	No/U	Yes			0
Evaluation of other potential causes
Antinuclear antibody
Blood culture
Serology for HAV/HBV/HCV
Chlamydia/mycoplasma
If none positive and ≥3 of above negative			Yes		1
Total score					8

Note. The diagnosis of DRESS syndrome is made based on the total score: <2 points: no case; 2-3 points: possible case; 4-5 points: probable case; >5 points: definite case. ANA = anti-nuclear antibody; BSA = body surface area; HAV = hepatitis A virus; HBV = hepatitis B virus; HCV = hepatitis C virus; N = no; U = unknown; WBC = white blood cell; Y = yes.

As it was urgent to restore tuberculosis treatment, we performed 1 month after the disappearance of the skin lesions, patch tests (PTs) to the 4 tuberculosis drugs (30% in petrolatum) that showed negative results at 48, 72, and 96-hour readings for rifampicin, pyrazinamide, isoniazid and ethambutol. An intradermal test with rifampicin 10% aq. yielded a negative result, too.

Thereupon, we have opted for a sequential and additive reintroduction of pyrazinamide, ethambutol then rifampicin with a 3-day interval and gradual dose escalation. We did not plan to reintroduce isoniazid as we thought it was the culprit drug. The patient tolerated pyrazinamide but when receiving ethambutol, she developed a fever (38.5 °C) with pruritic localized exanthema. Biological tests were within normal ranges. Ethambutol was withheld. However, pyrazinamide was continued. The rash resolved 2 days later. As ethambutol was considered the elicitor drug, we decided to reintroduce rifampicin and isoniazid. Rifampicin along with pyrazinamide, were administrated, without relapse of hypersensitivity. Nevertheless, after 10 hours from isoniazid (100 mg) administration, she manifested a fever (40°C), facial edema and a generalized pruritic eruption. Eosinophilia of 710 eosinophils with cholestatic liver injury were objected: AST = 465 IU-1(11.6 ULN) and ALT = 472 IU-1(11.8 ULN), GGT = 206 and APL = 364 IU l-1, with PT of 42%. All medications were withdrawn and cetirizine was started. The skin rash had improved and biological findings were without abnormalities, 2 weeks later. Pyrazinamide and rifampicin were reintroduced without incident. Streptomycin was added to the antituberculosis regimen. A rapid desensitization to ethambutol was performed with success. ⁸ However, the patient was unable to tolerate isoniazid desensitization, as she redeveloped fever with exanthema, despite the prescription of antihistaminic medication. Therefore, antituberculosis regimen including rifampicin, pyrazinamide, ethambutol and streptomycin was maintained without any untoward event.

Discussion

We describe a clinical observation of hypersensitivity relapse to ethambutol after DRESS induced by isoniazid. We believe that DRESS would obviously be related to isoniazid in view of (1) a clear temporal relationship between the administration of isoniazid and the onset of symptoms (40 days, typically 2-8 weeks), (2) Remission of the symptomological pattern after withdrawal of the drug, (3) Association of different symptoms evoking a clinical picture of DRESS syndrome, and (4) the relapse of symptoms with a severe pattern when isoniazid was reintroduced. Based on the Naranjo et al algorithm, it is probable that the systemic reaction was due to isoniazid ⁹ (Table 2).

Table 2.

Naranjo Adverse Drug Reaction Probability Applied to Our Case.

Naranjo Score
Question	Yes	No	Do Not Know or Not Done	Score In Our Case for ISONIAZID	Score In Our Case for Ethambutol
Are there previous conclusive reports on this reaction?	+l	0	0	+1	+1
Did the adverse event appear after the suspected drug was given?	+2	–1	0	+2	+2
Did the adverse reaction improve when the drug was discontinued, or a specific antagonist was given?	+l	0	0	+1	+1
Did the adverse reaction appear when the drug was re-administered?	+2	–1	0	+2	–1
Are there alternative causes that could have caused the reaction?	–1	+2	0	–1	–1
Did the reaction reappear when a placebo was given?	–1	+1	0	0	0
Was the drug detected in any body fluid in toxic concentrations?	+1	0	0	0	0
Was the reaction more severe when the dose was increased, or less severe when the dose was decreased	+1	0	0	0	0
Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0	0
Was the adverse event confirmed by objective evidence?	+l	0	0	+l	0
Total Score				7	2

Note. Score ≥9: definite; 5-8: probable; 1-4: possible; ≤0: doubtful.

Cutaneous adverse reactions were reported in 5.7% of patients receiving antituberculosis therapy. ¹⁰ With a mortality rate up to 10%, DRESS must be promptly diagnosed along with immediate withdrawal of the culprit drug and appropriately treated. Since its first description with phenytoin in 1950, several drugs, including sulfonamides, allopurinol and anticonvulsive agents, have been identified as responsible for inducing this adverse effect. According to a recent review published by Sharifzadeh et al, ¹¹ out of 254 cases of antibacterial-induced DRESS, 107 (42%) were related to ATD. Rifampicin was the most associated drug, followed by isoniazid, ethambutol and pyrazinamide. However, after excluding patients with a diagnosis other than tuberculosis who received rifampicin, Allouchery et al ⁵ have shown isoniazid was the most offending drug, followed by rifampicin, pyrazinamide, and then ethambutol. Unfortunately, the determination of the culprit drug in our case was not possible by skin tests, as the patch test to isoniazid was negative. Arruti et al ¹² have described a clinical observation of DRESS using a patch test that was positive for isoniazid (1% diluted in water). While, in a study conducted by Lehloenya et al, ¹³ only 4 out of 20 patients with isoniazid-induced DRESS had a positive patch test. This discrepancy could be explained by the difference in the dilution procedure (percentage, vehicle) and the time interval between the recovery and performance of the patch tests. In our patient, the patch test was performed 1 month after the recovery of the skin lesions. This time frame could be too short to obtain a positive patch test to isoniazid. It should be noted that Coster et al have concluded, in the context of DRESS, that patch tests can be performed earlier than recommended and with lower test concentrations, but false-negative results can occur. ¹⁴ Still, the sensitivity of patch testing in DRESS has been estimated to be 57% at 72-hour reading in a prospective study. ¹⁵ Despite being the “gold standard” for the diagnosis of drug hypersensitivity reactions, oral provocation test (OPT) is contraindicated in case of DRESS since it’s a life-threatening condition. Nevertheless, for adequate management of tuberculosis and as the benefit of treatment with ATD outweighs or at least equals the risk of morbidity and mortality from the drug reaction, OPT would be indicated guided by the negative results in the allergy tests. ¹⁶ OPTs were negative for the 4 ATD in our case.

Nevertheless, as the manifested features in our patient were at a mild extent after rechallenge with ethambutol compared to those occurring after isoniazid rechallenge, it could be argued that ethambutol was not the main drug involved in DRESS itself in our patient, but in hypersensitivity relapse. Moreover, it was tolerated by the patient after resolution of the DRESS framework. A relapse or flare-up is defined as a transient re-occurrence of clinical symptoms and/or laboratory signs during or following DRESS (such as exanthema, recurrent eosinophilia, elevation of liver enzymes), ¹⁷ related either to rapid reduction of systemic steroids, viral reactivation, administration of new drugs or administration of previously tolerated drugs. Drug-related relapse by a chemically distinct drug to the main elicitor is due to the fact that the drug responsible for the second reaction was administered during the immunological disturbances occurring in the initial DRESS episode. ¹⁸ It has been claimed that such drug-related relapse does not lead to sensitization because it does not stimulate the immune system long enough to cause T-cell expansion and permanent sensitization to this new drug. Therefore, skin and in-vitro tests for the second drug remain negative. With unproven sensitization, reintroduction of this drug may be allowed when needed, as this later can be tolerated when the activation of T cells caused by the initial DRESS episode has resolved. ¹⁷ In this regard, Jörg et al ¹⁷ have shown that out of 14 patients with drug-related relapse, 6 subjects needing the suspected drug were re-exposed without any reaction. This is the case of our patient who had taken ethambutol during the prodromal phase of DRESS and experienced a relapse when re-administered.

Still, re-exposure to this latter drug, after complete recovery, was well tolerated.

On the other hand, some patients could show a new T cell sensitization to the second drug, leading to MDHS. MDHS is defined as drug hypersensitivity confirmed either by skin tests or in-vitro tests to at least 2 chemically and pharmacologically unrelated drugs. ¹⁹ Consequently, with proof of an immune-mediated mechanism, drugs involved in MDHS must not be re-administered. Therefore, it is pivotal to differentiate drug-related relapse without proven sensitization from MDHS for subsequent therapeutic management. Consistently, as more than one antituberculosis drug is identified as the causative agent, it has been presumed that ATD are more often responsible for MDHS (48%). ²⁰ A recent study conducted by Jin et al showed that MDHS is frequently associated with ATD (42.9%). ²¹ ATD, given in fixed combination therapy, are at high risk of being associated with MDHS, as treatment schemes rely on a high dose of these drugs and a long-lasting regimen. ¹⁹ Barbaud et al underlined that MDHS can occur more frequently in case of SCARs, particularly in case of DRESS. In fact, MDHS to drugs administered at the onset of DRESS was observed in 13 out of 72 patients (18%). ²²

In conclusion, throughout this clinical observation, we point out a possible hypersensitivity relapse to ethambutol after isoniazid-induced DRESS in patients treated with first-line antituberculosis medication rather than a MDHS. As specific guidelines about allergological investigations and drug resumption strategies in DRESS induced by ATD remain inconclusive, clinicians should try to reintroduce the elicitor drug of hypersensitivity relapse under specific conditions to optimize tuberculosis treatment.