• drug reactions;
  • hypersensitivity syndrome;
  • isoniazid;
  • positive patch test

W e present a 34-year-old woman with a personal background of smoking and simple chronic bronchitis with bronchial hyperreactivity. Diagnosed with pulmonary tuberculosis, she began triple tuberculostatic treatment with isoniazid, rifampicin, and pyrazinamide. After 2 months, she visited the hospital with a severely pruritic urticariform cutaneous eruption; treatment was started with chlorpheniramine, and pyrazinamide was discontinued because the treatment had been completed. One week later, she was admitted to the emergency department because of a worsening of the eruption, with edema in the face and neck, purpuric lesions in the lower extremities, cutaneous-mucous icterus, lymphadenopathy, asthenia, and a fever of 38°C.

The emergency analysis was as follows. Leukocytes numbered 11.5×109; eosinophilia was 8%, 62% segmented, 15% lymphocytes (not activated), and 14% monocytes; and hepatic function was altered with aspartate aminotransferases of 1459 U/l and Quick time of 32%. The viral serologies and ANAs were negative; renal function, ionograma, proteinograma, α-fetoprotein, and urine sediment were normal; and the levels of lactate dehydrogenase were high. The abdominal ecograph showed homogeneous hepatomegaly, and the thorax radiology a decrease of the one tuberculous infiltrate of the right lower lobe. The patient was admitted to the unit and started treatment with methylprednisolone at 0.5 mg/kg body weight and endovenous dexchlorpheniramine. Cycloserine, ethambutol, and streptomycin replaced the initial antituberculostatic treatment. After 15 days, the hepatic function and the eosinophilia were normalized. The cutaneous eruption persisted with lower intensity for several months.

The patient was referred to our service for evaluation of hypersensitivity to tuberculostatics 1 year later, when she was asymptomatic. The test of lymphocytic transformation to rifampicin, pyrazinamide, and isoniazid was negative. Epicutaneous tests were also carried out with 1% rifampicin in vaseline, 10% pyrazinamide in alcohol (ethanol 70%), and 1% isoniazid in saline, showing a strongly positive reaction to isoniazid after 48 and 96 h. Ten epicutaneous tests with isoniazid administered to healthy controls were negative.

The incidence of cutaneous reactions to drugs in the general population is unknown and varies according to the drugs involved. The hypersensitivity syndrome refers to a serious specific reaction secondary to drugs that affects the skin ( 1). Clinically, patients present cutaneous rash and fever (87%), and often hepatitis (51%), arthralgia, lymphadenopathy (75%), and/or blood alterations (fundamentally eosinophilia of 30% and atypical lymphocytes). The histologic pattern shows a lymphocytic infiltrate, sometimes mimicking cutaneous lymphoma; thus, this syndrome is also called drug-induced pseudolymphoma. The diagnosis of the hypersensitivity syndrome is usually slow due to its late onset (3–6 weeks after the beginning of the treatment with the drug), and its slow, potentially fatal evolution. The mortality rate is about 10%. The drugs frequently involved are aromatic antiepileptic drugs (phenytoin, carbamazepine, and phenobarbital), and also allopurinol, gold salts, diapason, and sorbinil. Up to now, isoniazid has not been described as causing this syndrome. Recovery is usually complete except for the rash, and hepatitis can persist for several weeks. The culprit drug has to be stopped, and corticosteroid treatment is often given at the acute stage. The treatment with steroids has been shown to be effective ( 2), although the use of this treatment is not completely accepted, especially for patients with allopurinol-induced hypersensitivity syndrome. Some patients have relapses even without reintroduction, mostly during corticosteroid decrease ( 3).

The hypersensitivity syndrome seems to be an idiosyncratic reaction; individual differences in the metabolism of aromatic anticonvulsants or sulfonamides may predispose patients to such reactions. Some cross-reactions have been observed between the different anticonvulsant drugs.

This syndrome represents only a severe form of hypersensitivity reaction. It is necessary to change this term to decrease the ambiguity of this denomination.

Cutaneous eruptions from isoniazid have been described in less than 1% of the patients treated. Cases of contact dermatitis have also been described in nurses and workers in the pharmaceutic industry ( 4, 5).

The patch test has been shown to be effective in the diagnosis of contact eczema and also in some cases of reactions to drugs (carbamazepine, tetrazepam, practolol, and pyrazolines) ( 6). In the case of hypersensitivity syndrome due to antiepileptic drugs, several cases with a positive patch test to carbamazepine have been reported ( 7).

The case that we report clinically is presented as hypersensitivity syndrome, but the drug involved, as confirmed by the epicutaneous tests, was isoniazid.


  1. Top of page
  2. References
  1. A case of severe cutaneous reactions secondary to isoniazid.