Anti-tuberculous drug allergy: Diagnostic challenges.

Abstract

When the tubercle bacilli were first discovered in 1882 by Dr Robert Koch, tuberculosis (TB) killed one in every four persons in England.1 At that time, there had been no known cure for TB disease, apart from high altitude, fresh air and good nutrition in hundreds of sanatoria built worldwide. With the discovery of antibiotics in the 1940s, TB patients had a hope for cure. In 1945, the first drug known to be effective against the TB bacilli, streptomycin, was discovered by Schatz, Bugie and Waksman.2 Shortly after, para-aminosalicylic acid (PAS) was discovered. However, their use as monotherapy for TB led to the rapid development of drug resistance. In the years following, several drugs including isoniazid (1951), pyrazinamide (1952) and ethambutol (1961) were discovered. Combination therapy of isoniazid, PAS and streptomycin, the “triple therapy” regimen of 18–24 months became the standard of care.3 It was only after the discovery of rifampicin in 1966, that TB treatment regimens were able to be shortened to just 6 months, when used in combination with pyrazinamide. One would expect that the discovery of antimicrobials against TB would result in TB elimination. However, TB continues to be one of the leading causes of death worldwide, making it an infectious disease of global importance. The World Health Organization's (WHO) END-TB (Tuberculosis) strategy has set ambitious goals to reduce TB deaths by 95%, TB cases by 90% and catastrophic expenses on families due to TB by 2035.4 South-East Asia (43%), Africa (25%) and the Western Pacific (18%) contribute to the majority of TB patients globally. Other than curative treatment for confirmed TB, the diagnosis and preventive treatment of latent TB infection (LTBI) are recognized by WHO as an important strategy to accelerate the decline in global TB and achieve TB elimination. Careful differentiation between LTBI and subclinical TB where the need to proactively exclude active TB infection definitively is crucial, as downstream contact tracing activities, preventive and curative treatment drug regimens and duration between these two entities clearly differ. Tuberculosis may present to the clinical immunologist/allergist as primary and secondary immunodeficiency diseases,5 for example disseminated Bacille Calmette–Guérin (BCG) or late-onset complications of BCG including osteomyelitis, complication of treatment with tumour necrosis factor (TNF) inhibitors,6 or advanced/untreated human immunodeficiency virus (HIV) infection. Allergists may also encounter TB in the course of providing a specialist opinion to the TB physician regarding hypersensitivity reactions to TB drugs and the suitability/need for rechallenge and/or desensitization. Antimicrobials remain the leading cause of drug allergy in both developing and developed nations. The diagnosis and identification of the culprit drug in TB drug allergy remain challenging because standard curative treatment regimens for TB usually comprise four concurrent drugs in the 2-month intensive phase followed by two drugs in the 4- to 7-month continuing phase. Regimens which exclude rifampicin, for example rifampicin allergy or for multi-drug resistant TB (MDR-TB) require the use of second- and third-line drugs, which have a less favourable side effect profile, longer treatment duration (up to 20 months) and lower effectiveness. In the rare circumstance that a strong regimen cannot be formed owing to severe drug reactions, novel drugs such as bedaquiline, delamanid and pretomanid may need to be used. Although several regimens involving the use of these drugs have shown to have good short-term microbiological outcomes, long-term data regarding disease relapse remain poorly understood at this point.7 It is our hope that this review will be a practical resource and provide greater clarity for all clinical immunologists/allergists and TB physicians who manage patients with TB drug allergy. Shera Tan and Bernard Yu-Hor Thong were involved in writing the editorial and references. No conflict of interest to declare. Data sharing is not applicable to this article as no new data were created or analyzed in this study.