Diagnostic patch testing following tuberculosis-associated cutaneous adverse drug reactions induces systemic reactions in HIV-infected persons

Br J Dermatol. 2016 Jul;175(1):150-6. doi: 10.1111/bjd.14492. Epub 2016 May 26.

Abstract

Background: The incidence of cutaneous adverse drug reactions (CADRs) to first-line antituberculosis drugs (FLTDs) is higher in HIV-tuberculosis coinfection. However, the utility of patch testing to identify the offending drug in this patient subgroup has been poorly studied.

Objectives: To identify drugs causing adverse drug reactions in patients with HIV-tuberculosis coinfection.

Methods: Fourteen consecutive patients underwent diagnostic work-up (patch testing followed by a skin prick test and an oral rechallenge) to pinpoint the offending drug after developing FLTD-associated CADR, which included drug rash with eosinophilia and systemic symptoms (n = 12), Stevens-Johnson syndrome (SJS, n = 1) and toxic epidermal necrolysis/SJS overlap (n = 1). A positive reaction to any of the three diagnostic modalities eliminated that drug from the regimen. Once patients were clinically stable postreaction, sequential and additive rechallenge with FLTDs was initiated.

Results: Eleven of the 14 participants with FLTD-associated CADR were HIV infected (median CD4 count 149 cells mm(-3) ). In this subgroup, patch testing resulted in generalized systemic reactions in 10 of 11 patients (91%). These included rash in 10 of 13 reactions (77%), eosinophilia in eight (62%), transaminitis in seven (54%) and fever in five (38%). Isoniazid caused six of 13 (46%) generalized systemic reactions, rifampicin four (31%), ethambutol two (15%) and pyrazinamide one reaction. Using the Common Terminology Criteria for Adverse Events, five of 13 reactions were mild, six were moderate and two were severe. There were no life-threatening or fatal reactions.

Conclusions: In HIV-infected persons with tuberculosis-associated CADR, although patch-testing reactions to FLTD are common and tend to be associated with systemic features, they are not life threatening or fatal. These data inform clinical practice in HIV-endemic settings.

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Adolescent
  • Adult
  • Antitubercular Agents / adverse effects*
  • Coinfection / complications*
  • Drug Eruptions / diagnosis
  • Drug Eruptions / etiology*
  • Female
  • HIV Infections / complications*
  • Humans
  • Male
  • Patch Tests*
  • Stevens-Johnson Syndrome / diagnosis
  • Stevens-Johnson Syndrome / etiology
  • Tuberculosis / complications*
  • Tuberculosis / drug therapy
  • Young Adult

Substances

  • Antitubercular Agents