Objective: To determine acquired drug resistance among failure and relapse cases after treatment of new smear-positive tuberculosis.
Methods: A cohort of 2901 patients with new smear-positive tuberculosis was enrolled in Vietnam. Sputum samples were stored at enrolment. Upon failure or relapse, another sputum sample was collected. Both were cultured and underwent drug susceptibility testing and restriction fragment length polymorphism (RFLP) typing.
Results: Of 40 failure cases, 17 had multidrug resistance (MDR) at enrolment. At failure, 15 of the 23 (65%) patients without primary MDR had acquired MDR. Of 39 relapse cases and 143 controls, none had primary MDR.
Conclusion: Primary drug resistance was a strong risk factor for failure and relapse and for acquiring further resistance. As 80% of failure cases had MDR, the standard re-treatment regimen appears inadequate for failure cases in this control programme with a very high cure rate among new cases.