Table 3. Optimal DST characteristics depending on MDR TB management strategy, Peru*.

Programmatic and epidemiologic features	Optimal DST characteristics
Standardized versus individualized regimens
Standardized regimens for MDR based on regional resistance patterns	Centralized, complete DST (i.e., first- and second-line drugs) of representative samples to guide standardized treatment regimen; turnaround time less important
Individualized regimens	Rapid, point-of-care DST optimal to accommodate high demand and minimize turnaround time. Semi-individualized regimens may be constructed if only DST to first-line drugs performed.
Who is tested for DST?
Narrow DST indications (e.g., treatment failures only)	High pretest probability for MDR TB; therefore, optimal to perform DST to first- and second-line drugs to guide regimen design
Moderate DST indications (e.g., healthcare worker, smear-positive in second month of DOTS)	Rapid DST to first-line drugs to screen MDR TB versus non–MDR TB. If individualized treatment, drug-resistant samples may be referred for complete DST. Sensitivity may be more important than specificity because of greatest illness from failing to start appropriate treatment in patients with drug resistance.
Universal DST	Rapid DST to first-line drugs to screen MDR TB versus non–MDR TB. Rapid point-of-care testing (decentralized) optimal. If individualized treatment, drug-resistant samples may be referred for complete DST. Sensitivity may be more important than specificity.
Epidemiologic features
Patients with smear-negative disease (e.g., HIV, children)	Direct DST by using liquid medium or indirect DST after culture by liquid medium. Rapid turnaround time important given high illness rates in these risk groups.
High rates of resistance to second-line drugs (XDR TB)	Complete DST if high rates of resistance to second-line drugs, including XDR. If limited resources, DST to first-line drugs plus key second-line drugs (e.g., quinolone, kanamycin) to enable identification of XDR TB cases.
Management while awaiting DST results
Empiric first-line regimen	Greater risk for inadequate treatment of MDR TB cases; rapid testing more important
Empiric MDR TB regimen	Less risk for inadequate treatment of MDR TB cases, excess cost and toxicity for non–MDR TB cases. Complete DST results permit adjustment of empiric MDR TB therapy.

*DST, drug susceptibility testing; MDR TB, multidrug-resistant tuberculosis; XDR TB, extensively drug-resistant TB; DOTS, directly observed treatment, short course.