Table 6—

Summary of the variables analysed by the 13 studies reviewed

 First author [Ref.] Variables included 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Gandhi 9 Yes W Yes S Migliori 10 Yes Yes Yes Yes W Yes Yes Yes Yes Yes Yes Yes Yes I Migliori 11 Yes Yes Yes Yes W/L Yes Yes Yes Yes Yes Yes Yes Yes Yes I Kim 12 Yes Yes L Yes Yes Yes Yes Yes I Yes Yes Mitnick 16 Yes Yes Yes Yes L Yes Yes Yes Yes Yes Yes Yes Yes Yes I Yes Chan 17 L Yes Keshavjee 19 Yes Yes Yes L Yes Yes Yes Yes Yes Yes Yes I Yes Yes Eker 18 Yes Yes Yes Yes Yes L Yes Yes Yes Yes Yes Yes Yes Yes Yes I Yes Yes Kwon 20 Yes Yes L Yes Yes Yes Yes I Yes Lai 21 Yes Yes Yes Banerjee 22 Yes W Yes Yes Yes Yes I Bonilla 23 Yes Yes Yes L Yes Yes Yes Yes Yes S/I Kim 24 Yes Yes Yes Yes W Yes Yes Yes Yes Yes I Yes
• 1: representative sample (multidrug-resistant (MDR)-tuberculosis (TB) coverage); 2: history of previous treatment (MDR- versus extensively drug-resistant (XDR)-TB); 3: number with previous treatment for >1 month (MDR versus XDR); 4: number of drugs used in treatment (MDR versus XDR); 5: linezolid; 6: standard definitions according to Laserson et al. 28 (L) and/or the WHO (W); 7: number of drugs to which strains were resistant (MDR versus XDR); 8: 100% supranational reference laboratory quality assurance drug susceptibility testing; 9: all XDR-defining drugs tested; 10: times to sputum smear and/or culture conversion; 11: treatment success (MDR versus XDR); 12: failure (MDR versus XDR); 13: death (MDR versus XDR); 14: follow-up; 15: prevalence of HIV; 16: standard (S) or individualised (I) treatment; 17: surgery reported; 18: adverse events reported.