Résumé
Objective To compare the efficacy and safety among individualized regimens ( IR ) containing clofazimine ( Cfz ) , linezolid ( Lzd ) or meropenem-clavulanic acid ( MC ) in treatment of multidrug/extensively resistant tuberculosis (MDR/XDR-TB) by using network meta-analysis.Methods Randomized controlled trials ( RCTs) and observational studies of Cfz-IR, Lzd-IR and MC-IR regimens in the treatment of MDR/XDR-TB published from January 2000 to August 2017 at home and abroad were retrieved.The literature was screened according to inclusion and exclusion criteria.The quality of the literature was evaluated and valid data were extracted.The efficacy and safety of Cfz-IR, Lzd-IR and MC-IR in the treatment of MDR/XDR-TB were directly and indirectly compared by network meta-analysis.Relative risk ( RR) or relative comparative effect ( Mean) or adverse reaction rate and 95％confidence interval ( CI) were used as indicators of systematic evaluation , and Berg' s funnel plot was used for the existing publication bias.Results A total of 20 papers and 21 studies were included.There were 2490 cases in the study group and 2303 cases in the control group included.According to the direct comparison of the network meta-analysis, the efficacy for treatment of MDR/XDR-TB of Lzd-IR ( RR=1.18, 95％ CI 1.02-1.36, Z=2.28, P <0.05) and MC-IR(RR=1.23,95％CI 1.01-1.50,Z=2.10,P<0.05)was better than that of IR in control group.The rates of adverse reactions of Lzd-IR, Cfz-IR and MC-IR were 29％(95％CI 0.24-0.35), 21％(95％CI 0.13-0.28) and 7％ (95％ CI 0.03-0.10), respectively.The top-down efficacy outcomes of the 4 individualized chemotherapy regimens were MC-IR (66.4％), Lzd-IR(22.6％), Cfz-IR (10.0％) and IR without Cfz, Lzd or MC (1.0％).The fitting model showed that MC-IR (Z=3.04, P<0.05) and Lzd-IR (Z=2.31, P<0.05) significantly shortened the "off"time compared with IR without Cfz, Lzd or MC.Conclusion The network meta-analysis shows that the efficacy and safety of regimen MC-IR are significantly higher than those of Lzd-IR and Cfz-IR in treatment of MDR/XDR-TB.
Résumé
La XDR-TB (resistente a isoniazida, rifampicina, alguna fluoroquinolona y al menos una entre kanamicina, amikacina o capreomicina), ha causado efectos devastadores en pacientes con SIDA y es prácticamente incurable. Se presentan 12 casos de localización pulmonar en pacientes no SIDA. Se trataron con esquemas que incluyeron en todos linezolid y en 9 moxifloxacino, todos negativizaron el examen directo y cultivo del esputo. Nueve pacientes cumplieron criterios de curación, 1 está aún en tratamiento y 2 abandonaron. Ocho pacientes presentaron efectos adversos, en solo 1 caso debió suspenderse la tioridazina. La utilización de linezolid, moxifloxacina y tioridazina han contribuido a la evolución satisfactoria de estos pacientes. Estos fármacos son considerados de utilidad en la serie reportada, debiendo ser utilizados en centros especializados con experiencia en el manejo de la TB MR y XDR-TB.
The XDR-TB (resistant to isoniazid, rifampiN, fluorquinolone and at least of the following: kanamycina, amikacyna or capreomycin), has caused devastating effects in patients with AIDS and is practically incurable. Twelve cases of pulmonary XDR-TB in non AIDS are described. All were treated with schemes that included linezolid in all and moxifloxacin in 9, all respiratory specimens became negative. Nine patients fulfilled healing criteria, 1 is still under treatment and 2 abandoned the therapy. Eight patients presented adverse effects, thioridazine was stopped in only one patient. Linezolid, moxifloxacin and tioridazin contributed to the satisfactory evolution of these patients. These drugs were considered useful in the reported series of cases and should be used in specialized centres with experience in the management of MR TB and XDR-TB.