| Objective: To assess the effect and safety of therapies in common use for acute myocardial infarction (AMI) patients with failed thrombolytic therapy. Methods: We searched the Cochrane Central Register of Controlled Trials (CENTRAL, Issue 2, 2006), MEDLINE (1966 to July 2006), EMBASE (1984 to July 2006), China National Knowledge Infrastructure (CNKI, 1994 to July 2006), China Biomedicine Database disc (CBMdisc, 1980 to July 2006). In addition, we tracked down the reference lists of included papers related to therapies for acute myocardial infarction patients with failed thrombolytic therapy. We also handsearched several key Chinese journals related to cardiovascular field. Language was limited to Chinese and English. We included all randomized controlled trials (RCTs), quasi-randomized controlled trials and clinical controlled trials for acute myocardial infarction patients with failed thrombolytic therapy. Two authors (Zhang Xiaoli, Ai Changlin) independently assessed the methodological quality of included studies and extracted data and checked one another. Disagreement was resolved by discussion or the third person (Li Jing). The following outcomes were assessed to examine the effect and safety of therapies after thrombolytic therapy failed: mortality rate, occurrence of severe heart failure, reperfusion rate, thromboembolic stroke, revascularization rate and bleeding at the end of therapy or follow-up. Data were analyzed by RevMan 4.2.8 from the Cochrane Collaboration. Relative risk (RR) and 95% confidence interval (CI) were calculated for categorical variables while weighted mean difference (WMD) and 95% confidence interval were calculated for numerical variables.Results: Eighteen trials (9 RCTs and 9 non-RCTs) with 2497 cases met the inclusion criteria. Three RCTs mentioned randomization method; 6 RCTs mentioned allocation concealment; 5 RCTs used blinding; 2 RCTs mentioned intention-to-treat. Three non-RCTs mentioned how patients were allocated; 7 non-RCTs mentioned the method to balance groups' baseline; All 9 non-RCTs identified prognostic factors that might affect outcomes; 0 non-RCTs mentioned whether case-mix adjustment was used.Comparing rescue PCI with traditional treatment: Meta-analysis demonstrated that: (1) Mortality: Five RCTs with 800 cases (405 in treatment group and 395 in control group) and 5 non-RCTs with 492 cases (282 in treatment group and 210 in control group) reported mortality. Meta-analysis of 5 RCTs showed that there was a significant difference between the rescue PCI group and traditional group in mortality at the end of the follow-up [RR=0.64, 95%CI (0.41, 0.98)], but there was no significant difference [RR=0.89, 95%CI (0.51, 1.57)] from 5 non-RCTs recta-analysis. (2) Reinfarction. Meta-analysis from 3 RCTs with 621 cases (311 in treatment group and 310 in control group) showed that there was no significant difference between the rescue PCI group and traditional group in the occurrence of reinfarction at the end of the follow-up [RR=0.47, 95%CI (0.17, 1.26)]. Three non-RCTs with 425 cases (247 in treatment group and 178 in control group) reported occurrence of reinfarction. We didn't use meta-analysis to pool the data because of significant heterogeneity between the trials. (3) Improvement of left heart function: Two RCTs with 180 cases (92 in treatment group and 88 in control group) and 3 non-RCTs with 200 cases (68 in treatment group and 132 in control group) reported improvement of left heart function We didn't use meta-analysis to pool the data because of significant heterogeneity between the trials. (4) Severe heart failure: Four RCTs with 760 cases (387 in treatment group and 373 in control group) and 2 non-RCTs with 559 cases (256 in treatment group and 3 03 in control group) reported severe heart failure. Meta-analysis showed that there was no significant difference between the rescue PCI group and traditional group in the occurrence of severe heart failure at the end of the follow-up [RR=0.72, 95%CI (0.52, 1.02); RR=4.31, 95%CI (0.49, 37.96) respectively]. (5) Thromboembolic stroke: Meta-analysis from 4 RCTs with 637 cases (325 in treatment group and 312 in control group)showed that thromboembolic stroke in rescue PCI group was significantly higher than that in traditional group [RR=4.39, 95%CI (1.14, 16.87)]. (6) Bleeding: One RCT with 285 cases (144 in treatment group and 141 in control group) showed that bleeding in rescue PCI group was significantly higher than that in traditional group[RR=2.79, 95%CI (1.55, 5.02)], but meta-analysis from 2 non-RCTs with 597 cases (231 in treatment group and 366 in control group) showed that there was no significant difference between the two groups [RR=I:10, 95%CI (0.73, 1.65)].Comparing rescue thrombolytic treatment with traditional treatment: Meta-analysis demonstrated that: (1) Mortality: Two RCTs with 373 cases (187 in treatment group and 186 in control group) reported mortality. We didn't use meta-analysis to pool the data because of significant heterogeneity between the two trials. (2) Reperfusion: One RCT with 90 cases (45 in treatment group and 45 in control group) showed that reperfusion rate in rescue thrombolytic treatment group was significantly higher than that in traditional group [RR=2.92, 95%CI (1.75, 4.85)]. (3) Bleeding: Meta-analysis from 2 RCTs with 373 cases (187 in treatment group and 186 in control group) showed that there was no significant difference in the occurrence of bleeding between rescue thrombolytic group and traditional groups [RR=l.95, 95%CI (0.86, 4.42)].Comparing rescue PCI with rescue thrombolytic treatment: Meta-analysis demonstrated that: (1) Mortality: One RCT including 286 cases (144 in treatment group and 142 in control group) and 1 non-RCT including 157 cases (103 in treatment group and 54 in control group) showed that there was no significant difference between the rescue PCI and rescue thrombolytic groups in mortality rate at the end of the follow-up [RR=0.49, 95%CI (0.23, 1.05); RR=I.05, 95%CI (0.27, 4.03) respectively]. (2)Revascularization: One RCT including 286 cases (144 in treatment group and 142 in control group) showed that revascularization in rescue PCI group was significantly lower than that in rescue thrombolytic group [RR=0.57, 95%CI (0.34, 0.95)]. (3) Bleeding: One RCT including 286 cases (144 in treatment group and 142 in control group) showed that bleeding in rescue PCI group was significantly higher than that in rescue thrombolytic group [RR=2.15, 95%CI (1.27, 3.63)], but no difference [RR=I.31, 95%CI (0.73, 2.35)] from 1 non-RCT including 157 cases (103 in treatment group and 54 in control group).Comparing glycoprotein (GP)â…¡b/â…¢a receptor antagonists with standard treatment: Meta-analysis demonstrated that: (1) Mortality: Two RCTs including 166 cases (82 in treatment group and 84 in control group) and 2 non-RCTs including 95 cases (50 in treatment group and 50 in control group) showed that there was no significant difference between the glycoprotein (GP)â…¡b/â…¢a receptor antagonists and standard treatment groups in mortality rate at the end of the follow-up [RR =0.68, 95%CI (0.20, 2.33); RR=0.34, 95%CI (0.04, 3.13)]. (2) Bleeding: Three RCTs including 255 cases (126 in treatment group and 129 in control group) and 1 non-RCT including 59 cases (29 in treatment group and 30 in control group) showed that there was no significant difference between the glycoprotein (GP)â…¡b/â…¢a receptor antagonists group and standard treatment group in bleeding.Conclusion: Current evidence does not confirm the effect and safety of the therapies for AMI patients with failed thrombolytic therapy and does not support the routine use of these therapies in clinical practice except that rescue PCI reduces mortality compared with traditional treatment. Further high-quality randomized controlled trials are needed to provide reliable evidence on the treatments of AMI patients with failed thrombolytic therapy.
|
PDF Full Text Request