Minimally invasive direct coronary artery bypass versus percutaneous coronary stenting for stenosis of the left anterior descending artery.

We read with great interest a meta-analysis by Jaffery et al. [1] of randomized control trials (RCTs) comparing minimally invasive direct coronary artery bypass (MIDCAB) versus percutaneous coronary stenting (PCS) for stenosis of the proximal left anterior descending artery (LAD). Between PCS and MIDCAB, there was no difference in major adverse cardiac events (MACE), i.e. the composite end point of mortality, myocardial infarction, and target vessel revascularization: relative risk (RR) (95% confidence interval [CI]), 1.81 (0.80, 4.06). Excluding a trial with drug eluting stents (DES) by Hong et al. (reference number of the article by Jaffery et al. [1] [R-No] 22), however, resulted in a significantly higher risk of PCS in comparison to MIDCAB for MACE: RR (95% CI), 2.27 (1.32, 3.90). By searching Medline, etc. from January 1990 to January 2006, Jaffery et al. [1] finally identified five eligible RCTs by Cisowski et al. (published in 2002) (R-No 20), Drenth et al. (2002) (R-No 21), Thiele et al. (R-No 24), Hong et al. (R-No 22), and Reeves et al. (R-No 23). Cisowski et al. [2] and Drenth et al. [3], however, updated the results of their RCTs (R-No 20, 21). Furthermore, our comprehensive search identified another RCT by Kim et al. [4]. Therefore, we performed a meta-analysis of currently available RCTs by Cisowski et al. [2], Drenth et al. [3], Thiele et al. (R-No 24), Hong et al. (R-No 22), and Reeves et al. (R-No 23), and Kim et al. [4]. Three and two of the six individual trials demonstrated a statistically significant and nonsignificant benefit of MIDCAB over PCS for MACE: RR (95% CI), 4.00 (1.20, 13.32) [2]; 2.80 [1.09, 7.20] [3]; 1.76 (1.21, 2.55) (R-No 24); 2.00 (0.38, 10.43) (R-No 23); and 3.00 [0.86, 10.43] [4]. Merely one trial demonstrated a statistically nonsignificant MACE reduction with PCS (using DES) over MIDCAB: RR (95% CI), 0.46 (0.18, 1.17) (R-No 22). Pooled analysis of all the six trials demonstrated a statistically significant 83% increase in MACE with PCS relative to MIDCAB in a random-effect model: RR (95% CI), 1.83 (1.02, 3.26). There was significant trial heterogeneity of results by standard x tests ( p = 0.0459) but no evidence of significant publication bias by an adjusted rank-correlation test ( p = 0.5730). In conclusion, ourmeta-analysis of currently available RCTs showed that PCS was fraught with significantly increased risk of MACE in comparison to MIDCAB even including the trial with DES. References