Chronic total occlusion: no more meta-analysis, please—a randomized clinical trial is urgently needed

Percutaneous treatment of chronic total occlusions (CTO), defined as >3 months old, total obstruction of a coronary artery, is a phenomenon that has gained popularity in the portfolio of Cardiac Cath Lab Units, facilitated by the development of new technologies that allow addressing cases not feasible a few years ago. Despite its growing popularity, these are procedures that require a highly experienced operator, long sessions with increasing radiation dose to the patient and operator, and the risk of potentially serious complications. It is therefore very important to know the risk-benefit balance that this technique can provide in a given patient. The recent meta-analysis on the impact of percutaneous coronary intervention of chronic total occlusions on left ventricular function and clinical outcome by Hoebers (1), offers new and interesting facts that force once more to ponder. In an extensive review of all studies published in the literature, Hoebers concluded that a successful percutaneous treatment of a CTO is associated with an improvement in the ejection fraction (EF) with an absolute increase of 4.44%, a reduction in the adverse remodeling and an improvement in survival (OR: 0.52). Without wanting to question the validity of this meta-analysis, the data provided by Hoebers also allows another interpretation. When the authors selected only studies in which there is a clear definition of the treated population, confirming that all patients have a CTO at least 3 months old and an evaluation period post procedure for more than 4 months, the average difference between pre and post procedure EF is 4.71 (95% CI: 3.26-6.16) in the successfully treated group of patients and 2.21 (95% CI: −1.46 to 589) in the technique failure group of patients. So, in both groups the EF increases very slightly (only in the first case being statistically significant) and the difference between the increases in EF in both groups is actually 2.5 points. Whether this way of analyzing a meta-analysis can be very questionable from a statistical point of view, what we have no doubt about is that this small difference, quite probably less than the coefficient of variation of many techniques that analyze the EF, is clinically very poor. Therefore, despite the conclusions of Hoebers’s meta-analysis, we are not sure that percutaneous treatment of a CTO, associated with an improvement in the EF, can provide any clinical significance.