The Open Artery Hypothesis Revisited

Abstracts & Commentary

There has been considerable interest for many years regarding the potential short- and long-term benefits in patients who have had a myocardial infarction (MI) if they end up with a patent infarct related artery (IRA). No randomized trial to date has been of sufficient size to resolve the question as to whether the infarct vessel should be opened by an interventional procedure in the weeks to months following the MI. Much observational data support the hypothesis that an open IRA is beneficial, and would result in lower morbidity and mortality in post-MI patients, as well as attenuation of adverse left ventricular remodeling. The Mid-America Heart Institute in Kansas City, Mo, a large and experienced coronary intervention program, has reported an observational study of 2000 consecutive patients who underwent a PCI for a chronic total coronary occlusion (CTO) between 1980 and 1999. The CTO cohort was carefully matched to a comparable number of patients who underwent coronary angiography but did not have an occluded vessel. A propensity scoring method was used to optimally match the 2 cohorts for a wide variety of clinical factors. In addition, the 2007 patients who underwent a PCI for CTO were stratified for success or failure of the procedure; long-term (up to 10 year) outcomes were assessed. The matched non-CTO cohort was obtained from the Mid-America Heart Institute registry of almost 26,000 PCI patients. The mean follow-up for the entire group was 91 months; 94% of the CTO cohort were included in this analysis. All patients with a CTO had to be at least 7 days out from an acute MI; some had more than 1 PCI. Technical success was defined as less than 40% residual stenosis. Stent use was extremely low in this report (7% of the CTO cohort). Stented subjects were treated with coumadin for 1 month until 1996, at which time ticlodipine or clopidogrel was used. The end points for analysis included in-hospital complications, procedural success rates, and 10-year cumulative survival.

Results: Of the 2007 CTO patients there were 514 failures. The latter group were clinically similar, but had more CABG and multivessel disease. The CTO vessel was equally divided between the RCA and LAD, with slightly less circumflex procedures. Nintey-six percent of individuals had only a 1-vessel intervention. The overall technical procedural success rates were approximately 70%, with rates improving over time, without an increase in major coronary events. Major acute hospital coronary events (MACE) were similar between the CTO group and controls. Failed CTO was associated with a higher in-hospital MACE (4.5% vs 3.2%, P = 0.02); there was a trend toward more Q and non-Q wave infarctions with failed CTO. Long-term results showed no difference in survival between the total CTO cohort and the matched non-CTO individuals; 10-year mortality 71%, representing a 3% per year death rate. CTO individuals who had a successful PCI had a somewhat higher 10-year survival, 73.5% vs. 65%, P = 0.001. The CTO patent group had a 10-year survival comparable to the non-CTO matched controls. Furthermore, the patients with single vessel procedures fared better than those with multivessel CTO. 

After adjustment for baseline differences, an initial CTO success was a significant independent predictor of 10-year survival. Of interest, a small number (64) of individuals with a failed PCI underwent CABG; these had an improved 10-year survival compared to those with failure to open the vessel. Suero and colleagues emphasize that this is the largest series of patients reported to date who had a PCI for a chronically occluded vessel, and they note that successful opening of the CTO was associated with improved 10-year survival. They conclude that these findings justify an aggressive attempt at PCI in eligible CTO patients. The technical success rates recently have been 82%. Short-term morbidity and mortality showed little difference between the CTO intervention group and matched controls, but there were more adverse outcomes in the group with failed CTO. Suero et al state that "a striking survival advantage" occurred in patients with successful opening of an occluded vessel. They recommend careful consideration of selected PCI in chronically obstructed lesions (Suero JA, et al. J Am Coll Cardiol. 2001;38:409-414).

Comment by Jonathan Abrams, MD

This interesting and important database does not directly address the open artery hypothesis, in that it is unknown whether the obstructed coronary vessels subjected to PCI were or were not responsible for prior MI; however, one has to assume that many to most of these were. Their conclusions that a successfully opened CTO results in a 10-year survival comparable to individuals without an occlusion (and was substantially better than in those with a failed attempt), are persuasive that, in selected individuals, efforts to open a coronary occlusion should be considered. This is further supported by the surprisingly high success rates in the later years of the study. It should be noted that stent use was extremely low; it is possible and even likely that had stents been used more widely, the successful PCI cohort would have fared even better with respect to long-term survival.

These results do not clearly support (or refute) the open artery concept, but are indirectly supportive of the view that a patent IRA is in the best interest of the post- MI individuals. Much data in the literature indicate that MI patients who leave the hospital with a patent IRA do better than those with a closed artery, with decreased morbidity and mortality as well as attenuation of LV cavity expansion over time. It has been suggested that an open IRA results in a more sturdy structural framework for the LV myocardium, thus diminishing the likelihood of LV dilatation within and outside the infarct zone. Many other advantages of a patent artery can be postulated. Microvascular perfusion, however, was not studied by the Mid-America Heart Institute, and one must assume that at least some of the individuals who underwent opening of a CTO in this database did not benefit at all from the PCI due to failure of downstream myocardium to enhance its nutrient blood supply and/or undergo improved function of viable myocardial cells. 

A related study in the July issue of the Journal of American College of Cardiology examines long-term post-MI outcomes in a highly sophisticated manner, but without attention to the issue of an open IRA. Gaudron and colleagues were able to link the development of systolic dysfunction and cavity enlargement, with abnormal hemodynamics and electrical instability, all correlating with late sudden death. Gaudron et al do not provide information on the IRA status. However, 28% of the cohort of 434 patients developed significant LV dilatation with an increase in diastolic volume, as well as a markedly increased mortality. This paper is one of many that demonstrate an adverse event rate in individuals who undergo LV remodeling, which occurs less often in subjects with an open IRA. The data are persuasive that electrical as well as mechanical hemodynamic features correlate with each other. As the LV enlarges over time, QTC remains longer in the subjects who ultimately do not survive. The data from the Mid-America Heart Institute, as well as considerable prior animal and human research, indicate that a patent IRA is in the individuals’ best interest. The study by Gaudron et al from Germany provides considerable insights into the mechanistic relationships between LV cavity enlargement and death. It is believed by many, but still not proven, that an open artery will prevent the linkage of cavity expansion and sudden death.

Finally, the NHLBI will be conducting a 3-year Occluded Artery Trial (OAT), which will randomize post-MI patients with a closed IRA between 3 and 28 days, to PCI or none. That should finally resolve these perplexing but important issues (Gaudron P, et al. J Am Coll Cardiol. 2001;38:33-40).