Infarct Size Post-PCI Predictive of Mortality at One Year

September 19, 2014

WASHINGTON, DC — The size of the infarct following primary PCI for STEMI is strongly predictive of subsequent all-cause mortality and hospitalization for heart failure, according to a pooled analysis of 10 clinical trials.

Given the strong association between infarct size, which was measured within 30 days by MRI or single-photon-emission computed tomography (SPECT) and outcomes at one year, investigators say the size of damage done to the left ventricle might be a useful surrogate end point for future trials in STEMI patients.

"The study has major significance for clinical-trial design," lead investigator Dr Gregg Stone (Columbia University, New York) told heartwire , "because if we want to show differences in mortality or death and heart-failure hospitalization, often the trials need to include a thousand or a couple thousand patients. If we use a surrogate end point, we can reduce the sample size, but it has be a surrogate end point that you believe in, one that is linked to an end point you really care about."

The data, which included 2376 patients from 10 trials of primary PCI in STEMI patients, was presented this week at TCT 2014 . Of the studies, three used SPECT to measure infarct size and seven measured the end point using MRI. On average, it was five days post-PCI when the infarct size was measured. Stone noted that just 6.5% of patients included had had a prior MI, so this was unlikely to affect the infarct size assessment. The door-to-balloon time was 63 minutes and the symptom-to-balloon time was 232 minutes.

At one year, the mortality rate was 2.0%. In total, 2.5% of patients had a reinfarction, 2.7% were admitted to the hospital for heart failure, and 6.2% experienced the combined end point of death/MI/heart-failure hospitalization.

The median infarct size was 18.0% (assessed as a percentage of left ventricular mass). For patients who died by one year, the infarct size was significantly larger than those who survived. Similarly, those hospitalized for heart failure within one year had a significantly larger infarct size when compared with those who were not. For those readmitted for MI, infarct size was also significantly larger, although the relationship was not as strong.

Relationship Between Infarct Size and Outcomes

End point Infarct size, % (for those with clinical event) Infarct size, % (no clinical event) p
Death 29.0 17.8 0.0001
Reinfarction 21.0 18.0 0.05
Heart-failure hospitalization 28.9 17.3 <0.0001
Death, reinfarction, or heart-failure hospitalization 25.5 17.4 <0.0001

Dr Gregg Stone

When investigators analyzed infarct size by quartiles, they observed a significant graded association between the size of the infarction and mortality at one year. A similar graded association was observed for hospitalizations for heart failure and infarct size, but that relationship did not exist for reinfarctions. In a multivariate model, infarct size was predictive of death, heart-failure hospitalization, and the composite end point, but was not predictive of reinfarction.

"The study shows there is a very strong relationship between infarct size, as a surrogate end point, and subsequent death or heart-failure hospitalization," said Stone. "We think we can run studies that have 300 or 400 patients. If we show a very believable reduction in infarct size, then that should translate into better clinical outcomes."

The researchers found that SPECT and MRI were equally accurate at predicting clinical outcomes based on infarct size. During the TCT session, Stone said that previous studies have shown that baseline and final TIMI flow, time to balloon angioplasty, the presence of collateral arteries, and anterior vs nonanterior infarct-related arteries are all predictors of infarct size.

Stone told heartwire that he believes MRI and SPECT are both good means of assessing infarct size, although MRI does provide more information such as microvascular obstruction, ejection fraction, and left ventricular volumes. Resolution with MRI is better than SPECT, but it also more expensive.

Finally, Stone said the timing of left ventricular assessment is important because earlier measurements, such as within three to seven days, provide information on microvascular obstruction. However, the infarcts look larger at this stage, because edema is detected alongside the infarcted tissue. If the assessment is delayed until 30 days, the infarct shrinks, but by waiting some early clinical events can be missed, he noted.

Stone reports consulting fees from TherOx and VeloMedix.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.