ASCERT: CABG bested PCI in high-risk patients

Issue: March 2012

Society of Thoracic Surgeons 48th Annual Meeting

Partial results from the ASCERT trial presented at The Society of Thoracic Surgeons 48th Annual Meeting showed higher survival rates among patients who underwent CABG compared with percutaneous coronary intervention.

“Previous observational studies have shown a long-term survival advantage for CABG over PCI,” Fred H. Edwards, MD, principal investigator and head of The Society of Thoracic Surgeons (STS) Research Center, said in a press release. “These partial ASCERT results confirm that, in important high-risk clinical subsets, the CABG survival advantage can also be seen in a large nationwide population.”

To compare survival after PCI with CABG, Edwards and colleagues examined data from more than 185,000 Medicare patients undergoing heart revascularization from 2004 to 2008. Information on CABG patients (n=86,244) was taken from the STS National Database and information on PCI patients (103,549) was taken from the American College of Cardiology National Cardiovascular Data Registry.

According to study results, first-year survival favored PCI in all high-risk patient subgroups. However, after 1 year, survival progressively increased in these subgroups for patients who underwent CABG. After dividing the spectrum of propensity scores into quintiles of CABG and PCI patients with similar clinical characteristics, survival advantage for CABG persisted across each quintile (RR for mortality at 4 years ranged from 0.75-0.82). Researchers also found that high-risk groups (RR at 4 years=0.72; 95% CI, 0.62-0.81) and low-risk groups (RR at 4 years=0.74; 95% CI, 0.64-0.84) demonstrated CABG survival advantage.

Based on these data, David M. Shahian, MD, of Massachusetts General Hospital in Boston, and colleagues estimated a long-term CABG survival model. They linked CMS claims data with STS clinical data from 348,341 CABG patients aged at least 65 years who were discharged between 2002 and 2007.

Estimated mortality for these CABG patients was 3.2% at 30 days, 8.1% at 1 year and 23.3% at 3 years of follow-up. Study results showed that early mortality was predicted with traditional short-term risk factors such as emergency status, shock and reoperation. However, the short-term risk factors often became nonsignificant within 2 years among early survivors.

“By linking the two clinical registry patient records with the longitudinal CMA claims data, ASCERT offers patients and physicians a comprehensive look at the comparative effectiveness of open heart surgery and stent procedures,” Edwards said. “The results of this subset study and our forthcoming report of the overall ASCERT population, to be presented at ACC’s Scientific Session in March, should be used to improve the quality of care for heart patients.”

For more information:

Edwards FH. Survival analysis of clinical subsets from the ASCERT study (ACCF-STS Database Collaboration on the Comparative Effectiveness of Revascularization Strategies): CABG compared to percutaneous stent placement in 189,793 patients with multivessel coronary disease.
Shahian DM. Predictors of long-term survival following coronary artery bypass grafting: results from the ASCERT Study. Both presented at: The Society of Thoracic Surgeons 48th Annual Meeting; Jan. 29-Feb. 1, 2012; Fort Lauderdale, Fla.

Disclosure: Dr. Edwards is a consultant and on the advisory board for Humana.

The ASCERT program provides important comparative-effectiveness data for high-risk patients undergoing PCI vs. CABG. This is a large patient cohort with outcomes available by linkage to a CMS database. These data support the results of SYNTAX, BARI 2D and other trials, in that higher-risk patients likely benefit from CABG vs. PCI. It is important to remember, however, that this is not a randomized clinical trial and not a direct comparison of PCI to CABG. Patients were referred for a given revascularization procedure based upon physician and patient preference; they were not randomized. Additionally, there are other outcomes of importance to patients that strongly affect quality of life, such as stroke. It will be important to be able to review the full peer-reviewed manuscript when it is published.

– R. David Anderson, MD, MS, FACC, FSCAI
Director of Interventional Cardiology
University of Florida, Shands HealthCare

Disclosure: Dr. Anderson reports no relevant financial disclosures.

Follow CardiologyToday.com on Twitter.