NCDR Data Show Relationship Between Volume, Outcomes in Primary PCI for STEMI

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Data representing about 85% of all cardiac catheterization laboratories performing primary percutaneous coronary intervention (PCI) in the United States indicate that the volume of procedures performed annually at a given institution continues to serve as a predictor of mortality, according to a study published online November 5, 2013, ahead of print in Circulation: Cardiovascular Quality and Outcomes. The study backs up current guideline recommendations.

Michael C. Kontos, MD, of Virginia Commonwealth University (Richmond, VA), and colleagues analyzed data on 86,044 patients from 738 hospitals who underwent primary PCI from July 2006 through June 2009 and were included in the CathPCI Registry.

Hospitals were separated into 3 groups by volume:

  • Low (≤ 36 primary PCIs per year; n = 278)Intermediate (36-60 primary PCIs per year; n = 236) High (> 60 primary PCIs per year; n = 224)

Lower Volume, Higher Mortality

Low-volume hospitals performed fewer total PCIs and were less likely to have on-site cardiac surgery backup. High-volume hospitals were more likely to be located in urban areas and to be private/community hospitals rather than academic centers.

Patients in high-volume hospitals had shorter median door-to-balloon (D2B) times and a greater percentage achieved a D2B time of 90 minutes or less compared with patients in low-volume hospitals (72% vs. 66%; P < 0.0001). There were no differences in D2B times between intermediate- and high-volume hospitals.

Unadjusted in-hospital mortality was higher in low-volume vs. high-volume hospitals (5.6% vs. 4.8%; P < 0.001). This difference was maintained after multivariate analysis. After further adjustment for D2B time and in sensitivity analysis factoring in transfer patients, mortality remained different for low- vs. high-volume centers. In contrast, mortality was not different between intermediate- and high-volume hospitals (table 1).

Table 1. Mortality Risk vs. High Volume Centers


OR (95% CI)

P Value

Low-volume Centers
Unadjusted Mortality
Adjustment for D2B Times
Transfer Patients

1.17 (1.07-1.27)1.20 (1.08-1.33)
1.14 (1.005-1.26)

< 0.0001
< 0.04

Intermediate-volume Centers
Unadjusted Mortality
Adjustment for D2B Times

0.99 (0.93–1.07)
1.02 (0.94-1.11)


In-hospital mortality also was higher at low- vs. high-volume hospitals in patients presenting during off hours (5.6% vs. 4.9%; P = 0.012), but no significant difference was seen in patients presenting with cardiogenic shock (29% vs. 28%; P = 0.31).

After multivariate adjustment, low-volume hospitals were less likely to achieve a D2B time of ≤ 90 minutes compared with high-volume hospitals for all patient subgroups (P = 0.02).

Consistent with Current Guidelines

The study authors note that current guidelines from the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force recommend that primary PCI for STEMI be undertaken at hospitals that perform more than 200 elective PCIs per year and more than 36 primary PCI procedures for STEMI per year.

“The relationship between volume and outcomes seems to be continuous,” Dr. Kontos and colleagues write. “Our data indicate that current guideline-recommended volume numbers seem reasonable as hospitals with < 36 primary PCIs/year had significantly higher mortality than those with ≥ 36/year. The lack of a mortality difference between centers with a primary PCI volume of 30 to 60/year and those > 60/year suggests that selection of a higher volume requirement may not lead to significant differences in mortality.”


Kontos MC, Wang Y, Chaudhry SI, et al. Lower hospital volume is associated with higher in-hospital mortality in patients undergoing primary percutaneous coronary intervention for ST-segment-elevation myocardial infarction: A report from the NCDR. Circ Cardiovasc Qual Outcomes. 2013;Epub ahead of print.



Related Stories:

  • Dr. Kontos reports no relevant conflicts of interest.

We Recommend