UK Registry: No Relationship Between Hospital PCI Volume and Early Mortality


LONDON, England—Hospital PCI volume in the United Kingdom does not appear to affect 30-day mortality, according to results of a registry study presented September 1, 2015, at the European Society of Cardiology Congress.

Another View: UK Registry: No Relationship Between Hospital PCI Volume and Early Mortality

For the analysis, Darragh O'Neill, PhD, of University College London (London, England), and colleagues culled data on 427,467 PCI cases (22% primary PCI) performed between 2007 and 2013 at 93 hospitals in England and Wales recorded by the British Cardiovascular Intervention Society. Hospitals were classified by annual volume into 6 groups, although most fell between 200 and 2,000 cases (median 650) over the study period.

Overall mortality at 30 days was 1.9%, and 4.8% for primary PCI. Mortality increased from 1.4% to 2.2% over the study period, “suggesting that [hospitals] were taking on higher-risk patients,” Dr. O’Neill said, adding that this trend was not seen in the primary PCI subset.

STEMI cases were less prevalent in lower- compared with higher-volume centers (8.4% vs 24.7%), although lower-volume centers were more likely to perform PCI for STEMI with cardiogenic shock (8.4% vs 3.7%).

In a crude analysis, “there was a surprising increase in the number of deaths that occur in higher- volume vs lower-volume [centers],” Dr. O’Neill said. But after adjustment for age, sex, clinical characteristics, and year of procedure, there was no relationship between hospital PCI volume and risk-adjusted mortality across all 6 volume categories (0-199 through ≥ 2,000 cases). The same lack of relationship was seen in an analysis restricted to primary PCI cases.

“When we interpret these results in light of a priori evidence, we find no support for the notion that there is a volume/outcome relationship at the hospital-level in this English and Welsh dataset,” Dr. O’Neill observed.

Low Volume Might Not Correlate with Low Operator Experience

Dr. O’Neill suggested that operator-level experience might have an impact on procedural outcomes but added that those data were not available at the time the current analysis was conducted.

Session comoderator Petr Widimský, MD, DSc, of Charles University (Prague, Czech Republic), commented that previous US studies “found some relation to center volume but they didn’t find any relationship to operator volume.

“I would [have expected] you would find the same,” he continued, adding that the low-volume operators are often the most experienced.

Dr. O’Neill concluded by acknowledging that “the results presented here are based on a healthcare system that has already undergone important reconfiguration over the past decade and therefore… need to be interpreted in light of that.” In addition, he noted, “the volume range that we captured may not be representative of volumes performed in other healthcare systems.”


Source: 
O'Neill D. Percutaneous coronary intervention center volume and 30-day mortality: a contemporary national cohort study of 427,467 cases. Presented at: European Society of Cardiology Congress; September 1, 2015; London, England.

Related Stories:

Disclosures
  • Dr. O'Neill reports no relevant conflicts of interest.

Comments