No Link Between Operator Volume and Post-PCI Mortality in the UK
The registry study indicates that mortality may not be a good measure of quality. But it’s not clear what would be better.
In the current era in which PCI has been made safer with improvements in drugs, technologies, techniques, and patient selection, the annual volume of procedures performed by individual operators is not associated with 30-day mortality, an analysis out of the United Kingdom demonstrates.
That raises questions about the best metrics for assessing quality and about the relevance of recommended volume thresholds for maintenance of competency, according to senior author Mamas Mamas, BMBCh, DPhil (Royal Stoke Hospital, Stoke-on-Trent, England).
“I think it blows the discussion wide open,” he told TCTMD, noting that guidelines recommend performing at least 50 cases (in the United States) or 75 cases (in Europe) each year to maintain competency. In this study, splitting high- and low-volume operators with either of those cutoffs did not show a significant difference in 30-day mortality.
The discussion, then, needs to move beyond mortality—which is “a very insensitive measure of quality,” particularly considering that PCI is much safer now than it was a decade ago—to other potential metrics, such as the overall complication rate, the degree of incomplete revascularization, unplanned readmissions, or patient-reported outcomes, Mamas urged.
As for the recommended volume thresholds, Mamas said, “I don’t think they are appropriate in the UK. I can’t say whether that’s the case in the US, but certainly we don’t find any data that we should be having a minimum of 75 cases per year.”
Need for Newer Data
Although it makes intuitive sense that interventional cardiologists performing a greater number of procedures would be better at them, Mamas said, prior studies assessing volume-outcome relationships have provided mixed results. Many of those analysis were conducted before 2010, however, and there are few data from more contemporary, large, national data sets.
Mamas, along with lead author William Hulme, MSc (University of Manchester, England), and colleagues, aimed to help fill that gap with their study, which was published online March 22, 2018, ahead of print in the European Heart Journal. The analysis, drawing on the British Cardiovascular Intervention Society registry, included 133,970 procedures performed in England and Wales in 2013 and 2014. Median operator volume across all procedures was 178 per year, with an average of 124 per year.
The 30-day mortality rate was 2.6%, ranging from 2.9% in the lowest-volume operators to 2.5% in the highest-volume operators. After adjustment for case mix, however, there was no relationship between volume and mortality (OR per 100 procedures 0.99; 95% CI 0.93-1.05). Additional analyses confined to high-risk patient subsets, looking at in-hospital mortality or MACE, or evaluating dichotomized high- and low-volume groups based on recommended thresholds also failed to show a significant relationship between volume and outcomes.
The investigators note that two other studies have examined volume-outcome association in contemporary practice. One using the Japanese PCI registry showed no relationship between operator volume and in-hospital mortality or periprocedural complications. In contrast, another using the CathPCI Registry of the American College of Cardiology’s National Cardiovascular Data Registry showed a small increase in in-hospital mortality for each 50-case decrease in annual volume. Because radial access was used in only 15.2% of procedures in the US registry, the relevance to UK practice is limited, the researchers say.
“Our study adds a European perspective, reflecting a recent PCI era and showing that increasing volume is not associated with better mortality outcomes, and also that the majority of operators have caseloads exceeding previously defined volume thresholds,” they write. Only 8.9% and 16.9% of operators in the current study performed less than 50 and 75 cases, respectively, per year.
Mamas said he doesn’t think the findings would necessarily apply to countries like the United States, where a higher proportion of operators perform low volumes of procedures based on those cutoffs.
‘Not a Lot of Science’ Backing a Volume Threshold
In comments to TCTMD, Bonnie Weiner, MD (Saint Vincent Hospital, Worcester, MA), echoed Mamas’s skepticism about using mortality as a quality metric, pointing out that many in the interventional cardiology community have been wary of doing so due to concerns about risk adjustment. Moreover, she added, “event rates are sufficiently low these days that being able to demonstrate any difference between operators is almost impossible.”
Other metrics besides mortality, such as bleeding or the need for repeat interventions, might be better at differentiating between operators, she said, but evidence is lacking.
There are issues, too, with associating procedural volume with outcomes because high- and low-volume operators are likely taking on different types of cases, a phenomenon that cannot be fully accounted for with statistical adjustments, said Weiner, a past president of the Society for Cardiovascular Angiography and Interventions.
Asked whether volume thresholds should still be recommended for maintenance of competency, Weiner said that’s a tricky question, indicating that most people would want a family member to undergo PCI by an operator performing a certain number of cases each year. But what that number is remains unclear, she said, noting that the 50-case threshold recommended in the United States was largely derived through consensus.
“There’s not a lot of science supporting that number,” Weiner said, adding that late-career operators who have already performed tens of thousands of cases may be able to maintain competency while performing even less than that on an annual basis. “But again, I don’t think we have the data to support any different number.”
Where Do We Go From Here?
To get to a volume threshold with more scientific backing, Weiner said the community needs to focus on getting more information on the acute complications of PCI beyond mortality, including bleeding, MI, and stroke.
Mamas agreed that these types of complications, particularly those occurring after discharge, need to be captured better by existing registries so researchers can start to assess volume-outcome relationships that are not focused just on mortality.
He added that he’d like to see more discussions in healthcare systems like that in the United States about whether having high proportions of low-volume operators is appropriate and in the best interest of patients. “I think those sorts of discussions, whilst uncomfortable, are discussions that really need to be had,” Mamas said.
In terms of what can be done right now to maintain high-quality PCI at the local level, Weiner said the importance of review from an operator’s peers cannot be overestimated. By keeping track of case selection and outcomes, this process will ensure that outliers are identified and addressed.
Hulme W, Sperrin M, Curzen N, et al. Operator volume is not associated with mortality following percutaneous coronary intervention: insights from the British Cardiovascular Intervention Society registry. Eur Heart J. 2018;Epub ahead of print.
- Hulme, Mamas, and Weiner report no relevant conflicts of interest.