Feature: Inappropriate PCI use varies, patient selection needs improvement
 |
“Measurement of hospital quality has traditionally focused on processes of care and post procedure outcomes,” Bradley and colleagues from the VA Eastern Colorado HealthCare System in Denver, wrote.
The study, published online May 10 in Circulation, looked at data from 203,531 patients who underwent elective PCI at 779 hospitals and were included in the National Cardiovascular Data Registry between July 2009 and April 2011. The researchers aimed to better understand the association between patient selection for PCI and postprocedural outcomes.
“Appropriate use criteria [AUC] are evolving quality metrics in PCI,” Bradley said. “The question becomes whether or not this new and emerging quality metric is in any way related to traditional quality metrics of post-procedural outcomes such as inhospital mortality, bleeding complications and processes of care such as provisional guidelines directed medications.”
During the study, Bradley et al studied the association between a hospital’s proportion of nonacute PCIs that were deemed “inappropriate” by 2009 AUC standards for revascularization and other factors such as in-hospital mortality, bleeding complications and medical therapy post-PCI.
The researchers reported that 50 percent of PCIs during the study were deemed “appropriate,” while 12.1 percent were deemed “inappropriate.” Of the total procedures, 35.5 percent were classified as “uncertain.” The median proportion of PCIs classified as inappropriate was 10.9 percent, ranging from 0.0 percent to 58.6 percent. The median number PCIs classified in the lowest hospital tertile was 5.3 percent. This proportion in the highest tertile was recorded at 20 percent.
Bradley et al reported that 453 patients died in the hospital, 3,699 experienced periprocedural bleeds and 173,847 patients were discharged on medical therapy post-procedure.
“We found that a hospital’s tertile of inappropriate PCI was not related to in-hospital mortality or provisions of guideline-driven medications,” Bradley offered. “This suggests that the appropriate use criteria or appropriateness of PCI and traditional quality metrics measures are independent aspects of PCI quality.” However, he added that both of these metrics systems should be used to ensure quality PCI at a facility.
“It’s important to understand that both measures are needed to inform the quality of PCI,” he added. During the trial, the researchers found a broad variation among hospitals in the proportion of inappropriate PCI. Bradley added that these findings suggest that there is much room for improvement in terms of nonacute PCI in the clinical setting, including improving patient selection.
While this variation may pinpoint the need for better patient selection, Bradley offered that “the process of patient selection is not related to how well the procedure was performed.”
The authors noted that the measurement of PCI appropriateness and postprocedural outcomes remain important metrics to help depict PCI quality. Bradley said that patient selection is unrelated to the processes that ensure that patients receive high quality procedures such as bleeding avoidance strategies, renal protection measures and pathways post-PCI that help patients receive the most appropriate medications.
Additionally, Bradley and colleagues said that hospitals should utilize decision-making tools to help improve care. He noted that using decision-support tools prior to the patient arriving at the cardiac cath lab rather than after, could help improve care.
“Furthermore, proper patient selection to avoid inappropriate PCI may be disincentivized by monetary reimbursement, referral structures and the expectation of colleagues,” they wrote.