Higher volumes tied to better outcomes in endovascular therapy for stroke
Operator and hospital volume were associated with lower rates of inpatient mortality and better discharge disposition among patients who underwent endovascular therapy for acute ischemic stroke, according to data presented at the International Stroke Conference.
Laura K. Stein, MD, assistant professor of neurology at Icahn School of Medicine at Mount Sinai, and colleagues analyzed 2016-2017 Medicare data to determine whether total cases predicted inpatient mortality, good disposition at discharge and 30-day readmissions.
“As it stands now, physicians at centers seeking endovascular certification from the Joint Commission must have performed 15 cases in the preceding 12 months or 30 in the preceding 24 months,” Stein said during a presentation. “We hypothesized that for endovascular thrombectomy, a volume threshold associated with better outcomes can be identified by analysis of national Medicare data.”
The analysis included 13,311 patients treated by 2,754 operators at 641 hospitals. The mean number of cases per operator was 4.8 (range, 1-82; standard deviation, 7.7) and the mean number of cases per hospital was 20.8 (range, 1-160; standard deviation, 23.6).
Among the cohort, 56% were discharged with good disposition and 16% were readmitted within 30 days, Stein said.
Regarding hospital volume, at a threshold of 28 cases, there were significantly lower odds of inpatient mortality, according to Stein. “This odds ratio becomes even lower and more favorable at each successive case volume threshold,” she said.
For example, Stein said, hospitals with at least 25 cases had 13% lower odds of inpatient mortality than hospitals with fewer than 25 cases (OR = 0.87; 95%, 0.79-0.97) while hospitals with at least 55 cases had 23% lower odds of inpatient mortality than hospitals with fewer than 55 cases (OR = 0.77; 95%, 0.69-0.86).
Odds of good disposition at discharge became significant at a threshold of 34 cases per hospital, Stein said. “Again, odds of good outcome increase with every successive case volume threshold,” she said.
Hospitals with at least 25 cases had 10% higher odds of good disposition at discharge than hospitals with fewer than 25 cases (OR = 1.1; 95%, 1.02-1.19) while hospitals with at least 55 cases had 13% higher odds of good disposition at discharge than hospitals with fewer than 55 cases (OR = 1.13; 95%, 1.04-1.22), she said.
Regarding operator volume, odds of inpatient mortality became lower at a threshold of five cases (OR = 0.86; 95% CI, 0.77-0.97), Stein said, noting that odds improved at higher volumes (OR for 35 cases vs. < 35 cases = 0.65; 95% CI, 0.55-0.77).
Odds of good disposition at discharge became significant at an operator volume of 22 cases, “with a successive increase in odds of a good outcome with each additional proceduralist case,” she said.
Extrapolating the results to all insurers, the hospital 1-year case volume threshold was 24 for lower mortality odds and 29 for higher odds of good disposition at discharge, whereas the operator 1-year case volume threshold was four for lower mortality odds and 19 for higher odds of good disposition at discharge, Stein said.
There was no relationship between operator or hospital volume and 30-day readmission rates, according to the researchers.
“These data demonstrate a linear relationship between endovascular thrombectomy case volume and outcomes of inpatient mortality and good outcome,” Stein said during the presentation. “They suggest that volume should considered in stroke center certification. I want to make it very clear that we in no way mean to suggest that we should be limiting access because of volume anywhere in this country.” – by Erik Swain
Stein LK, et al. LB11. Presented at: International Stroke Conference; Feb. 19-21, 2020; Los Angeles.
Disclosure: Stein reports no relevant financial disclosures. Two other authors report financial ties with various device companies.