MILWAUKEE – Less invasive lower-extremity endovascular interventions do not reduce hospital readmissions among patients with peripheral artery disease, according to an analysis of the Cerner Health Facts database.
The 30-day readmission rate was 13.9% for patients who underwent open surgery and 15.3% for those who had an endovascular procedure.
Lead author Dr. Todd Vogel expressed surprise that the two approaches were relatively equal, adding that, "I thought with endo, we’re doing less, they’d come back more."
The common practice of staging lower-limb endovascular interventions is creating concerns that use of hospital readmissions as a quality outcome measure for reimbursement may not accurately identify planned readmissions or quality of care.
Session moderator Dr. Patrick Geraghty said in an interview that lower-extremity intervention outcomes "are probably the most complex and difficult to define outcomes issue for all of vascular surgery," as compared with carotid and aortic aneurysms, and that this is already reflected in efforts proposed by the National Surgical Quality Improvement Program (NSQIP) and the Centers for Medicare and Medicaid Services (CMS).
"Lower-extremity readmission is going to be a real hot-button discussion because we already know it’s substantial," he said. "If I do a stent graft for someone with leg ischemia and the flow improves and they go home on post-op day 1, and I bring them back 10 days later for a planned debridement of a toe ulcer that we’d been looking at, was that bad? Was that poor care, something I should be penalized for?
"Or was it just good care, but it didn’t fit into CMS’s box of everything should be done within one admission and that any readmission is therefore bad?"
The current analysis is unique in that utilizes electronic medical record (EMR) data to provide real-world outcomes for lower-limb interventions, said Dr. Geraghty, a vascular surgeon with Barnes-Jewish Hospital in St. Louis.
"I think we’re seeing here maybe the first fruits of good EMR design, and it’s a prod for surgeons to look into EMR design and ask whether we can design EMR notes for vascular follow-up in the ER such that we pull good EMR data over great numbers of patients," he said at the annual meeting of the Midwestern Vascular Surgical Society.
Dr. Vogel said that the Cerner database is not as population based as Medicare, capturing observational patient EMR data on more than 84 million admissions and ambulatory visits at roughly 187 participating hospitals, albeit primarily urban. Cerner is the second largest EMR in the United States after Epic.
The analysis encompassed 1,458 elective first admissions with a diagnosis of peripheral artery disease (PAD) undergoing a lower-extremity procedure from October 2008 to December 2010. Of these, 777 had open surgery and 681 an endovascular procedure.
Intermittent claudication was the most common indication for any procedure, present in 56.2% of open and 43.8% of endovascular patients.
The overall readmission rate at 30 days was surprisingly high at 14.5%, and was also unexpectedly high for for those with claudication, at 10.2% in the open and 11.3% in the endovascular group, said Dr. Vogel, chief of vascular surgery at the University of Missouri Hospitals and Clinics, Columbia.
"The frightening number to think about is that, in the claudicant group, we have a 10% readmission rate within 30 days," he said, noting that rates were very similar between groups. "So that’s a number we should all begin to think about."
As expected, readmission rates in the open and endovascular groups increased with disease severity. Rates for rest pain and gangrene were 14% vs. 18.2%, and 22% vs. 24%, respectively.
In bivariate analysis, blacks were significantly more likely to be readmitted 30 days after discharge (odds ratio, 1.56), as were patients discharged to a skilled nursing facility or nursing home (OR, 2.59), he said.
There was a nonsignificant trend toward higher readmissions at teaching hospitals (OR, 1.20), while a hospital stay of more than 7 days was a strong, significant predictor of 30-day readmission (OR, 2.54).
Readmissions also were increased in patients with a Charlson Comorbidity Index score of 3-5 (OR, 1.56) or score of 6-10 (OR, 1.90), diabetes (OR, 1.41), or sepsis (OR, 2.99), he said.
The risk of 30-day readmission was increased more than fivefold among patients with poor liver function, as indicated by total bilirubin levels greater than 2 mg/dL (OR, 5.15) or AST over 100 U/L (OR, 5.56). Risk was also more than twofold higher among patients with renal disease, as indicated by hemoglobin (nadir) less than 8 g/dL (OR, 2.17) and serum creatinine of at least 2 mg/dL (OR, 2.07), as well as those dispensed a staggering 30 medications or more (OR, 2.63), Dr. Vogel reported.