Readmission Rate After Carotid Revascularization an Opportunity for Improvement
Nearly 1 in 10 Medicare patients who undergo carotid revascularization returns to the hospital within 30 days, according to an observational study published in the April 14, 2015, issue of the Journal of the American College of Cardiology. Moreover, the risk is higher for those who receive stenting compared with surgery.
The study aims to raise “awareness among clinicians, hospital administrators, and policymakers regarding this issue and [illustrate] the extent to which readmission rates differ by the type of revascularization procedure,” write Jeptha P. Curtis, MD, of the Yale University School of Medicine (New Haven, CT), and colleagues. They say that further research is needed to inform strategies for reducing readmission after carotid revascularization.
The investigators used Medicare fee-for-service administrative claims data to identify 168,323 patients (mean age 76.3 years; 56.4% male; 93.2% Caucasian) who underwent 180,059 carotid interventions (6.8% had more than 1) at 2,287 hospitals from 2009 through 2011. More than three-quarters (81.5%) of the procedures were endarterectomies.
Of the entire cohort, 9.7% were symptomatic; diabetes, hypertension, and congestive heart failure were present in 43.3%, 89.3%, and 19.7%, respectively. In addition, 43.1% had other peripheral vascular diseases and 17.6% had renal failure.
Reasons for Readmission Vary Broadly
The unadjusted 30-day readmission rate was 9.6%, with 44.3% of instances occurring within 7 days of discharge. By 30 days, 1.2% of patients had died.
Reasons for readmission varied widely, with the top 10 accounting for only half of cases. The most common causes were complications of care (8.6%), heart failure (6.6%), and pneumonia (5.2%), while cerebral complications including ischemic stroke, TIA, and hemorrhage collectively accounted for 10.7% of cases. In addition, readmissions occurring during the first week after discharge were more likely to stem from complications of care, heart failure, and acute stroke, while those occurring in the third and fourth weeks arose more often from peripheral vascular diseases, chronic angina, and CAD.
Predictors of readmission were:
- Being symptomatic: adjusted OR 1.18 (95% CI 1.12-1.24)
- Female sex: adjusted OR 1.12 (95% CI 1.08-1.15)
- Non-Caucasian race: adjusted OR 1.12 (95% CI 1.06-1.19)
- Age over 65 years: adjusted OR 1.01 (95% CI 1.01-1.02)
CAS Patients More Likely to Be Readmitted
Patients who underwent carotid endarterectomy (CEA) were more likely to be male and Caucasian and to have diabetes, hypertension, and CAD than those who received stents (P < .001 for all). In contrast, carotid artery stenting (CAS) patients more often were symptomatic and had congestive heart failure, ACS, other peripheral vascular diseases, and chronic renal failure (P < .001 for all).
On multivariate analysis, the risk of 30-day readmission was higher for the CAS group compared with the CEA group (OR 1.13; 95% CI 1.08-1.18). The same relationship held true in a propensity-matched cohort (OR 1.18; 95% CI 1.07-1.23) and regardless of symptomatic status, age, sex, and race.
The median hospital 30-day risk-standardized readmission rate was 9.5% (range 7.5%-12.5%). Analysis of the 1,500 centers that performed more than 25 carotid revascularizations showed that risk-standardized readmission rates were similar regardless of their proportion of percutaneous procedures (P = .771).
Public Reporting Is Coming
According to the authors, the Centers for Medicare & Medicaid Services (CMS) has announced plans to publicly report 30-day readmission rates for patients undergoing vascular procedures. In the future, they note, hospitals with higher-than-expected rates may be subject to payment penalties. In this light, the current data provide “a benchmark [by] which hospitals can evaluate efforts to reduce readmission rates in this population,” they observe.
Dr. Curtis and colleagues note that “almost one-third of readmission diagnoses were potentially due to procedural complications…, [which] may represent high-yield areas for targeted efforts to reduce readmission.”
Nonetheless, they say, the diversity of reasons for readmission suggests that, rather than targeting specific diagnoses or time periods, “general strategies leading to improvement in discharge planning, medication reconciliation, and early follow-up after discharge may better address underlying vulnerabilities in the transitional care process.”
Importantly, Dr. Curtis and colleagues point out, the absence of a correlation between readmission rates and a center’s proportional performance of stenting vs surgery “suggests that hospitals using CAS more frequently will not necessarily be disadvantaged if and when measures of vascular readmission are publicly reported and included in payment programs.”
The researchers acknowledge that administrative claims data have limitations in that they:
- Are subject to coding errors and coding differences across providers and institutions
- May not be able to differentiate between planned and unplanned readmissions
- Do not provide detailed information about the procedures
Is the Difference Between CAS and CEA Meaningful?
“Readmissions are increasingly being monitored as a benchmark of both the quality of and the costs to the health care system,” Barry T. Katzen, MD, of the Miami Cardiac and Vascular Institute (Miami, FL), writes in an accompanying editorial. “As such, identifying diseases and procedures that are associated with significant readmission rates allows physicians and administrators to focus attention on identifying variables and creating opportunities for improvement in the management of these patients.”
Dr. Katzen cautions against concluding that CAS imparts higher readmission risk than CEA, in large part because there may have been important clinical differences between the patients who received the 2 therapies despite propensity matching.
In a telephone interview with TCTMD, William A. Gray, MD, of Columbia University Medical Center (New York, NY), said the almost 10% readmission rate was surprisingly high and did not jibe with his experience.
He agreed with the editorial that the apparently poorer showing by CAS was likely due to selection bias, with CAS performed in patients who are at high surgical risk and thus typically sicker than those who undergo surgery. “There are clearly [differences between these groups] that the researchers cannot adjust for,” he observed.
Moreover, he questioned whether the small difference in readmission rates between CAS and CEA was meaningful or potentially an artifact of a study with a very large sample size.
Broader Context Needed to Judge Readmissions
Dr. Gray noted that the 30-day stroke and death outcomes in postmarket CAS registries are significantly better than in Medicare administrative databases. The reason, he suggested, is that the registries included only 300 to 400 vetted and monitored sites, while Medicare data draws from far more hospitals with little or no monitoring.
“The issue becomes: What’s really representative of quality care?” he commented. Dr. Gray said that he and his fellow CAS investigators have “argued very strongly to CMS” that coverage should be restricted to a limited number of sites with appropriate data reporting, certification, and accreditation.
The authors make a similar proposal, suggesting that this strategy “could have the extra benefit of facilitating the transition of care from the inpatient to the outpatient setting.”
Some kind of benchmark for readmission rates is warranted, Dr. Gray acknowledged, because readmission is a “sentinel event” related to cost and quality of care. “But in isolation, it’s hard to say how meaningful [the observed 10%] rate is. You have to contextualize it,” he stressed. “Let’s find out how this stacks up against other procedure-based readmission rates.
“This kind of study is valuable,” Dr. Gray concluded. “But I would like not to see [any benchmark] become punitive before we get a chance to improve quality.”
1. Al-Damluji MS, Dharmarajan K, Zhang W, et al. Readmissions after carotid artery revascularization in the Medicare population. J Am Coll Cardiol. 2015;65:1398-1408.
2. Katzen BT. Readmissions after carotid artery revascularization in the Medicare population: a word of caution [editorial]. J Am Coll Cardiol. 2015;65:1409-1410.
- Dr. Curtis reports receiving support from CMS and holding stock options in Medtronic.
- Dr. Katzen reports serving on the scientific advisory boards for Boston Scientific, Medtronic Vascular, and WL Gore.
- Dr. Gray reports no relevant conflicts of interest.