Marlene Busko

May 22, 2017

NEW ORLEANS, LA — Hospitals that performed 100 or more transcatheter aortic-valve replacement (TAVR) procedures in a year (high-volume centers) had a 25% lower rate of hospital readmissions at 30 days compared with other hospitals, in new research[1].

"We report for the first time, to our knowledge, an inverse association between hospital TAVR volume and 30-day readmissions," Dr Sahil Khera (New York Medical College, Valhalla) and colleagues write.

Their study was presented on May 11, 2017 as a "Best of the Best" abstract at the Society for Cardiovascular Angiography and Interventions (SCAI) 2017 Scientific Sessions and simultaneously published online in JAMA Cardiology.

"Examining practices at high-volume hospitals or other hospitals identified to have lower-than-expected readmission rates may provide systemic approaches to improve care for TAVR patients," senior author Dr J Dawn Abbott (Brown University, Providence, RA) told heartwire from Medscape in an email.

"In all centers . . . improvements can be made to reduce readmissions from congestive heart failure [by] paying more attention to diuretic and antihypertension medication dosing, employing heart-failure transition teams, and close follow-up," she suggested. Noncardiac readmissions could be reduced by paying closer attention to general anesthesia and indwelling lines and catheters.

In an accompanying editorial[2], Dr John D Carroll (University of Colorado, Denver) cautions that these data do not identify the root causes for readmission but adds that all TAVR centers should strive to improve. "For low-volume and higher-volume [TAVR] programs, the most constructive approach is to promote best practices, analyze outcomes, including readmission rates, compare them with national benchmarks, and take actions that are likely to improve performance."

"It's reasonable for all centers performing TAVR to assess cardiac and noncardiac comorbidities," Dr Sunil V Rao (Durham VA Medical Center, NC), who was not involved in this study, agreed in an email to heartwire .

Similarly, in an accompanying editor's note[3], Dr Patrick T O'Gara (Brigham and Women's Hospital, Boston, MA) writes that "despite the controversies surrounding the readmission rate metric, scrutiny of processes designed to optimize transitions of care should be promoted."

Observational Study in 129 Hospitals

Now that TAVR is approved for patients with severe symptomatic aortic stenosis at intermediate surgical risk, the volume of TAVR procedures is expected to rise exponentially, according to Khera and colleagues.

To examine the association between hospital TAVR volume and 30-day patient readmissions, they identified 129 US hospitals in the Nationwide Readmission Database that had established TAVR programs.

Each hospital had performed at least five TAVRS in the first 3 months of 2014, for a total of 16,252 TAVR procedures in 2014.

The researchers divided the hospitals into three categories:

  • High-volume centers with >100 TAVR/year (62 hospitals).

  • Medium-volume centers with >50 to <100 TAVR/year (47 hospitals).

  • Low-volume centers with <50 TAVR/year (20 hospitals).

More than three-quarters of the TAVR procedures were performed in high-volume centers (77%), and fewer were done in medium-volume centers (19%) or low-volume centers (4%).

The mean 30-day readmission rate was lower in high-volume hospitals (15.6%) and similar in medium-volume hospitals (19.0%) and low-volume hospitals (19.5%).

The adjusted odds of 30-day readmission after TAVR was much lower in high-volume centers than in medium- of low-volume centers.

30-Day Readmission, High- vs Lower-Volume TAVR Hospitals

TAVR volume Adjusted OR (95% CI) P
High vs medium 0.76 (0.68–0.85) <0.001
High vs low 0.75 (0.60–0.92) 0.007

The readmissions occurred at median of 9 days after surgery.

More Noncardiac Readmissions at Low-Volume Hospitals

Low-volume-TAVR hospitals had more readmissions for noncardiac causes than medium- or high-volume hospitals (66% vs 60% vs 61% ), but these differences were not statistically significant (P=0.36)

Readmissions for infection or respiratory, endocrine/metabolic, renal, or trauma problems were more common in low-volume hospitals, whereas readmissions for gastrointestinal issues and TIA/stroke were more common in high- or medium-volume hospitals.

In addition, readmissions for heart failure, arrhythmias, and conduction disorders were more common in medium- and high-volume hospitals, whereas readmissions for coronary artery disease, chest pain, and syncope were more common in low-volume hospitals.

Low-volume hospitals also had more readmissions that were not related to TAVR—that is, not related to cardiac causes, bleeding, TIA/stroke, acute kidney injury, postoperative infections, or infections due to cardiac/vascular device—but this difference was also not statistically significant.

There were also no statistical differences between high- and lower-volume TAVR centers for the average hospital-readmission stay (5.5 days to 6 days) or cost ($13,400 to $14,100).

However high-volume centers had "substantially reduced" healthcare expenditures for readmissions, Khera and colleagues note.

Patients who had TAVR at low-volume hospitals had more comorbidities, and the procedure was more often done using a transapical rather than endovascular approach.

"Potentially, the drivers of the observed inverse hospital volume–30-day readmissions association are patient comorbidities, TAVR-approach selection, proficiency in [endovascular] TAVRs by more experienced operators at higher-volume hospitals (operator factors), and better postoperative care coordination at higher-volume hospitals," the researchers suggest.

Associations Do Not Prove Causation, But Can Help Guide Policy

However, all of these associations do not prove causation, Carroll cautions.

Similarly, Rao observed that "while these are very important findings," it is important to keep in mind that unmeasured confounders may be driving readmission.

The authors do acknowledge that they lacked information such as valve type, echocardiographic variables, postprocedural paravalvular leak, patient risk scores, medication use, or Society of Thoracic Surgeons scores.

"TAVR patients tend to have a lot of comorbid conditions, but how exactly readmissions can be reduced is not known," Rao noted. "There are no strategies that have been proven to reduce readmissions in this patient population."

"A robust and validated model for [reducing] readmission rates after TAVR has not been created," Carroll echoes.

In a companion paper[4], the researchers identified that a longer index hospital stay, acute kidney injury, greater comorbidities, transapical TAVR, chronic kidney disease, chronic lung disease, and discharge to a nursing facility all predicted readmission, he noted.

"I think this paper and the prior paper by this group support both the development of a risk model for readmission and the study of approaches to reduce readmission among TAVR patients," said Rao.

In the meantime, the authors hope that "as new TAVR programs open across the country, these data will guide policymakers to identify targets for optimizing and standardizing TAVR outcomes across hospitals."

Khera and Abbott had no relevant financial relationships. Disclosures for the coauthors are listed in the paper. Carroll is a member of the Steering Committee of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. O'Gara had no disclosures.

For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.