Author + information
- Sorin J. Brener, MD∗ ()
- Department of Medicine, Division of Cardiology, NewYork-Presbyterian Brooklyn Methodist Hospital, Brooklyn, New York
- ↵∗Address for correspondence:
Dr. Sorin J. Brener, Department of Medicine, Division of Cardiology, Cardiac Catheterization Laboratory, NYP Brooklyn Methodist Hospital, 506 6th Street, Brooklyn, New York 11215.
More than 15 years after its introduction, transcatheter aortic valve replacement (TAVR) has become a mature procedure, with well-documented processes of screening, decision making, implementation, and follow-up (1,2). Since the end of 2011, when TAVR became commercially available in the United States, the American College of Cardiology and the Society of Thoracic Surgeons (STS) joined efforts to gather data on all TAVR procedures performed at more than 400 sites using the TVT (Transcatheter Valve Therapeutics) registry. TVT mandates reporting of all procedures, but has minimal auditing of input data, relying heavily on algorithms for identification of out-of-range values. It collects a large number of baseline, procedural, and follow-up variables, and provides site-specific reports on a periodic basis for in-hospital, 30-day, and 1-year outcomes.
In this issue of JACC: Cardiovascular Interventions, Kolte et al. (3) use this essentially administrative tool to report on the outcomes of patients undergoing urgent/emergent TAVR over 4.5 years and compare them with those undergoing elective TAVR. This is an important issue to address for a number of reasons. Among them are the question of whether TAVR is sufficiently reliable to “rescue” patient with severe deterioration of heart failure due to aortic stenosis, the dilemma of performing balloon aortic valvuloplasty as a temporizing measure versus “going for broke” with TAVR, and the uncertainty of whether urgent TAVR can help patients with other severe conditions from which recovery would be easier in the absence of severe aortic stenosis.
The investigators culled the records of over 72,000 patients but were able to use only ∼40,000 for this study, mostly because of age <65 years, lack of linkage with the Center for Medicaid & Medicare Services data, and ineligibility of patients with fee-for-service procedures (3). It is quite unclear to this editorialist why so many (n = ∼23,000) had to be excluded since their outcomes are documented in TVT even without the linkage to Center for Medicaid & Medicare Services or private insurance records. They found that 3,952 patients—approximately 1 in 10—had urgent TAVR (emergent indication was extremely rare). These patients, as expected, were much sicker than their elective counterparts in every measurable way. As a representative data point, their STS risk score for surgical aortic valve replacement was 11.8% (25th and 75th percentile: 7.6% to 17.9%) versus 6.1% (25th and 75th percentile: 4.1% to 9.1%) in elective patients. This range of STS scores places almost 75% of the urgent cohort in the high-risk and 35% in the inoperable category.
Gratifyingly, we learn though that the immediate results of urgent TAVR are quite good. Device success is >90%, and paravalvular aortic regurgitation, stroke rate, major bleeding, and need to convert to surgical AVR are similar to elective cases—all this despite a 2-fold higher rate of valve-in-valve procedures for degenerated aortic prostheses. After adjusting for all baseline and procedural characteristics (the authors appropriately performed a nonparsimonious multivariable model including all the variables we have data on including input methodology), the incidence of acute kidney injury and dialysis was also similar between the groups. The excess in mortality at 30 days and at 1 year (adjusted hazard ratio: 1.20, 95% confidence interval: 1.10 to 1.31; p < 0.001, and adjusted hazard ratio: 1.28, 95% confidence interval: 1.10 to 1.48; p = 0.001) is much smaller than one would expect from the great disparity in baseline risk. Many of the independent predictors of mortality at 30 days and at 1 year reflect chronic conditions rather than causes for the acute deterioration. Also, importantly, the observed mortality in both the urgent/emergent TAVR and in the elective TAVR groups was less than one-half the expected mortality, reaffirming the urgent need to update our risk scores for percutaneous AVR and maybe for surgical AVR, too.
Although technically speaking, the report deals with a mix of urgent and emergent TAVR, the latter was <1% of the total cohort (3), and thus, we are learning predominantly about what happens when urgent TAVR is performed. The definition for urgent used by TVT is relatively vague: need to perform TAVR before hospital discharge to prevent further deterioration in the patient’s condition. And yet, surprisingly, <10% of the urgent TAVR cohort received inotropes before procedure, and their average ejection fraction was 53%. Only 1% had a mechanical assist device placed at the start of the procedure and only one-quarter had moderate or severe aortic regurgitation. In fact, the latter 2 are predictors of improved survival, as they usually reflect prosthesis malfunction. There is no information on noncardiac comorbidity. These characteristics do not, in my opinion, go along with the need for urgent TAVR. It may be that logistical considerations—difficulty in discharge to home or to a facility and readmission for TAVR, outpatient testing that requires repeat visits to the hospital, or inability to ensure continuity of care (uncertainty if patient returns for TAVR)—were important considerations in the designation of urgent and the implantation of the new valve at the end of a stabilization period. If that is the case, the report describes more accurately what happens to recently destabilized and treated patients with severe aortic stenosis versus those undergoing more elective evaluation for TAVR as outpatients, rather the outcome of urgent TAVR as a solution for the very sick patients. The authors correctly identify the limitations of such an administrative and reporting tool, and the lack of identification of the exact reason for performing TAVR urgently.
It is interesting to note that 382 valve-in-valve procedures—292 of them planned—were performed in the urgent TAVR group, but only 202 patients had failed bioprostheses. This points again to the lack of auditing in this dataset.
The choice of valve in the urgent TAVR group was similar to the one seen in the overall population: nearly three-quarters of the implants were balloon-expandable valves. The valve choice seemed to independently predict early and late mortality in this group, a finding that is novel and not seen in any other comparison of the 2 systems. We are not told whether that holds true for the elective TAVR group or not. The authors attempt to explain these findings by the possible advantage of the balloon-expandable valves—faster deployment which is better tolerated and less paravalvular aortic regurgitation—but appropriately recognize that the choice of the device may be influenced by the severity of the hemodynamic status and the ability to withstand rapid pacing. It is unclear how often self-expanding valves were used for valve-in-valve procedures for which an appropriate size balloon-expandable valve may not exist.
In summary, this report (3), despite its limitations, is very important because it demonstrates that TAVR can be safely and effectively performed in patients having various reasons for being already hospitalized, without need for a separate hospitalization. The authors do not address the reimbursement implications of such a strategy, particularly important because the length of stay and the discharge rate to a nursing facility are markedly higher than in the elective population. I hope that this does not translate into a payer position that claims that “your urgency is not my urgency.”
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Brener has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Cribier A.,
- Eltchaninoff H.,
- Bash A.,
- et al.
- Kolte D.,
- Khera S.,
- Vemulapalli S.,
- et al.