Even after 3 years of follow-up, redo transcatheter aortic valve replacement (TAVR) performs about as well as the first procedure, whether compared for hard endpoints, such as death and stroke, or for softer endpoints, such as function and quality of life, new registry data suggest.
reported Rajendra Makkar, MD, associate director, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles.
The results were presented at the annual meeting of the European Association of Percutaneous Cardiovascular Interventions.
Data for this analysis were drawn from 348,338 TAVR procedures with the Edwards balloon-expandable valves in the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Replacement Registry.
Of these, 1,216 were redo procedures. In 475 of the cases, the redo was performed in a patient whose first procedure was with an Edwards device. In the remaining 741 cases, the Edwards device replaced a different prosthetic heart valve. The median time to the redo from the first procedure was 26 months.
For the analysis, the redo-TAVRs were compared with native TAVR patients through 1:1 propensity matching employing 35 covariates, such as age, body mass index (BMI), baseline comorbidities, prior cardiovascular procedures, valve size, and Society of Thoracic Surgeons risk score.
Low death and stroke rates following TAVR redos
The rates of all-cause death or stroke within hospital (4.7% vs. 3.9%; P = .32) and at 30 days (6.1% vs. 5.9%; P = .77) were numerically but not statistically higher in the redo group.
At 1 year, the rates of death (17.3% vs. 17.7%; P = .961) and stroke (3.3% vs. 3.5%; P = .982) were numerically but not significantly lower among those who underwent a redo procedure.
The secondary endpoints told the same story. The one exception was the higher aortic valve reintervention rate (0.61% vs. 0.09%; P = .03) at 30 days in the redo group. This did reach statistical significance, but Dr. Makkar pointed out rates were very low regardless. The rates climbed in both groups by 1 year (1.09% vs. 0.21%; P = .01).
No other secondary endpoints differed significantly at 30 days or at 1 year. Even though some were numerically higher after redo at 1 year, such as major vascular complications (1.25 vs. 1.60; P = .51), others were lower, such as new-start dialysis (1.62 vs. 0.98; P = .26). All-cause readmission rates at 1 year were nearly identical (32.56% vs. 32.23%; P = .82).
Consistent with the comparable outcomes on the hard endpoints, major and similar improvements were seen in both the redo and the propensity-matched native TAVR patients on the Kansas City Cardiomyopathy Questionnaire Overall Summary. The slight advantage for the redo group was not significant at 30 days, but the degree of improvement was greater after the redo than after native TAVR at 1 year (15% vs. 10%; P = .03).
“You bring good news,” said Alain G. Cribier, MD, director of cardiology, Charles Nicolle Hospital, University of Rouen, France. Widely regarded as the father of TAVR for his first-in-human series in 2002, Dr. Cribier said that there are several reassuring take-home messages from this study.
“First, these data tell us that the redo rate is extremely low,” he said, noting that the registry data suggests a risk well below 1%. “Second, we are seeing from this data that there are no more complications [than TAVR in a native valve] if you need to do this.”