A new model may help predict mortality risk after transcatheter aortic valve replacement (TAVR).
Researchers used data from the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry (TVTR) to develop a model to predict in-hospital mortality risk after TAVR. Data were included for 13,718 consecutive U.S. patients who had TAVR from Nov. 1, 2011, to Feb. 28, 2014. The model was validated by using records from 6,868 consecutive patients who had TAVR from March 1 to Oct. 8, 2014. Expert opinion and statistical analysis were used to select covariates. The results of the study were published online by JAMA Cardiology on March 9.
Of the development sample, 13,672 of 13,718 patients had data available, and of these, 6,680 (48.9%) were men and 6,992 (51.5%) were women; mean age was 82.1 years. In the validation cohort, 3,554 (51.7%) were men and 3,314 (48.3%) were women; mean age was 81.6 years. Patients who had TAVR were generally considered to be unsuitable candidates for or at extreme risk with surgical aortic valve replacement and were offered TAVR only if they were expected to survive for at least 1 year. Seven hundred thirty patients (5.3%) died in the hospital after TAVR.
In the final model, the characteristics associated with in-hospital death were age (odds ratio [OR], 1.13; 95% CI, 1.06 to 1.20); glomerular filtration rate per 5-U increments (OR, 0.93; 95% CI, 0.91 to 0.95); hemodialysis (OR, 3.25; 95% CI, 2.42 to 4.37); New York Heart Association functional class IV (OR, 1.25; 95% CI, 1.03 to 1.52); severe chronic lung disease (OR, 1.67; 95% CI, 1.35 to 2.05); nonfemoral access site (OR, 1.96; 95% CI, 1.65 to 2.33); and procedural acuity categories 2 (OR, 1.57; 95% CI, 1.20 to 2.05), 3 (OR, 2.70; 95% CI, 2.05 to 3.55), and 4 (OR, 3.34; 95% CI, 1.59 to 7.02). Procedural acuity was defined as urgency of the procedure as determined by the patient's clinical state. The C-statistic for discrimination was 0.67 in the development group and 0.66 in the validation group.
Among other limitations, the study authors noted that they could not adjust for all potential risk factors, such as frailty and quality of life. They also noted that the model should not be the only measure used to determine which patients receive TAVR but should be used together with such factors as history, physical examination, laboratory data, and clinical judgment. The model, they wrote, “should be a valuable adjunct for patient counseling, performance assessment, local quality improvement, and national monitoring of the appropriateness of patient selection for TAVR.”
The authors of an accompanying invited commentary reiterated the model's limitations and also pointed out that it is unclear how expert opinion informed model development. “In its present iteration, the TVTR risk score provides a tool to identify individual patients with elevated in-hospital mortality,” the commentary authors wrote. “To improve its accuracy and relevance, future iterations of the risk score should stem from more complete ascertainment of patient and site factors and incorporate data-driven findings.” They also noted that models predicting postdischarge survival would be helpful.