Preoperative risk score for mortality within 3 years of nonemergent endovascular repair for descending thoracic aortic aneurysm.

Abstract

The purpose of this study was to create a validated risk score for mortality within 3 years of elective endovascular repair (TEVAR) of descending thoracic aortic aneurysm (TAA) based on variables existing at the time of preoperative clinical presentation. The Vascular Quality Initiative was queried for nonemergent TEVAR for degenerative, non-dissection-related descending TAAs limited to aortic landing zones 2 to 5. The inclusion criteria were met by 3750 patients. Patients were divided randomly into a 70% testing cohort for risk score model creation and a 30% internal validation cohort to verify the function of the risk score. Binary logistic regression analysis was performed for the outcome of 3-year mortality using variables that achieved a univariable P value of ≤.05. Using this regression, it was determined which variables have a multivariable association for the outcomes as defined by a regression P value of ≤.05. A risk score was then created for the primary outcome. Variables with a multivariable a P value of ≤.05 from the above-mentioned regression were included in the risk score and weighted based on their respective regression beta-coefficient in a point scale. Variables with a beta-coefficient of <0.25 were assigned 0 points, and then a point was added for each rise in beta-coefficient at 0.25 intervals. Significant (P < .05) multivariable association with 3-year mortality was noted for advancing age decade (P < .001), nursing home status (adjusted odds ratio [aOR], 2.63; P = .029), body mass index of <20 kg/m² (aOR, 1.74; P = .008), active smoking status (aOR, 1.44; P = .023), 6.0 to 6.9 cm TAA diameter (aOR, 1.40; P = .015 relative to <6 cm), ≥7 cm TAA diameter (aOR, 1.97; P < .001 relative to <6 cm), chronic obstructive pulmonary disease (COPD) on medication (aOR, 1.42; P = .029), COPD on oxygen (aOR, 2.29; P < .001), renal insufficiency (aOR, 1.72; P < .001), anemia (aOR, 1.74; P = .001), and lack of statin medication preoperatively (aOR, 1.54; P = .001). There is noted to be significantly upsloping mortality with statistically significant escalation between each risk score bundle. Patients with risk scores of <3 experienced 3-year mortality at just a 2.3% rate, whereas patients with risk scores of ≥13 had a mean 32.8% mortality at 3 years (OR, 21.1; P < .001). There was outstanding agreement between the testing and validation groups at all risk score bundles with nearly identical mortality rates and no statistically significant difference at each level. Area under curve analysis revealed 0.7 and 0.67 for the testing and validation cohorts, respectively. Hosmer-Lemeshow goodness of fit for the source testing regression was 88.2% overall accuracy and for the validation cohort regression was 87.7%. A risk score for 3-year mortality after TEVAR for non-dissection-related descending TAA has been created which has both good accuracy and outstanding internal Vascular Quality Initiative validation. The most impactful deleterious variables toward survival included advancing age, COPD, aneurysm diameter, anemia, renal insufficiency, tobacco use, lack of statin therapy and nursing home status.