Publications

BACKGROUND: Ineligibility for surgical revascularization is increasingly prevalent and associated with increased mortality after percutaneous coronary intervention (PCI). High-quality, contemporary, multicenter data regarding clinical outcomes after PCI is scarce and poses a barrier to clinical decision-making for surgically ineligible patients. The aim of this study was to describe and compare the clinical characteristics, institutional variation, and longitudinal outcomes of PCI among surgically eligible and ineligible patients in the Veterans Affairs Healthcare System.

METHODS: Patients with left main and/or multivessel coronary artery disease undergoing index PCI between October 1, 2017 and September 30, 2022 were identified and the prevalence of surgical ineligibility determined by review of the electronic medical record. The association between surgical ineligibility and mortality and major adverse cardiovascular events (MACE; death, myocardial infarction, repeat revascularization, and stroke) was assessed.

RESULTS: A total of 6192 patients with left main and/or multivessel coronary artery disease (842 surgically ineligible and 5350 surgically eligible) underwent PCI during the study period. After adjustment, surgical ineligibility was associated with a significantly decreased time to mortality (time ratio, 0.801 [95% CI, 0.662-0.970]) over a median 1045-day (interquartile range, 583-1600) follow-up period, though not associated with composite MACE (time ratio, 0.859 [95% CI, 0.685-1.078]). After adjustment for target lesion characteristics and procedural complexity, the association between surgical ineligibility and mortality was attenuated (time ratio, 0.842 [95% CI, 0.688-1.030]).

CONCLUSIONS: Ineligibility for surgical revascularization was associated with increased risk of long-term mortality after PCI. The risk of adverse outcomes after PCI, however, was similar among surgically eligible and ineligible patients after adjusting for measured comorbidities, coronary anatomic features, and procedural complexity.

BACKGROUND: There may be variability in willingness to perform percutaneous coronary intervention (PCI) in higher-risk patients who present with ST-segment-elevation myocardial infarction (STEMI). We sought to describe current treatment selection patterns and hospital-level variability.

METHODS AND RESULTS: We identified patients presenting with STEMI with a culprit lesion on coronary angiography between January 1, 2019, and March 31, 2023, using the NCDR (National Cardiovascular Data Registry) CPMI (Chest Pain-Myocardial Infarction) registry. We compared patient-level characteristics of patients who did and did not undergo PCI at each hospital. There were 178 984 patients from 582 US hospitals presenting with STEMI who were included. Among patients with STEMI and a culprit lesion, 6180 did not undergo PCI (3.5%). Patients with a presentation of STEMI and a culprit lesion who did not undergo PCI were older (67 [interquartile range, 58-76]) years versus 62 ([interquartile range, 54-71] years, P<0.001), more likely to present with heart failure (15.0% versus 7.4%, P<0.001), and more likely to have cardiac arrest before arrival (9.7% versus 5.1%, P<0.001) than patients who underwent PCI. Patients who did not undergo PCI had higher predicted mortality rates (12.5%±17.9% versus 6.5%±11.5%, P<0.001) and observed mortality rates (21.7% versus 6.4%, P<0.001) compared with patients who underwent PCI.

CONCLUSIONS: There is variability in the percentage of patients with culprit lesions on invasive coronary angiography undergoing PCI for STEMI, with 3.5% of patients with STEMI not receiving PCI overall, and >5% of patients not undergoing PCI in a quarter of US hospitals. Differences in observed versus predicted mortality rates for patients who did or did not undergo PCI may highlight the effects of risk-avoidant behavior.

BACKGROUND: Catheter ablation is an effective therapy for ventricular tachycardia (VT) and is increasing in use. Assessment of contemporary real-world outcomes of VT ablation requires data inclusive of both inpatient and outpatient encounters.

OBJECTIVE: We aimed to assess factors associated with 1-year in-hospital mortality, all-cause readmission, and recurrent readmission for VT after VT ablation along with the associated costs of care.

METHODS: Inpatient and outpatient VT ablations were captured in the Healthcare Cost and Utilization Project data in Florida, Maryland, and New York from 2016 to 2020 with 1-year follow-up through 2021. Cox proportional hazards regression was used to identify risk factors associated with 1-year in-hospital mortality, all-cause readmission, and recurrent VT readmission. Costs of inpatient cases and readmission were captured.

RESULTS: Of 3899 patients undergoing VT ablation, 2% died during the initial episode of care. The 1-year in-hospital mortality rate, all-cause readmission rate, and recurrent VT readmission rate were 6.8%, 43.4%, and 16.9%, respectively, and were broadly stable during the study period. Coronary artery disease, chronic kidney disease, and heart failure were independently associated with an increased risk of in-hospital mortality and all-cause readmission within 1 year (P < .05). Recurrent VT was the most common cause of readmission. All-cause readmissions increased costs by 55%.

CONCLUSION: Mortality, long-term readmission, and recurrent VT remain high after VT ablation and include measurably important costs. Strategies to improve freedom from recurrent VT while managing comorbid conditions may serve as targets for improving the efficacy and cost-effectiveness of an important procedure applied to a vulnerable population.

BACKGROUND: Although Medicare Advantage (MA) plans provide coverage to >50% of Medicare beneficiaries, it is unclear whether MA claims can be used similarly to Medicare Fee-For-Service (FFS) claims for clinical outcomes assessment. In this study, we evaluate the accuracy of claims algorithms previously validated in FFS to assess comorbidities and outcomes in MA patients after aortic valve replacement.

METHODS: We compared the concordance of 11 claims-based covariates (diabetes, hypertension, atrial flutter/fibrillation, myocardial infarction) and outcomes (stroke, disabling stroke, transient ischemic attack, major vascular complication, bleeding, permanent pacemaker implantation, death) among FFS and MA patients with the covariates and adjudicated outcomes in the multinational Evolut Low-Risk Trial (2016-2018). We used claims algorithms for 1-year outcomes and calculated sensitivity, specificity, positive predictive value, negative predictive value, and kappa, using adjudicated outcomes as the reference. We compared the kappa for MA versus FFS using the 2-sample z-test with a significance level of P<0.05.

RESULTS: Among 1139 US patients aged 65+ years old in the Evolut Low-Risk Trial, 782 patients (175 MA and 607 FFS) were linked to claims data and had complete comorbidity data. Among all covariates, claims algorithms for covariates had sensitivities ≥85% for identifying diabetes, atrial flutter/fibrillation, and hypertension in MA and FFS. For the outcomes, sensitivities were ≥85% for bleeding (comprehensive), permanent pacemaker implantation, and death. The kappa was higher in MA versus FFS for diabetes (P=0.03) and hypertension (P=0.025) but was lower in myocardial infarction (P<0.0001). There was no statistically significant difference in the kappa agreement between MA versus FFS for any of the selected outcomes.

CONCLUSIONS: Medicare claims have a similar level of kappa agreement in MA versus FFS for most covariates and outcomes. As patients shift to MA, ascertainment of outcomes using Medicare claims in postapproval studies remains valid for select outcomes.

The objective of this study was to measure the contemporary patency rates and frequency of interventions required for arteriovenous fistula (AVF) care in a representative US population of patients with end-stage kidney disease, including by age, race, and gender. All Medicare beneficiaries aged >20 years who underwent AVF graft creation for end-stage kidney disease between 2017 and 2019 were included for analysis. The primary end points included primary patency, primary assisted patency, postintervention patency, and fistula functionality up to 1 year after AVF placement. The secondary end point included admission for an associated adverse event after AVF creation. Multivariate analysis of patency rates was also assessed. Of 43,457 patients included in the analysis, the cumulative primary patency at 90 days was 68.4% and at 1 year, 31.5%. At 1 year, the primary assisted patency rate, postintervention patency, and fistula use were 70.4%, 30.2%, and 59.1%, respectively. There was no difference in primary patency rates when comparing age groups (age 40 to 59 years: hazard ratio [HR] 1.01, 95% confidence interval [CI] 0.95 to 1.06, p = 0.84 or age ≥60 years: HR 0.99, 95% CI 0.93 to 1.04, p = 0.61) with the reference of age group 20 to 39 years. Women were at greater risk of experiencing primary patency failure than were men (HR 1.16, 95% CI 1.14 to 1.20, p <0.001), and Black patients were at greater risk of experiencing primary patency failure than were White patients (HR 1.34, 95% CI 1.31 to 1.38, p <0.001). The cumulative incidence of admissions for adverse events was 32.6% at 1 year. In conclusion, our findings suggest that the real-world AVF patency rates remain low, with disproportionately low rates in women and Black patients.

BACKGROUND: Many research investigations for pulmonary embolism (PE) rely on the International Classification of Diseases 10th Revision (ICD-10) codes for analyses of electronic databases. The validity of ICD-10 codes in identifying PE remains uncertain.

OBJECTIVES: The objective of this study was to validate an algorithm to efficiently identify pulmonary embolism using ICD-10 codes.

METHODS: Using a prespecified protocol, patients in the Mass General-Brigham hospitals (2016-2021) with ICD-10 principal discharge codes for PE, those with secondary codes for PE, and those without PE codes were identified (n = 578 from each group). Weighting was applied to represent each group proportionate to their true prevalence. The accuracy of ICD-10 codes for identifying PE was compared with adjudication by independent physicians. The F1 score, which incorporates sensitivity and positive predictive value (PPV), was assessed. Subset validation was performed at Yale-New Haven Health System.

RESULTS: A total of 1712 patients were included (age: 60.6 years; 52.3% female). ICD-10 PE codes in the principal discharge position had sensitivity and PPV of 58.3% and 92.1%, respectively. Adding secondary discharge codes to the principal discharge codes improved the sensitivity to 83.2%, but the PPV was reduced to 79.1%. Using a combination of ICD-10 PE principal discharge codes or secondary codes plus imaging codes for PE led to sensitivity and PPV of 81.6% and 84.7%, respectively, and the highest F1 score (83.1%; P < .001 compared with other methods). Validation yielded largely similar results.

CONCLUSION: Although the principal discharge codes for PE show excellent PPV, they miss 40% of acute PEs. A combination of principal discharge codes and secondary codes plus PE imaging codes led to improved sensitivity without severe reduction in PPV.