Merging machine learning and patient preference: a novel tool for risk prediction of percutaneous coronary interventions.

BACKGROUND AND AIMS: Predicting personalized risk for adverse events following percutaneous coronary intervention (PCI) remains critical in weighing treatment options, employing risk mitigation strategies, and enhancing shared decision-making. This study aimed to employ machine learning models using pre-procedural variables to accurately predict common post-PCI complications. METHODS: A group of 66 adults underwent a semiquantitative survey assessing a preferred list of outcomes and model display. The machine learning cohort included 107 793 patients undergoing PCI procedures performed at 48 hospitals in Michigan between 1 April 2018 and 31 December 2021 in the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) registry separated into training and validation cohorts. External validation was conducted in the Cardiac Care Outcomes Assessment Program database of 56 583 procedures in 33 hospitals in Washington. RESULTS: Overall rate of in-hospital mortality was 1.85% (n = 1999), acute kidney injury 2.51% (n = 2519), new-onset dialysis 0.44% (n = 462), stroke 0.41% (n = 447), major bleeding 0.89% (n = 942), and transfusion 2.41% (n = 2592). The model demonstrated robust discrimination and calibration for mortality {area under the receiver-operating characteristic curve [AUC]: 0.930 [95% confidence interval (CI) 0.920-0.940]}, acute kidney injury [AUC: 0.893 (95% CI 0.883-0.903)], dialysis [AUC: 0.951 (95% CI 0.939-0.964)], stroke [AUC: 0.751 (95%CI 0.714-0.787)], transfusion [AUC: 0.917 (95% CI 0.907-0.925)], and major bleeding [AUC: 0.887 (95% CI 0.870-0.905)]. Similar discrimination was noted in the external validation population. Survey subjects preferred a comprehensive list of individually reported post-procedure outcomes. CONCLUSIONS: Using common pre-procedural risk factors, the BMC2 machine learning models accurately predict post-PCI outcomes. Utilizing patient feedback, the BMC2 models employ a patient-centred tool to clearly display risks to patients and providers (https://shiny.bmc2.org/pci-prediction/). Enhanced risk prediction prior to PCI could help inform treatment selection and shared decision-making discussions.

Statewide Initiative to Reduce Patient Radiation Doses During Percutaneous Coronary Intervention.

BACKGROUND: Improved radiation safety practices are needed across hospitals performing percutaneous coronary intervention (PCI). This study was performed to assess the temporal trend in PCI radiation doses concurrent with the conduct of a statewide radiation safety initiative. METHODS: A statewide initiative to reduce PCI radiation doses was conducted in Michigan between 2017 and 2021 and included focused radiation safety education, reporting of institutional radiation doses, and implementation of radiation performance metrics for hospitals. Using data from a large statewide registry, PCI discharges between July 1, 2016, and July 1, 2022, having a procedural air kerma (AK) recorded were analyzed for temporal trends. A multivariable regression analysis was performed to determine whether declines in procedural AK over time were attributable to changes in known predictors of radiation doses. RESULTS: Among 131 619 PCI procedures performed during the study period, a reduction in procedural AK was observed over time, from a median dose of 1.46 (0.86-2.37) Gy in the first year of the study to 0.97 (0.56-1.64) Gy in the last year of the study (P<0.001). The proportion of cases with an AK ≥5 Gy declined from 4.24% to 0.86% over the same time period (P<0.0001). After adjusting for variables known to impact radiation doses, a 1-year increase in the date of PCI was associated with a 7.61% (95% CI, 7.38%-7.84%) reduction in procedural AK (P<0.0001). CONCLUSIONS: Concurrent with the conduct of a statewide initiative to reduce procedural radiation doses, a progressive and significant decline in procedural radiation doses was observed among patients undergoing PCI in the state of Michigan.

Contemporary Trends and Outcomes of Intravascular Lithotripsy in Percutaneous Coronary Intervention: Insights From BMC2.

BACKGROUND: With an aging population and an increase in the comorbidity burden of patients undergoing percutaneous coronary intervention (PCI), the management of coronary calcification for optimal PCI is critical in contemporary practice. OBJECTIVES: This study sought to examine the trends and outcomes of coronary intravascular lithotripsy (IVL), rotational/orbital atherectomy, or both among patients who underwent PCI in Michigan. METHODS: We included all PCIs between January 1, 2021, and June 30, 2022, performed at 48 Michigan hospitals. Outcomes included in-hospital major adverse cardiac events (MACEs) and procedural success. RESULTS: IVL was used in 1,090 patients (2.57%), atherectomy was used in 1,743 (4.10%) patients, and both were used in 240 patients (0.57% of all PCIs). IVL use increased from 0.04% of PCI cases in January 2021 to 4.28% of cases in June 2022, ultimately exceeding the rate of atherectomy use. The rate of MACEs (4.3% vs 5.4%; P = 0.23) and procedural success (89.4% vs 89.1%; P = 0.88) were similar among patients treated with IVL compared with atherectomy, respectively. Only 15.6% of patients treated with IVL in contemporary practice were similar to the population enrolled in the pivotal IVL trials. Among such patients (n = 169), the rate of MACEs (0.0%) and procedural success (94.7%) were similar to the outcomes reported in the pivotal IVL trials. CONCLUSIONS: Since its introduction in February 2021, coronary IVL use has steadily increased, exceeding atherectomy use in Michigan by February 2022. Contemporary use of IVL and atherectomy is generally associated with high rates of procedural success and low rates of complications.

Cause and preventability of in-hospital mortality after PCI: A statewide root-cause analysis of 1,163 deaths.

BACKGROUND: Mortality is the most devastating complication of percutaneous coronary interventions (PCI). Identifying the most common causes and mechanisms of death after PCI in contemporary practice is an important step in further reducing periprocedural mortality. OBJECTIVES: To systematically analyze the cause and circumstances of in-hospital mortality in a large, multi-center, statewide cohort. METHODS: In-hospital deaths after PCI occurring at 39 hospitals included in the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) between 2012 and 2014 were retrospectively reviewed using validated methods. A priori PCI-related mortality risk was estimated using the validated BMC2 model. RESULTS: A total of 1,163 deaths after PCI were included in the study. Mean age was 71±13 years, and 507 (44%) were women. Left ventricular failure was the most common cause of death (52% of cases). The circumstance of death was most commonly related to prior acute cardiovascular condition (61% of cases). Procedural complications were considered contributing to mortality in 235 (20%) cases. Death was rated as not preventable or slightly preventable in 1,045 (89.9%) cases. The majority of the deaths occurred in intermediate or high-risk patients, but 328 (28.2%) deaths occurred in low-risk patients (<5% predicted risk of mortality). PCI was considered rarely appropriate in 30% of preventable deaths. CONCLUSIONS: In-hospital mortality after PCI is rare, and primarily related to pre-existing critical acute cardiovascular condition. However, approximately 10% of deaths were preventable. Further research is needed to characterize preventable deaths, in order to develop strategies to improve procedural safety.

Quality of Life After Transcatheter Aortic Valve Replacement in Normal-Flow, Low-Gradient Aortic Stenosis.

Background: Prior studies of aortic valve replacement (AVR) in patients with normal-flow, low-gradient aortic stenosis (NF-LG AS) have demonstrated conflicting results regarding the survival benefit of AVR. Changes in quality of life (QoL) after transcatheter AVR (TAVR) have not been reported in this population. Objectives: The purpose of this study was to compare changes in QoL after TAVR for patients with NF-LG AS to patients with high-gradient aortic stenosis (HG-AS). Methods: Patients who underwent TAVR for severe aortic stenosis (AS) were divided into 4 hemodynamic profiles of AS, including NF-LG AS. Changes in Kansas City Cardiomyopathy Questionnaire-12 score from baseline to 1 year were compared between AS groups. The primary composite outcome indicating clinical improvement consisted of survival to 1 year and improved Kansas City Cardiomyopathy Questionnaire overall summary score of ≥5 points while adjusting for relevant baseline factors. Results: Out of 860 patients who underwent TAVR, high gradient AS was present in 368 (42.8%) patients and NF-LG AS in 245 (28.5%). HG-AS and NF-LG AS groups had a similar proportion of patients who met the primary unadjusted outcome of clinical improvement (70.4% vs 63.9%, respectively; P = 0.189). One-year Kaplan-Meier mortality estimates were higher for NF-LG AS patients than HG-AS patients (12.9% vs 5.8%, P < 0.001). In the primary adjusted analysis, there was no significant difference in the composite outcome between HG and NF-LG AS groups (adjusted OR: 0.72, 95% CI: 0.47-1.11). Conclusions: Selected patients with NF-LG AS experienced similar improvement in QoL after TAVR compared with HG-AS. Further investigation of patients with NF-LG AS will help to inform optimal selection for treatment with TAVR.

Renal Function-Based Contrast Threshold Predicts Kidney Injury in Transcatheter Aortic Valve Replacement.

Background: Acute kidney injury (AKI) after contrast-guided interventions is associated with adverse outcomes, but the role of contrast in the context of renal function is less well described for patients undergoing transcatheter aortic valve replacement (TAVR). Methods: Patients from the Michigan TAVR registry between January 2016 and December 2019 were included. AKI was defined using Valve Academic Research Consortium 2 definitions. An integer cut point for the ratio of contrast volume (CV) to renal function (estimated glomerular filtration rate [eGFR]) as a predictor of AKI was calculated. Results: Of 7112 cases, AKI occurred in 629 (8.8%) patients. Unadjusted mortality was higher among patients with AKI (32.5% vs 9.0%, P â€‹< â€‹.0001). AKI remained significantly associated with the risk of mortality after multivariable adjustment (hazard ratio = 4.50, P â€‹< â€‹.001). Procedural characteristics associated with AKI included CV/eGFR >2 (adjusted odds ratio [aOR] = 1.36, P = .003, 95% CI = 1.10-1.67), CV/eGFR >3 (aOR = 1.38, P â€‹= â€‹.009, 95% CI = 1.09-1.77), and use of general anesthesia (aOR = 1.67, P â€‹< â€‹.0001, 95% CI = 1.38-2.03). Conclusions: CV in the context of renal function administrated during TAVR is a robust tool to predict AKI. AKI after TAVR is associated with an increased risk of mortality. Incorporation of thresholds of >2× and > 3× eGFR into procedural planning should be considered as a quality initiative.