Gil Metser

BACKGROUND: Artificial intelligence, particularly deep learning (DL), has immense potential to improve the interpretation of transthoracic echocardiography (TTE). Mitral regurgitation (MR) is the most common valvular heart disease and presents unique challenges for DL, including the integration of multiple video-level assessments into a final study-level classification. METHODS: A novel DL system was developed to intake complete TTEs, identify color MR Doppler videos, and determine MR severity on a 4-step ordinal scale (none/trace, mild, moderate, and severe) using the reading cardiologist as a reference standard. This DL system was tested in internal and external test sets with performance assessed by agreement with the reading cardiologist, weighted κ, and area under the receiver-operating characteristic curve for binary classification of both moderate or greater and severe MR. In addition to the primary 4-step model, a 6-step MR assessment model was studied with the addition of the intermediate MR classes of mild-moderate and moderate-severe with performance assessed by both exact agreement and ±1 step agreement with the clinical MR interpretation. RESULTS: A total of 61 689 TTEs were split into train (n=43 811), validation (n=8891), and internal test (n=8987) sets with an additional external test set of 8208 TTEs. The model had high performance in MR classification in internal (exact accuracy, 82%; κ=0.84; area under the receiver-operating characteristic curve, 0.98 for moderate or greater MR) and external test sets (exact accuracy, 79%; κ=0.80; area under the receiver-operating characteristic curve, 0.98 for moderate or greater MR). Most (63% internal and 66% external) misclassification disagreements were between none/trace and mild MR. MR classification accuracy was slightly higher using multiple TTE views (accuracy, 82%) than with only apical 4-chamber views (accuracy, 80%). In subset analyses, the model was accurate in the classification of both primary and secondary MR with slightly lower performance in cases of eccentric MR. In the analysis of the 6-step classification system, the exact accuracy was 80% and 76% with a ±1 step agreement of 99% and 98% in the internal and external test set, respectively. CONCLUSIONS: This end-to-end DL system can intake entire echocardiogram studies to accurately classify MR severity and may be useful in helping clinicians refine MR assessments.

Purpose: To investigate the clinical implication of additional below-the-ankle (BTA) intervention in patients with chronic limb-threatening ischemia (CLTI) undergoing below-the-knee (BTK) intervention. Materials and Methods: A sub-analysis was performed using data from the LIBERTY trial ( ClinicalTrials.gov identifier NCT01855412), a prospective, observational, core-laboratory adjudicated, multicenter study of endovascular intervention in 1204 patients. Patients with CLTI (Rutherford Classification 4-6) who underwent BTK intervention were included in this sub-analysis. Participants were then stratified into 2 treatment groups according to whether at least one lesion intervened on was BTA (n=66) or not (n=273). The decision on whether and where to intervene was made during the procedure. The main outcome measures included major amputation, target vessel revascularization (TVR), major adverse events (MAE), survival, amputation-free survival, major adverse limb events or peri-operative death (MALE-POD), and all-cause death. Other outcome measures included procedural success, procedural complications, and wound healing rate. Results: There were no differences in procedural success or severe angiographic complications between the 2 groups. At 1-year post-procedure, patients in the BTK group had a higher rate of freedom from major amputation (95.0% vs. 86.9%, respectively; HR: 2.87, 95% CI: 1.17-7.03), a higher rate of freedom from TVR (80.1% vs. 66.9%, respectively; HR: 1.94, 95% CI: 1.14-3.32), a higher rate of freedom from MALE-POD (94.6% vs. 86.9%, respectively; HR: 2.65, 95% CI: 1.10-6.41), and a higher rate of freedom from MAE at both 1 (76.0% vs. 60.1%, respectively; HR: 2.00, 95% CI: 1.24-3.22) and 3 years post procedure (67.5% vs. 55.8%, respectively; HR: 1.69, 95% CI: 1.08-2.65). There was a significantly lower rate of survival in the BTK group at 3 years (74.3% vs. 91.1%, respectively; HR: 0.35, 95% CI: 0.14-0.87). After risk adjustment, there was a higher rate of all-cause death in the BTK group at 3 years (19.4% vs. 9.1%, respectively; p=0.023) post-intervention. Conclusion: Patients with disease requiring intervention to BTA lesions have a potential increased amputation rate in the short term, but BTA intervention carries a potential survival benefit in the long term when compared to BTK intervention alone.

Background: Telemedicine visit use vastly expanded during the COVID-19 pandemic, and this has had an uncertain impact on cardiovascular care quality. Objective: We sought to examine the association between telemedicine visits and the failure to meet the Controlling High Blood Pressure (BP) quality measure from the Centers for Medicare & Medicaid Services. Methods: This was a retrospective cohort study of 32,727 adult patients with hypertension who were seen in primary care and cardiology clinics at an urban, academic medical center from February to December 2020. The primary outcome was a failure to meet the Controlling High Blood Pressure quality measure, which was defined as having no BP recorded or having a last recorded BP of ≥140/90 mm Hg (ie, poor BP control). Multivariable logistic regression was used to assess the association between telemedicine visit use during the study period (none, 1 telemedicine visit, or ≥2 telemedicine visits) and poor BP control; we adjusted for demographic and clinical characteristics. Results: During the study period, no BP was recorded for 2.3% (486/20,745) of patients with in-person visits only, 27.1% (1863/6878) of patients with 1 telemedicine visit, and 25% (1277/5104) of patients with ≥2 telemedicine visits. After adjustment, telemedicine use was associated with poor BP control (1 telemedicine visit: odds ratio [OR] 2.06, 95% CI 1.94-2.18; P<.001; ≥2 telemedicine visits: OR 2.49, 95% CI 2.31-2.68; P<.001; reference: in-person visits only). This effect disappeared when the analysis was restricted to patients with at least 1 recorded BP (1 telemedicine visit: OR 0.89, 95% CI 0.83-0.95; P=.001; ≥2 telemedicine visits: OR 0.91, 95% CI 0.83-0.99; P=.03). Conclusions: Increased telemedicine visit use is associated with poorer performance on the Controlling High Blood Pressure quality measure. However, telemedicine visit use may not negatively impact BP control when BP is recorded.

BACKGROUND Telemedicine visit use vastly expanded during the COVID-19 pandemic, and this has had an uncertain impact on cardiovascular care quality. OBJECTIVE We sought to examine the association between telemedicine visits and the failure to meet the Controlling High Blood Pressure (BP) quality measure from the Centers for Medicare & Medicaid Services. METHODS This was a retrospective cohort study of 32,727 adult patients with hypertension who were seen in primary care and cardiology clinics at an urban, academic medical center from February to December 2020. The primary outcome was a failure to meet the Controlling High Blood Pressure quality measure, which was defined as having no BP recorded or having a last recorded BP of ≥140/90 mm Hg (ie, poor BP control). Multivariable logistic regression was used to assess the association between telemedicine visit use during the study period (none, 1 telemedicine visit, or ≥2 telemedicine visits) and poor BP control; we adjusted for demographic and clinical characteristics. RESULTS During the study period, no BP was recorded for 2.3% (486/20,745) of patients with in-person visits only, 27.1% (1863/6878) of patients with 1 telemedicine visit, and 25% (1277/5104) of patients with ≥2 telemedicine visits. After adjustment, telemedicine use was associated with poor BP control (1 telemedicine visit: odds ratio [OR] 2.06, 95% CI 1.94-2.18; P<.001; ≥2 telemedicine visits: OR 2.49, 95% CI 2.31-2.68; P<.001; reference: in-person visits only). This effect disappeared when the analysis was restricted to patients with at least 1 recorded BP (1 telemedicine visit: OR 0.89, 95% CI 0.83-0.95; P=.001; ≥2 telemedicine visits: OR 0.91, 95% CI 0.83-0.99; P=.03). CONCLUSIONS Increased telemedicine visit use is associated with poorer performance on the Controlling High Blood Pressure quality measure. However, telemedicine visit use may not negatively impact BP control when BP is recorded.