A Boosted Ensemble Algorithm for Determination of Plaque Stability in High-Risk Patients on Coronary CTA

ObjectivesThis study sought to identify culprit lesion (CL) precursors among acute coronary syndrome (ACS) patients based on qualitative and quantitative computed tomography–based plaque characteristics.BackgroundCoronary computed tomography angiography (CTA) has been validated for patient-level prediction of ACS. However, the applicability of coronary CTA to CL assessment is not known.MethodsUtilizing the ICONIC (Incident COroNary Syndromes Identified by Computed Tomography) study, a nested case-control study of 468 patients with baseline coronary CTA, the study included ACS patients with invasive coronary angiography–adjudicated CLs that could be aligned to CL precursors on baseline coronary CTA. Separate blinded core laboratories adjudicated CLs and performed atherosclerotic plaque evaluation. Thereafter, the study used a boosted ensemble algorithm (XGBoost) to develop a predictive model of CLs. Data were randomly split into a training set (80%) and a test set (20%). The area under the receiver-operating characteristic curve of this model was compared with that of diameter stenosis (model 1), high-risk plaque features (model 2), and lesion-level features of CL precursors from the ICONIC study (model 3). Thereafter, the machine learning (ML) model was applied to 234 non-ACS patients with 864 lesions to determine model performance for CL exclusion.ResultsCL precursors were identified by both coronary angiography and baseline coronary CTA in 124 of 234 (53.0%) patients, with a total of 582 lesions (containing 124 CLs) included in the analysis. The ML model demonstrated significantly higher area under the receiver-operating characteristic curve for discriminating CL precursors (0.774; 95% confidence interval [CI]: 0.758 to 0.790) compared with model 1 (0.599; 95% CI: 0.599 to 0.599; p < 0.01), model 2 (0.532; 95% CI: 0.501 to 0.563; p < 0.01), and model 3 (0.672; 95% CI: 0.662 to 0.682; p < 0.01). When applied to the non-ACS cohort, the ML model had a specificity of 89.3% for excluding CLs.ConclusionsIn a high-risk cohort, a boosted ensemble algorithm can be used to predict CL from non-CL precursors on coronary CTA.     

The QT interval in clinical hypercalcemia

Corrected QT intervals were determined in 13 patients with severe, chronic hypercalcemia. The Q_OT_C interval was short in only 2 of 14 instances; Q_AT_C in 5 of 15 instances, and Q_ET_C in 5 of 16 instances. The correlations between serum calcium and the QT_C measurement were not significant when evaluating either linear or curvilinear (quadratic) relationships. Small and inconsistent changes were found when comparing the QT intervals before the development of the hypercalcemic episode, during hypercalcemia, or after successful treatment. We conclude that shortening of the QT interval is an unreliable index of clinical (chronic) hypercalcemia.     

The Utility of Rapid Atrial Pacing Immediately Post-TAVR to Predict the Need for Pacemaker Implantation

ObjectivesThe aim of this study was to determine the utility of rapid atrial pacing immediately after transcatheter aortic valve replacement (TAVR) to predict the need for permanent pacemaker implantation (PPI).BackgroundRisk stratification for patients without high-grade atrioventricular block (AVB) after TAVR is imprecise and based on anatomic considerations, electrocardiographic characteristics, and clinical suspicion. A more reliable assessment is necessary to minimize inpatient rhythm monitoring and/or reduce unnecessary PPI.MethodsConsecutive patients undergoing TAVR at 2 centers were included. After valve implantation in patients without pacemakers who did not have complete heart block or atrial fibrillation, the temporary pacemaker was withdrawn from the right ventricle and placed in the right atrium. Rapid atrial pacing was performed from 70 to 120 beats/min, and patients were assessed for the development of Wenckebach AVB. Patients were then followed for clinical outcomes, including PPI.ResultsA total of 284 patients were included. Of these, 130 (45.8%) developed Wenckebach AVB. There was a higher rate of PPI within 30 days of TAVR among the patients who developed Wenckebach AVB (13.1% vs. 1.3%; p < 0.001), with a negative predictive value for PPI in the group without Wenckebach AVB of 98.7%. A greater percentage of patients receiving self-expanding valves required PPI than those receiving a balloon-expandable valves (15.9% vs. 3.7%; p = 0.001), though these rates were still relatively low among patients who did not develop Wenckebach AVB (2.9% and 0.8%).ConclusionsAtrial pacing post-TAVR is easily performed and can help identify patients who may benefit from extended rhythm monitoring. Patients who did not develop pacing-induced Wenckebach AVB demonstrated an extremely low likelihood of PPI.     

Implementation of cancer prevention guidelines in clinical practice

Data from several sources, including consumer surveys, physician surveys, and medical record audits, indicate that consumers do not receive cancer screening tests as recommended by the National Cancer Institute, the American Cancer Society, and the U.S. Preventive Services Task Force. Performance rates are consistently below published standards for all tests except Pap tests. Major reasons physicians do not perform the recommended tests include physician forgetfulness, disagreement with recommendations, lack of time, and patient refusal. Physicians also tend to overestimate their own performance rates. Barriers to screening test performance can be categorized into patient factors, physician factors, test factors, and health care delivery system factors. Interventions, such as computerized reminder systems, physician audits with feedback, and patient education and reminders, can be effective in promoting performance of such screening. Interventions that target both physician and patient may be particularly effective. Presented at the conference, Frontiers in Disease Prevention, The Johns Hopkins University, June 5–6, 1989.     

Molecular screening of the glucokinase gene in familial Type 2 (non-insulin-dependent) diabetes mellitus

Summary The glucokinase locus has been implicated by linkage studies in several Caucasian pedigrees with early onset, autosomal dominant diabetes, and mutations have been identified in a large number of these pedigrees. Although mutations have been reported in some pedigrees with late onset Type 2 (non-insulin-dependent) diabetes mellitus, linkage studies of typical familial Type 2 diabetes did not suggest a major role for this locus. Nonetheless, linkage studies were consistent with the hypothesis that mutations of the glucokinase gene were responsible for the pathogenesis of Type 2 diabetes in a minority of pedigrees or one gene in a polygenic disorder. To systematically address this hypothesis, we examined 60 diabetic members of 18 pedigrees ascertained for two or more Type 2 diabetic siblings and eight unrelated diabetic spouses. Initially, the coding regions from each of the 11 glucokinase exons were examined by the sensitive technique of single strand conformation polymorphism analysis to screen for single nucleotide substitutions. Subsequently, we also sequenced each exon from an affected member of the single pedigree in which a glucokinase allele was most likely to segregate with diabetes. Single strand conformation polymorphism analysis detected only three variants, none of which altered the amino acid sequence. No coding or splice site mutations were detected. Likewise, no additional mutations were detected upon direct sequence analysis. However, additional screening of promoter and 3′ untranslated regions detected a variant pattern in the untranslated region of exon 10 which appeared to segregate with diabetes and impaired glucose tolerance in one pedigree. Sequence analysis demonstrated the deletion of a cytosine in exon 10 at position 906, but this deletion was not associated with Type 2 diabetes among unrelated spouses, was not linked to diabetes, and was not associated with significant elevations of fasting glucose or insulin among non-diabetic pedigree members. Similarly, two common variants in the islet promoter did not segregate with diabetes. We conclude that among typical familial Type 2 diabetes in a population representative of Northern European Caucasians, glucokinase mutations are an unlikely cause of diabetes. [Diabetologia (1994) 37: 182–187] Received: 10 June 1993 and in revised form: 20 August 1993