Pressure loss recovery in aortic valve stenosis: Contemporary relevance.

Pressure loss recovery (PLR) is a hydrodynamic phenomenon that occurs when blood flow encounters a narrowing typified by aortic valve stenosis (AS). Multiple factors contribute to the magnitude of PLR including the volumetric rate of flow, the geometry of the entrance to the vena contracta (VC) or point of minimum dimension, including that of the left ventricular outflow tract and valve orifice, and the geometry of the proximal aorta. In the majority of clinical circumstances, PLR results in echocardiographic Doppler gradient estimates that are modestly, but generally not clinically important, greater than those derived from rigorously performed catheter measurements. The contribution of PLR to the echocardiographically-measured gradient may not differ significantly between currently available valve prostheses and is likely to be small in patients with mild AS following TAVR.

Clinical Effectiveness of Sacubitril/Valsartan Among Patients Hospitalized for Heart Failure With Reduced Ejection Fraction.

Background Sacubitril/Valsartan has been highly efficacious in randomized trials of heart failure with reduced ejection fraction (HFrEF). However, the effectiveness of sacubitril/valsartan in older patients hospitalized for HFrEF in real-world US practice is unclear. Methods and Results This study included Medicare beneficiaries age ≥65 years who were hospitalized for HFrEF ≤40% in the Get With The Guidelines-Heart Failure registry between October 2015 and December 2018, and eligible for sacubitril/valsartan. Associations between discharge prescription of sacubitril/valsartan and clinical outcomes were assessed after inverse probability of treatment weighting and adjustment for other HFrEF medications. Overall, 1551 (10.9%) patients were discharged on sacubitril/valsartan. Of those not prescribed sacubitril/valsartan, 7857 (62.0%) were prescribed an angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker. Over 12-month follow-up, compared with a discharge prescription of angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker, sacubitril/valsartan was independently associated with lower all-cause mortality (adjusted hazard ratio [HR], 0.82; 95% CI, 0.72-0.94; P=0.004) but not all-cause hospitalization (adjusted HR, 0.97; 95% CI, 0.89-1.07; P=0.55) or heart failure hospitalization (adjusted HR, 1.04; 95% CI, 0.91-1.18; P=0.59). Patients prescribed sacubitril/valsartan versus those without a prescription had lower risk of all-cause mortality (adjusted HR, 0.69; 95% CI, 0.60-0.79; P<0.001), all-cause hospitalization (adjusted HR, 0.90; 95% CI, 0.82-0.98; P=0.02), but not heart failure hospitalization (adjusted HR, 0.94; 95% CI, 0.82-1.08; P=0.40). Conclusions Among patients hospitalized for HFrEF, prescription of sacubitril/valsartan at discharge was independently associated with reduced postdischarge mortality compared with angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker, and reduced mortality and all-cause hospitalization compared with no sacubitril/valsartan. These findings support the use of sacubitril/valsartan to improve postdischarge outcomes among older patients hospitalized for HFrEF in routine US clinical practice.

Patient Perceptions and Familiarity With Medical Therapy for Heart Failure.

Importance: There are major gaps in use of guideline-directed medical therapy (GDMT) for patients with heart failure (HF). Patient-reported data outlining patient goals and preferences associated with GDMT are not available. Objective: To survey patients with chronic HF to better understand their experiences and perceptions of living with HF, including their familiarity and concerns with important GDMT therapies. Design, Setting, and Participants: Study participants were recruited from the GfK KnowledgePanel, a probability-sampled online panel representative of the US adult population. English-speaking adults who met the following criteria were eligible if they were (1) previously told by a physician that they had HF; (2) currently taking medications for HF; and (3) had no history of left ventricular assist device or cardiac transplant. Data were collected between October and November 2018. Analysis began in December 2018. Main Outcomes and Measures: The survey included 4 primary domains: (1) relative importance of disease-related goals, (2) challenges associated with living with HF, (3) decision-making process associated with HF medication use, and (4) awareness and concerns about available HF medications. Results: Of 30 707 KnowledgePanel members who received the initial survey, 15 091 (49.1%) completed the screening questions, 440 were eligible and began the survey, and 429 completed the survey. The median (interquartile range) age was 68 (60-75) years and most were white (320 [74.6%]), male (304 [70.9%]), and had at least a high school education (409 [95.3%]). Most survey responders reported familiarity with ß-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and diuretics. Overall, 107 (24.9%) reported familiarity with angiotensin receptor-neprilysin inhibitors or mineralocorticoid receptor antagonists. Overall, 136 patients (42.5%) reported have safety concerns regarding angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, and 133 (38.5%) regarding ß-blockers, 35 (37.9%) regarding mineralocorticoid receptor antagonists, 38 (36.5%) regarding angiotensin receptor-neprilysin inhibitors, and 123 (37.2%) regarding diuretics. Between 27.7% (n = 26) and 38.5% (n = 136) reported concerns regarding the effectiveness of ß-blockers, angiotensin receptor-neprilysin inhibitors, mineralocorticoid receptor antagonists, or diuretics, while 41% (n = 132) were concerned with the effectiveness of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers. Conclusions and Relevance: In this survey study, many patients were not familiar with GDMT for HF, with familiarity lowest for angiotensin receptor-neprilysin inhibitors and mineralocorticoid receptor antagonists. Among patients not familiar with these therapies, significant proportions questioned their effectiveness and/or safety. Enhanced patient education and shared decision-making support may be effective strategies to improve the uptake of GDMT for HF in US clinical practice.

The Burden of Congestion in Patients Hospitalized With Acute Decompensated Heart Failure.

Congestion is associated with adverse outcomes in heart failure (HF) patients. We characterized congestion in patients hospitalized for HF and examined the association between congestion severity at admission and postdischarge outcomes. Using the OPTIMIZE-HF registry linked to Medicare claims, we analyzed patients ≥65 years old hospitalized for HF from 2003 to 2004. Congestion severity was measured using a 15-point scale that scores dyspnea, orthopnea, fatigue, jugular venous pressure, rales, and edema. Patient characteristics and outcomes were described by congestion strata. Proportional hazards models were fit to examine associations between congestion and 1-year outcomes. Congestion scores for the 24,724 patients ranged from 0 to 14, with a median of 5 (Q1, Q3: 3, 7). At baseline, patients with the highest scores (≥7) had the highest rates of recent HF hospitalizations, EF ≤40%, and co-morbidities, including arrhythmias, diabetes mellitus, and renal insufficiency. Adjusting for patient characteristics, a 3-point congestion score increase was positively associated with mortality (hazard ratio [HR] 1.06, 95% confidence interval [CI] 1.03, 1.09), all-cause rehospitalization (HR 1.02, 95% CI 1.00, 1.04), and HF rehospitalization (HR 1.09, 95% CI 1.06, 1.12), but not emergency department visits (HR 0.99, 95% CI 0.97, 1.01). In conclusion, for patients hospitalized with HF, congestion was associated with rehospitalization and mortality.

Relationship Between Hospital Characteristics and Early Adoption of Angiotensin-Receptor/Neprilysin Inhibitor Among Eligible Patients Hospitalized for Heart Failure.

Background The angiotensin-receptor/neprilysin inhibitor ( ARNI ) sacubitril/valsartan reduces hospitalization and mortality for patients with heart failure with reduced ejection fraction. However, adoption of ARNI into clinical practice has been slow. Factors influencing use of ARNI have not been fully elucidated. Using data from the Get With The Guidelines-Heart Failure registry, Hospital Compare, Dartmouth Atlas, and the American Hospital Association Survey, we sought to identify hospital characteristics associated with patient-level receipt of an ARNI prescription. Methods and Results We analyzed patients with heart failure with reduced ejection fraction who were eligible for ARNI prescription (ejection fraction≤40%, no contraindications) and hospitalized from October 1, 2015 through December 31, 2016. We used logistic regression to estimate the associations between hospital characteristics and patient ARNI prescription at hospital discharge, accounting for clustering of patients within hospitals using generalized estimating equation methods and adjusting for patient-level covariates. Of 16 674 eligible hospitalizations from 210 hospitals, 1020 patients (6.1%) were prescribed ARNI at discharge. The median hospital-level proportion of patients prescribed ARNI was 3.3% (Q1, Q3: 0%, 12.6%). After adjustment for patient-level covariates, for-profit hospitals had significantly higher odds of ARNI prescription compared with not-for-profit hospitals (odds ratio, 2.53; 95% CI , 1.05-6.10; P=0.04), and hospitals located in the Western United States had lower odds of ARNI prescription compared with those in the Northeast (odds ratio, 0.33; 95% CI , 0.13-0.84; P=0.02). Conclusions Relatively few hospital characteristics were associated with ARNI prescription at hospital discharge, in contrast to what has been observed in early adoption in other disease areas. Additional evaluation of barriers to implementing new evidence into heart failure practice is needed.

2018 ACC/HRS/NASCI/SCAI/SCCT Expert Consensus Document on Optimal Use of Ionizing Radiation in Cardiovascular Imaging-Best Practices for Safety and Effectiveness, Part 2: Radiological Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection.

The stimulus to create this document was the recognition that ionizing radiation-guided cardiovascular procedures are being performed with increasing frequency, leading to greater patient radiation exposure and, potentially, to greater exposure to clinical personnel. While the clinical benefit of these procedures is substantial, there is concern about the implications of medical radiation exposure. ACC leadership concluded that it is important to provide practitioners with an educational resource that assembles and interprets the current radiation knowledge base relevant to cardiovascular procedures. By applying this knowledge base, cardiovascular practitioners will be able to select procedures optimally, and minimize radiation exposure to patients and to clinical personnel. "Optimal Use of Ionizing Radiation in Cardiovascular Imaging - Best Practices for Safety and Effectiveness" is a comprehensive overview of ionizing radiation use in cardiovascular procedures and is published online. To provide the most value to our members, we divided the print version of this document into 2 focused parts. "Part I: Radiation Physics and Radiation Biology" addresses radiation physics, dosimetry and detrimental biologic effects. "Part II: Radiologic Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection" covers the basics of operation and radiation delivery for the 3 cardiovascular imaging modalities (x-ray fluoroscopy, x-ray computed tomography, and nuclear scintigraphy). For each modality, it includes the determinants of radiation exposure and techniques to minimize exposure to both patients and to medical personnel.

2018 ACC/HRS/NASCI/SCAI/SCCT Expert Consensus Document on Optimal Use of Ionizing Radiation in Cardiovascular Imaging-Best Practices for Safety and Effectiveness, Part 1: Radiation Physics and Radiation Biology: A Report of the American College of Cardiology Task Force on Expert Consensus Decision Pathways Developed in Collaboration With Mended Hearts.

The stimulus to create this document was the recognition that ionizing radiation-guided cardiovascular procedures are being performed with increasing frequency, leading to greater patient radiation exposure and, potentially, to greater exposure for clinical personnel. Although the clinical benefit of these procedures is substantial, there is concern about the implications of medical radiation exposure. The American College of Cardiology leadership concluded that it is important to provide practitioners with an educational resource that assembles and interprets the current radiation knowledge base relevant to cardiovascular procedures. By applying this knowledge base, cardiovascular practitioners will be able to select procedures optimally, and minimize radiation exposure to patients and to clinical personnel. Optimal Use of Ionizing Radiation in Cardiovascular Imaging: Best Practices for Safety and Effectiveness is a comprehensive overview of ionizing radiation use in cardiovascular procedures and is published online. To provide the most value to our members, we divided the print version of this document into 2 focused parts. Part I: Radiation Physics and Radiation Biology addresses the issue of medical radiation exposure, the basics of radiation physics and dosimetry, and the basics of radiation biology and radiation-induced adverse effects. Part II: Radiological Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection covers the basics of operation and radiation delivery for the 3 cardiovascular imaging modalities (x-ray fluoroscopy, x-ray computed tomography, and nuclear scintigraphy) and will be published in the next issue of the Journal.

Home-Time After Discharge Among Patients Hospitalized With Heart Failure.

BACKGROUND: Surveys of patients with cardiovascular disease have suggested that "home-time"-being alive and out of any health care institution-is a prioritized outcome. This novel measure has not been studied among patients with heart failure (HF). OBJECTIVES: This study sought to characterize home-time following hospitalization for HF and assess its relationship with patient characteristics and traditionally reported clinical outcomes. METHODS: Using GWTG-HF (Get With The Guidelines-Heart Failure) registry data, patients discharged alive from an HF hospitalization between 2011 and 2014 and ≥65 years of age were identified. Using Medicare claims, post-discharge home-time over 30-day and 1-year follow-up was calculated for each patient as the number of days alive and spent outside of a hospital, skilled nursing facility (SNF), or rehabilitation facility. RESULTS: Among 59,736 patients, 57,992 (97.1%) and 42,153 (70.6%) had complete follow-up for home-time calculation through 30 days and 1 year, respectively. The mean home-time was 21.6 ± 11.7 days at 30 days and 243.9 ± 137.6 days at 1 year. Contributions to reduced home-time varied by follow-up period, with days spent in SNF being the largest contributor though 30 days and death being the largest contributor through 1 year. Over 1 year, 2,044 (4.8%) patients had no home-time following index hospitalization discharge, whereas 8,194 (19.4%) had 365 days of home-time. In regression models, several conditions were associated with substantially reduced home-time, including chronic obstructive pulmonary disease, renal insufficiency, and dementia. Through 1 year, home-time was highly correlated with time-to-event endpoints of death (tau = 0.72) and the composite of death or HF readmission (tau = 0.59). CONCLUSIONS: Home-time, which can be readily calculated from administrative claims data, is substantially reduced for many patients following hospitalization for HF and is highly correlated with traditional time-to-event mortality and hospitalization outcomes. Home-time represents a novel, easily measured, patient-centered endpoint that may reflect effectiveness of interventions in future HF studies.

Early impact of guideline publication on angiotensin-receptor neprilysin inhibitor use among patients hospitalized for heart failure.

BACKGROUND: On May 20, 2016, US professional organizations in cardiology published joint treatment guidelines recommending the use of angiotensin-receptor neprilysin inhibitor (ARNI) for eligible patients with heart failure with reduced ejection fraction (HFrEF). Using data from the Get With The Guidelines-Heart Failure registry, we evaluated the early impact of this update on temporal trends in ARNI prescription. METHODS: We analyzed patients with HFrEF who were eligible for ARNI prescription (EF ≤40%, no contraindications) and hospitalized from February 20, 2016, through August 19, 2016-allowing for 13weeks before and after guideline publication. We quantified trends in ARNI use associated with guidelines publication with an interrupted time-series design using logistic regression and accounting for correlations within hospitals using general estimating equation methods. RESULTS: Of 7,200 eligible patient hospitalizations, 51.9% were discharged in the period directly preceding publication of the guidelines, and 48.1% were discharged after. Odds ratios of ARNI prescription at discharge were significantly higher in the postguideline period compared with the preguideline period in adjusted models (adjusted odds ratio 1.29, 95% CI 1.06-1.57, P=.01). However, there was no significant interaction between observed and expected ARNI use after guideline publication (Pinteraction=.14). Results were consistent using a 6-month before and after time frame. CONCLUSIONS: The model suggested a small increase in ARNI use in HF patients being discharged from the hospital immediately after guideline release. However, the publication of national guidelines recommending ARNI use seemed to have little influence on the adoption of this evidence-based medication in the first 3 to 6months.

Sustained sex-based treatment differences in acute coronary syndrome care: Insights from the American Heart Association Get With The Guidelines Coronary Artery Disease Registry.

BACKGROUND: Sex-based differences in acute coronary syndrome (ACS) mortality may attenuate with age due to better symptom recognition and prompt care. HYPOTHESIS: Age is a modifier of temporal trends in sex-based differences in ACS care. METHODS: Among 104 817 eligible patients with ACS enrolled in the AHA Get With the Guidelines-Coronary Artery Disease registry between 2003 and 2008, care and in-hospital mortality were evaluated stratified by sex and age. Temporal trends within sex and age groups were assessed for 2 care processes: percentage of STEMI patients presenting to PCI-capable hospitals with a DTB time ≤ 90 minutes (DTB90) and proportion of eligible ACS patients receiving aspirin within 24 hours. RESULTS: After adjustment for clinical risk factors and sociodemographic and hospital characteristics, 2276 (51.7%) women and 6276 (56.9%) men with STEMI were treated with DTB90 (adjusted OR: 0.85, 95% CI: 0.80-0.91, P < 0.0001 for women vs men). Time trend analysis showed an absolute increase ranging from 24% to 35% in DTB90 rates among both men and women (P for trend <0.0001 for each group), with consistent differences over time across the 4 age/sex groups (3-way P-interaction = 0.93). Despite high rate of baseline aspirin use (87%-91%), there was a 9% to 11% absolute increase in aspirin use over time, also with consistent differences across the 4 age/sex groups (all 3-way P-interaction ≥0.15). CONCLUSIONS: Substantial gains of generally similar magnitude existed in ACS performance measures over 6 years of study across sex and age groups; areas for improvement remain, particularly among younger women.

Late-stage left ventricular dysfunction in adult survivors of tetralogy of Fallot repair in childhood.

Objective: Left ventricular systolic dysfunction (LVSD) in adult survivors of tetralogy of Fallot (TOF) repair in childhood has been observed, although the relationship with long-term outcome remains inadequately described. Methods: A cohort of 44 consecutive adult patients with TOF repair in childhood were followed prospectively from January 2001 through June 2016. LVSD was defined as an echocardiographically derived left ventricular (LV) ejection fraction <0.55. Clinical and demographic characteristics in patients with and without LVSD were compared. Event-free survival (all-cause death or hospitalisation) was estimated using the product limit method. Results: The average time from childhood surgical repair to cohort inception was similar between groups (LVSD, 33.7±12.7 years; normal LV function, 36.1±14.9 years; P=0.62) as were their mean ages (LVSD, 36.5±14.5 years; normal LV function, 40.7±15.2 years; P=0.73). Patients with LVSD (n=13) had more prior surgeries, more frequent history of significant pulmonic regurgitation, right ventricular systolic dysfunction and more implantable cardiac devices. Over a total observation time of 15.5 years, patients with LVSD were at significantly higher risk of all-cause death or hospitalisation (P=0.006). Onset of LVSD frequently preceded an adverse outcome. Conclusions: In this cohort of adult patients with TOF repair in childhood followed for a total of 550 patient-years, the frequency of LVSD was 30%. LVSD was associated with lower event-free survival. The appearance of LVSD many years after TOF repair may herald the onset of an adverse outcome.

Seasonal and circadian variations of acute myocardial infarction: Findings from the Get With The Guidelines-Coronary Artery Disease (GWTG-CAD) program.

BACKGROUND: Seasonal variation with winter preponderance of myocardial infarction incidence has been described decades ago, but only a few small studies have classified myocardial infarction based on ST-segment elevation. It is unclear whether seasonal and circadian variations are equally present in warmer and colder regions. We investigated whether seasonal and circadian variations in acute myocardial infarction (AMI) are more prominent in colder northern states compared with warmer southern states. We also investigated the peak time of admission to better understand the circadian rhythm. METHODS: Data from the GWTG-CAD database were used. We analyzed 82,971 consecutive acute myocardial infarction (AMI) patients treated at 276 US centers from 2003 to 2008. The country was geographically divided into warmer southern and colder northern states using latitude 35 degrees for this purpose. RESULTS: Overall, acute myocardial infarction (AMI) admissions varied across seasons (P < .01), and were higher in winter (winter vs. spring n = 21,483 vs. 20,291, respectively). When stratified based on type of AMI, non-ST-segment elevation myocardial infarction (NSTEMI) admissions varied across seasons (P < .01) and were highest in winter and lowest in spring. Seasonal variation was not significant in STEMI admissions (P = .30). Seasonal variation with winter predominance was noted in AMI patients in warmer southern states (P < .01), but not in colder states. The distributions of length of stay for AMI patients and door to balloon times for STEMI patients were minimally different across all four seasons (P < .01) with longest occurring in winter. Most patients with AMI presented during daytime with a peak close to 11 am and a nadir at approximately 4 am. CONCLUSIONS: Seasonal variation with winter predominance exists in AMI admissions and was significant in NSTEMI admissions but not in STEMI admissions. Seasonal variation was only significant in warmer southern states.

Compliance with guideline-directed therapy in diabetic patients admitted with acute coronary syndrome: Findings from the American Heart Association's Get With The Guidelines-Coronary Artery Disease (GWTG-CAD) program.

BACKGROUND: To evaluate and compare baseline characteristics, outcomes and compliance with guideline based therapy at discharge among diabetic and non-diabetic patients admitted with acute coronary syndromes (ACS). METHODS AND RESULTS: Study population consisted of 151,270 patients admitted with ACS from 2002 through 2008 at 411 sites participating in the American Heart Association's Get with the Guidelines (GWTG) program. Demographic variables, physical exam findings, laboratory data, left ventricular ejection fraction, length of stay, in-hospital mortality and discharge medications were compared between diabetic and non-diabetic patients. Temporal trends in compliance with guidelines directed therapy were evaluated. Of 151,270 patients, 48,938 (32%) had diabetes. Overall, diabetic patients were significantly older and more likely non-white. They had significantly more hypertension, atherosclerotic disease, CKD, and LV dysfunction and were more likely to present as NSTEMI. They had longer hospital stay and higher hospital mortality than non-diabetic patients. Diabetic patients were less likely to get LDL checks (65% vs 70%) and less frequently prescribed statins (85% vs 89%), RAAS blockers for LV dysfunction (80% vs 84%) and dual-antiplatelet therapy (69% vs 74%). Diabetic patients were less likely to achieve BP goals before discharge (75% vs 82%). Fewer diabetic patients met first medical contact to PCI time for STEMI (44% vs 52%). Temporal trends, however, showed continued progressive improvement in most performance measures from 2002 to 2008 (all P<.001). CONCLUSIONS: These data from a large cohort of ACS patients demonstrate gaps in compliance with guidelines directed therapy in diabetic patients but also indicate significant and continued improvement in most performance measures over time. Concerted efforts are needed to continue this positive trend.

Early Adoption of Sacubitril/Valsartan for Patients With Heart Failure With Reduced Ejection Fraction: Insights From Get With the Guidelines-Heart Failure (GWTG-HF).

OBJECTIVES: The aim of this study was to assess the prevalence and variation in angiotensin receptor/neprilysin inhibitor (ARNI) prescription among a real-world population with heart failure with reduced ejection fraction (HFrEF). BACKGROUND: The U.S. Food and Drug Administration approved sacubitril/valsartan for patients with HFrEF in July 2015. Little is known about the early patterns of use of this novel therapy. METHODS: The study included patients discharged alive from hospitals in Get With the Guidelines-Heart Failure (GWTG-HF), a registry of hospitalized patients with heart failure, between July 2015 and June 2016 who had documentation of whether ARNIs were prescribed at discharge. Patient and hospital characteristics were compared among patients with HFrEF (ejection fraction ≤40%) with and without ARNI prescription at discharge, excluding those with documented contraindications to ARNIs. To evaluate hospital variation, hospitals with at least 10 eligible hospitalizations during the study period were assessed. RESULTS: Of 21,078 patients hospitalized with HFrEF during the study period, 495 (2.3%) were prescribed ARNIs at discharge. Patients prescribed ARNIs were younger (median age 65 years vs. 70 years; p < 0.001), had lower ejection fractions (median 23% vs. 25%; p < 0.001), and had higher use of aldosterone antagonists (45% vs. 31%; p < 0.001) at discharge. At the 241 participating hospitals with 10 or more eligible admissions, 125 (52%) reported no discharge prescriptions of ARNIs. CONCLUSIONS: Approximately 2.3% of patients hospitalized for HFrEF in a national registry were prescribed ARNI therapy in the first 12 months following Food and Drug Administration approval. Further study is needed to identify and overcome barriers to implementing new evidence into practice, such as ARNI use among eligible patients with HFrEF.

Prediction of 30-Day All-Cause Readmissions in Patients Hospitalized for Heart Failure: Comparison of Machine Learning and Other Statistical Approaches.

Importance: Several attempts have been made at developing models to predict 30-day readmissions in patients with heart failure, but none have sufficient discriminatory capacity for clinical use. Machine-learning (ML) algorithms represent a novel approach and may have potential advantages over traditional statistical modeling. Objective: To develop models using a ML approach to predict all-cause readmissions 30 days after discharge from a heart failure hospitalization and to compare ML model performance with models developed using "conventional" statistically based methods. Design, Setting, and Participants: Models were developed using ML algorithms, specifically, a tree-augmented naive Bayesian network, a random forest algorithm, and a gradient-boosted model and compared with traditional statistical methods using 2 independently derived logistic regression models (a de novo model and an a priori model developed using electronic health records) and a least absolute shrinkage and selection operator method. The study sample was randomly divided into training (70%) and validation (30%) sets to develop and test model performance. This was a registry-based study, and the study sample was obtained by linking patients from the Get With the Guidelines Heart Failure registry with Medicare data. After applying appropriate inclusion and exclusion criteria, 56 477 patients were included in our analysis. The study was conducted between January 4, 2005, and December 1, 2010, and analysis of the data was conducted between November 25, 2014, and June 30, 2016. Main Outcomes and Measures: C statistics were used for comparison of discriminatory capacity across models in the validation sample. Results: The overall 30-day rehospitalization rate was 21.2% (11 959 of 56 477 patients). For the tree-augmented naive Bayesian network, random forest, gradient-boosted, logistic regression, and least absolute shrinkage and selection operator models, C statistics for the validation sets were similar: 0.618, 0.607, 0.614, 0.624, and 0.618, respectively. Applying the previously validated electronic health records model to our study sample yielded a C statistic of 0.589 for the validation set. Conclusions and Relevance: Use of a number of ML algorithms did not improve prediction of 30-day heart failure readmissions compared with more traditional prediction models. Although there will likely be further applications of ML approaches in prognostic modeling, our study fits within the literature of limited predictive ability for heart failure readmissions.

Association of radial versus femoral access with contrast-induced acute kidney injury in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction.

BACKGROUND: Patients with ST-elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI) are at an increased risk of developing contrast-induced acute kidney injury (CI-AKI). Data on the association between transradial (TRA) vs. transfemoral (TFA) access and the risk of CI-AKI in this setting are limited. METHODS: We analyzed data on 1162 patients undergoing primary PCI for STEMI at two tertiary care centers between 2010 and 2014. Primary outcome was CI-AKI, defined as a relative rise in serum creatinine of ≥25%, or an absolute increase of ≥0.5mg/dL, within 48h of primary PCI. We used multivariable logistic regression and propensity analysis to determine the association between vascular access site and CI-AKI. RESULTS: Of 1162 patients who underwent primary PCI for STEMI, TFA was used in 857 (73.8%), and TRA in 305 (26.2%) patients. In the unmatched cohort, TRA was associated with numerically lower rates of CI-AKI as compared with TFA; however, this difference did not reach statistical significance (5.9% vs. 7.0%; unadjusted OR 0.83, 95%CI 0.48-1.44, p=0.510; adjusted OR 0.84, 95%CI 0.44-1.62, p=0.610). Similar results were seen in a propensity matched cohort of 508 patients (254 TRA and 254 TFA; CI-AKI 5.5% vs. 8.3%, OR 0.65, 95% CI 0.32-1.30, p=0.220). CONCLUSIONS: In patients with STEMI undergoing primary PCI, TRA was not associated with a lower risk of CI-AKI, as compared with TFA. Randomized controlled trials are needed to definitely assess the role of vascular access site in reducing the risk of CI-AKI in patients undergoing primary PCI for STEMI. SUMMARY: In patients with STEMI undergoing primary PCI, the overall incidence of contrast-induced acute kidney injury (CI-AKI) was low (6.7%). Transradial access was not associated with a lower risk of CI-AKI as compared with transfemoral access.