Effect of hypertrophic cardiomyopathy on the prediction of thromboembolism in patients with nonvalvular atrial fibrillation

Prediction of new-onset atrial fibrillation in patients with hypertrophic cardiomyopathy using machine learning

AIMS

Atrial fibrillation (AF) is the most common sustained arrhythmia among patients with hypertrophic cardiomyopathy (HCM), leading to increased symptom burden and risk of thromboembolism. The HCM-AF score was developed to predict new-onset AF in patients with HCM, though sensitivity and specificity of this conventional tool are limited. Thus, there is a need for more accurate tools to predict new-onset AF in HCM. The objective of the present study was to develop a better model to predict new-onset AF in patients with HCM using machine learning (ML).

METHODS AND RESULTS

In this prospective, multicentre cohort study, we enrolled 1069 patients with HCM without a prior history of AF. We built a ML model (logistic regression with Lasso regularization) using clinical variables. We developed the ML model using the cohort from one institution (training set) and applied it to an independent cohort from a separate institution (test set). We used the HCM-AF score as a reference model. We compared the area under the receiver-operating characteristic curve (AUC) between the ML model and the reference model using the DeLong's test. Median follow-up time was 2.1 years, with 128 (12%) patients developing new-onset AF. Using the ML model developed in the training set to predict new-onset AF, the AUC in the test set was 0.84 (95% confidence interval [CI] 0.77-0.91). The ML model outperformed the reference model (AUC 0.64; 95% CI 0.54-0.73; DeLong's p < 0.001). The ML model had higher sensitivity (0.82; 95% CI 0.65-0.93) than that of the reference model (0.67; 95% CI 0.52-0.88). The ML model also had higher specificity (0.76; 95% CI 0.71-0.81) than that of the reference model (0.57; 95% CI 0.41-0.70). Among the most important clinical variables included in the ML-based model were left atrial volume and diameter, left ventricular outflow tract gradient with exercise stress and at rest, late gadolinium enhancement on cardiac magnetic resonance imaging, peak heart rate during exercise stress, age at diagnosis, positive genotype, diabetes mellitus, and end-stage renal disease.

CONCLUSION

Our ML model showed superior performance compared to the conventional HCM-AF score for the prediction of new-onset AF in patients with HCM.

Effects of Longitudinal Changes in Anemia Status on Clinical Outcomes in Patients With Non-Valvular Atrial Fibrillation - Analysis From the Hokuriku-Plus AF Registry

BACKGROUND

Anemia, a common comorbidity in older patients with heart failure (HF) and atrial fibrillation (AF), is associated with an increased risk of adverse events. This study evaluated the prognostic effects of longitudinal changes in anemia status on clinical outcomes in patients with AF.

METHODS AND RESULTS

We prospectively evaluated data of 1,388 patients with AF from the Hokuriku-Plus AF Registry (1,010 men; mean [±SD] age 72.3±9.7 years) and recorded the incidence of death, HF, thromboembolism, and major bleeding. Of these patients, the 1,233 for whom hemoglobin levels were available at baseline and at the 1-year follow-up were further evaluated. Patients were categorized into 3 groups based on longitudinal changes in 1-year anemia status: Group 1, AF without anemia; Group 2, AF with improved anemia; and Group 3, AF with sustained or new-onset anemia. Over the 1-5 years of follow up, the incidences of death, HF, thromboembolism, and major bleeding were significantly higher among patients with than without anemia. In addition, the incidence of death or HF was significantly higher in Group 3 than in Groups 1 and 2. Multivariate analysis revealed no anemia or improvement in anemia in 1 year as an independent predictor for a favorable prognosis for cardiovascular death and HF.

CONCLUSIONS

Recovery from anemia may be associated with a favorable clinical course of AF.

Atrial Fibrillation in Hypertrophic Cardiomyopathy

Atrial fibrillation (AF) is common among patients with hypertrophic cardiomyopathy (HCM) with a prevalence greater than 25%. AF in HCM is associated with a high risk of stroke and can be a marker of more advanced cardiomyopathy. Although, it frequently results in cardiac hemodynamic changes which are poorly tolerated, it can be subclinical. Thus, prompt diagnosis and adequate management of AF are essential to minimizing AF-related adverse outcomes in HCM. All HCM patients should be screened for AF regularly, and those with high-risk features should be screened more frequently preferably with extended ambulatory monitoring. Once AF is detected, oral anticoagulation should be initiated. Both general and HCM-specific modifiable risk factors should be addressed and assessment for cardiomyopathy progression should be performed. Although no randomized controlled studies have compared rate versus rhythm control in HCM, early rhythm control could be considered to prevent further LA remodeling.

Clinical Significance of Postoperative Atrial Fibrillation in Hypertrophic Cardiomyopathy Patients Undergoing Septal Myectomy

BACKGROUND

The optimal management of hypertrophic cardiomyopathy (HCM) patients with postoperative atrial fibrillation (POAF) after surgical myectomy remains unknown. We sought to investigate the association between POAF and atrial fibrillation (AF) or cardioembolic events during follow-up to bridge this gap.

METHODS

Patients undergoing surgical myectomy at 2 HCM referral centres in North America from 2002 to 2020 were included in this study. Patients with preoperative AF were excluded. POAF was defined as any episode of AF within 30 days after surgery.

RESULTS

Of 1176 patients, 375 (31.9%) had POAF. Age (adjusted hazard ratio [HR] 1.05, 95% confidence interval [CI] 1.03-1.06; P < 0.001), premyectomy left atrial diameter (LAD; adjusted HR 1.6, 95% CI 1.32-2.02; P < 0.001), and smoking (adjusted HR 1.60, 95% CI 1.17-2.20; P = 0.001) were associated with POAF on multivariable analysis. Of 934 patients with follow-up data, of duration 4.3 ± 4.1 years, AF was detected in 86 (9.2%). Only POAF (HR 4.20, 95% CI 2.44-7.23; P < 0.001), previous history of stroke (HR 4.81, 95% CI 1.63-14.17; P = 0.01), and postmyectomy LAD (HR 1.80, 95% CI 1.21-2.70; P = 0.004) were associated with AF incidence during follow-up. Cardioembolic events occurred in only 15 patients (1.6%). POAF was not associated with increased cardioembolic risk, with only 3 patients with POAF suffering such an event, all more than 4 years after surgery.

CONCLUSIONS

POAF is common in HCM patients undergoing myectomy and is a predictor of AF during follow-up. Over long-term follow-up, cardioembolic events are uncommon. These findings suggest that routine long-term anticoagulation for all HCM patients with postmyectomy AF is not justified after the initial postoperative period.

Hypertrophic Cardiomyopathy Predicts Thromboembolism and Heart Failure in Patients With Nonvalvular Atrial Fibrillation - A Prospective Analysis From the Hokuriku-Plus AF Registry

BACKGROUND

The prognostic effect of concomitant hypertrophic cardiomyopathy (HCM) on adverse events in patients with atrial fibrillation (AF) has not been evaluated in a multicenter prospective cohort study in Japan.

METHODS AND RESULTS

Using the Hokuriku-Plus AF Registry, 1,396 patients with nonvalvular AF (1,018 men, 72.3±9.7 years old) were assessed prospectively; 72 (5.2%) had concomitant HCM. During a median follow-up of 5.0 years (interquartile range 3.5-5.3 years), 79 cases of thromboembolism (1.3 per 100 person-years) and 192 of heart failure (HF) (3.2 per 100 person-years) occurred. Kaplan-Meier analysis revealed that the HCM group had a significantly greater incidence of thromboembolism (P=0.002 by log-rank test) and HF (P<0.0001 by a log-rank test) than the non-HCM group. The Cox proportional hazards model demonstrated that persistent AF (adjusted hazard ratio 2.98, 95% confidence interval 1.56-6.21), the CHA2DS2-VASc score (1.35, 1.18-1.54), and concomitant HCM (2.48, 1.16-4.79) were significantly associated with thromboembolism. Conversely, concomitant HCM (2.81, 1.72-4.43), older age (1.07, 1.05-1.10), lower body mass index (0.95, 0.91-0.99), a history of HF (2.49, 1.77-3.52), and lower left ventricular ejection fraction (0.98, 0.97-0.99) were significantly associated with the development of HF.

CONCLUSIONS

Concomitant HCM predicts the incidence of thromboembolism and HF in AF patients.

Effect of Catheter Ablation for Atrial Fibrillation in Heart Failure With Mid-Range or Preserved Ejection Fraction - Pooled Analysis of the AF Frontier Ablation Registry and Hokuriku-Plus AF Registry

BACKGROUND

A recent randomized trial demonstrated that catheter ablation for atrial fibrillation (AF) in patients with heart failure with reduced ejection fraction (EF) is associated with a reduction in death or heart failure. However, the effect of catheter ablation for AF in patients with heart failure with mid-range or preserved EF is unclear.

METHODS AND RESULTS

We screened 899 AF patients (72.4% male, mean age 68.4 years) with heart failure and left ventricular EF ≥40% from 2 Japanese multicenter AF registries: the Atrial Fibrillation registry to Follow the long-teRm Outcomes and use of aNTIcoagulants aftER Ablation (AF Frontier Ablation Registry) as the ablation group (525 patients who underwent ablation) and the Hokuriku-Plus AF Registry as the medical therapy group (374 patients who did not undergo ablation). Propensity score matching was performed in these 2 registries to yield 106 matched patient pairs. The primary endpoint was a composite of cardiovascular death and hospitalization for heart failure. At 24.6 months, the ablation group had a significantly lower incidence of the primary endpoint (hazard ratio 0.32; 95% confidence interval 0.13-0.70; P=0.004) than the medical therapy group.

CONCLUSIONS

Compared with medical therapy, catheter ablation for AF in patients with heart failure and mid-range or preserved EF was associated with a significantly lower incidence of cardiovascular death or hospitalization for heart failure.

Hypertrophic Cardiomyopathy and Atrial Fibrillation: A Systematic Review and Meta-analysis of Anticoagulation Strategy

BACKGROUND

Atrial fibrillation (AF) frequently complicates hypertrophic cardiomyopathy (HCM), and anticoagulation significantly decreases the risk of stroke in this population. To date, no randomized controlled trials (RCTs) have compared direct oral anticoagulants (DOACs) and vitamin K antagonists (VKAs). The present study aimed to systematically compare the two anticoagulation strategies in terms of effectiveness and safety.

METHOD

We performed a systematic literature search and meta-analysis in the PubMed, MEDLINE, and EMBASE databases for studies reporting all-cause mortality, major bleeding, or thromboembolic events (TEs). Since no RCTs were available, we included observational studies only. The overall hazard ratio (HR) and 95% confidence interval (CI) for each analyzed parameter were pooled using a random-effects model.

RESULTS

Five observational studies including 6919 patients were eligible for inclusion. Compared with VKAs, DOACs were associated with statistically significant lower rates of all-cause mortality (HR 0.64, 95% CI 0.35-0.54; p < 0.00001), comparable major bleeding events (HR 0.64, 95% CI 0.40-1.03; p = 0.07), and TEs (HR 0.94, 95% CI 0.73-1.22; p = 0.65).

CONCLUSIONS

Compared with VKAs, a DOAC-based strategy might represent an effective and safe strategy regarding all-cause mortality, major/life-threatening bleeding complications, and TEs in HCM patients with concomitant AF. However, further prospective studies are necessary to reinforce a DOAC-based anticoagulation strategy in this population.

Characterization of baseline clinical factors associated with incident worsening kidney function in patients with non-valvular atrial fibrillation: the Hokuriku-Plus AF Registry

Evidence suggests that atrial fibrillation (AF) could increase the risk of worsening kidney function (WKF) which is linked to an increased risk of stroke, bleeding, and death in AF patients. However, limited data exist regarding the factors that could lead to WKF in these patients. Therefore, we sought to identify the potential factors associated with the development of WKF in patients with non-valvular AF (NVAF). We analyzed prospectively recruited 1122 NVAF patients [men 71.9%, median age 73.0 years (interquartile range: 66.0-79.0)] with a baseline estimated glomerular filtration rate (eGFR) ≥ 15 mL/min/1.73 m² from the Hokuriku-Plus AF Registry. The primary outcome was incident WKF, defined as the %eGFR change from the baseline ≥ 30% during the follow-up period. We evaluated the association between baseline variables and incident WKF using univariate and multivariate Cox proportional hazard models. We also evaluated the non-linear association between the identified factors and incident WKF. During a median follow-up period of 3.0 years (interquartile range: 2.7-3.3), incident WKF was observed in 108 patients (32.6 per 1000 person-years). Compared to the patients without incident WKF, the patients with incident WKF were older and had a higher prevalence of heart failure (HF), diabetes mellitus (DM), and vascular disease at baseline. Those who experienced incident WKF also had higher diastolic blood pressure, lower hemoglobin, lower eGFR, higher B-type natriuretic peptide (BNP) and used warfarin more frequently. Upon multivariate analysis, age ≥ 75 years, HF, DM, and anemia were independently associated with incident WKF. Additionally, age and hemoglobin were linearly associated with the risk of incident WKF, whereas a J- or U-shaped association was observed for HbA1c and BNP. Age ≥ 75 years, HF, DM, and anemia were associated with the development of WKF in Japanese patients with NVAF. In patients with these risk factors, a careful monitoring of the kidney function and appropriate interventions may be important when possible.

Comprehensive comparison of stroke risk score performance: a systematic review and meta-analysis among 6 267 728 patients with atrial fibrillation

AIMS

Multiple risk scores to predict ischaemic stroke (IS) in patients with atrial fibrillation (AF) have been developed. This study aims to systematically review these scores, their validations and updates, assess their methodological quality, and calculate pooled estimates of the predictive performance.

METHODS AND RESULTS

We searched PubMed and Web of Science for studies developing, validating, or updating risk scores for IS in AF patients. Methodological quality was assessed using the Prediction model Risk Of Bias ASsessment Tool (PROBAST). To assess discrimination, pooled c-statistics were calculated using random-effects meta-analysis. We identified 19 scores, which were validated and updated once or more in 70 and 40 studies, respectively, including 329 validations and 76 updates-nearly all on the CHA2DS2-VASc and CHADS2. Pooled c-statistics were calculated among 6 267 728 patients and 359 373 events of IS. For the CHA2DS2-VASc and CHADS2, pooled c-statistics were 0.644 [95% confidence interval (CI) 0.635-0.653] and 0.658 (0.644-0.672), respectively. Better discriminatory abilities were found in the newer risk scores, with the modified-CHADS2 demonstrating the best discrimination [c-statistic 0.715 (0.674-0.754)]. Updates were found for the CHA2DS2-VASc and CHADS2 only, showing improved discrimination. Calibration was reasonable but available for only 17 studies. The PROBAST indicated a risk of methodological bias in all studies.

CONCLUSION

Nineteen risk scores and 76 updates are available to predict IS in patients with AF. The guideline-endorsed CHA2DS2-VASc shows inferior discriminative abilities compared with newer scores. Additional external validations and data on calibration are required before considering the newer scores in clinical practice.

CLINICAL TRIAL REGISTRATION

ID CRD4202161247 (PROSPERO).

Clinical Characteristics, Outcomes, and Risk Factors for Adverse Events in Elderly and Non-Elderly Japanese Patients With Non-Valvular Atrial Fibrillation ― Competing Risk Analysis From the Hokuriku-Plus AF Registry ―

Background: Few studies in Japan have reported on follow-up data regarding the clinical course and risk factors for adverse outcomes in elderly patients with non-valvular atrial fibrillation (NVAF), vs. younger patients, when considering the competing risk of death.

Methods and Results: We prospectively studied 1,328 patients with NVAF (965 men; mean [±SD] age 72.4±9.7 years) from the Hokuriku-Plus AF Registry with a median follow-up of 5.0 years (interquartile range 3.5–5.3 years) and evaluated the incidence of thromboembolism or major bleeding in elderly (age ≥75 years; n=595) and non-elderly (age <75 years; n=733) patients. Analysis using the Gray method showed no significant difference in the incidence of thromboembolism; however, the incidence of major bleeding was significantly higher in the elderly than non-elderly group. The Fine-Gray model, after adjustment for age and sex in the elderly group, showed that age (hazard ratio [HR] 1.08; 95% confidence interval [CI] 1.02–1.13; P=0.004) and warfarin use (HR 1.87; 95% CI 1.12–3.14; P=0.02) were significantly associated with major bleeding. In the elderly group, those using warfarin had a higher incidence of thromboembolism and major bleeding than those using direct oral anticoagulants (DOACs).

Conclusions: The efficacy and safety of DOACs were remarkable in elderly compared with non-elderly patients with NVAF considering the competing risk of death. DOACs may be a favorable choice in elderly patients with NVAF.

Machine Learning Methods for Identifying Atrial Fibrillation Cases and Their Predictors in Patients With Hypertrophic Cardiomyopathy: The HCM-AF-Risk Model

Background

Hypertrophic cardiomyopathy (HCM) patients have a high incidence of atrial fibrillation (AF) and increased stroke risk, even with low CHA₂DS₂-VASc (congestive heart failure, hypertension, age diabetes, previous stroke/transient ischemic attack) scores. Hence, there is a need to understand the pathophysiology of AF/stroke in HCM. In this retrospective study, we develop and apply a data-driven, machine learning–based method to identify AF cases, and clinical/imaging features associated with AF, using electronic health record data.

Methods

HCM patients with documented paroxysmal/persistent/permanent AF (n = 191) were considered AF cases, and the remaining patients in sinus rhythm (n = 640) were tagged as No-AF. We evaluated 93 clinical variables; the most informative variables useful for distinguishing AF from No-AF cases were selected based on the 2-sample t test and the information gain criterion.

Results

We identified 18 highly informative variables that are positively (n = 11) and negatively (n = 7) correlated with AF in HCM. Next, patient records were represented via these 18 variables. Data imbalance resulting from the relatively low number of AF cases was addressed via a combination of oversampling and undersampling strategies. We trained and tested multiple classifiers under this sampling approach, showing effective classification. Specifically, an ensemble of logistic regression and naïve Bayes classifiers, trained based on the 18 variables and corrected for data imbalance, proved most effective for separating AF from No-AF cases (sensitivity = 0.74, specificity = 0.70, C-index = 0.80).

Conclusions

Our model (HCM-AF-Risk Model) is the first machine learning–based method for identification of AF cases in HCM. This model demonstrates good performance, addresses data imbalance, and suggests that AF is associated with a more severe cardiac HCM phenotype.

Thromboembolism in Patients with Hypertrophic Cardiomyopathy

Hypertrophic cardiomyopathy (HCM) is an inherited cardiac disease, which has a marked heterogeneity in clinical expression, natural history, and prognosis. HCM is associated with a high prevalence of thromboembolic events (stroke and systemic embolic events), even if taking no account of atrial fibrillation (AF), leading to unexpected disability and death in patients of all ages. Several risk factors of thromboembolism such as AF, greater age, left atrial diameter, heart failure and others have been confirmed in patients with HCM. Conventional thromboembolic predictive models were estimated by several trials in HCM population but it turned out to be unsatisfactory. Based on those previous explorations, researchers tried to modify or develop novel models suitable for HCM population in thromboembolism prediction. In consideration of catastrophic advent events of thromboembolism, current guidelines have recommended life-long anticoagulant therapy after a single short AF. Therefore, early identification of risk factors for thromboembolism, accurate risk stratification, timely preventive measures and aggressive management may help to avoid serious adverse thromboembolic events in HCM population.

Clinical impact of atrial fibrillation in a nationwide cohort of hypertrophic cardiomyopathy patients

Background

The clinical implications of atrial fibrillation (AF) in hypertrophic cardiomyopathy (HCM) patients are incompletely characterized. We investigated the impact of AF on stroke and mortality, assessed the performance of the CHA2DS2-VASc score, and explored the predictors of stroke in HCM patients.

Methods

A nationwide cohort of HCM patients (n=8,349, mean age 60.7) identified from 2010 to 2015 from the Korean National Health Insurance database were followed up for occurrence of ischemic stroke or all-cause death.

Results

During a mean follow-up of 2.5 years, the incidence rate of stroke was 2.69 and 5.87, and mortality rate was 2.06 and 4.44 per 100 person-years in non-AF and AF patients, respectively. AF was independently associated with a 60% and 50% increased risk for stroke and all-cause mortality, respectively. The AF-associated increase in risk of stroke was more prominent in HCM patients with no or few risk factors. The CHA2DS2-VASc score showed poor discrimination of stroke risk in HCM patients with AF, mostly due to the high incidence of stroke in patients with scores of 0 or 1. Traditional risk factors were not always associated with stroke in HCM patients with AF; age, heart failure, high blood pressure and GGT were the strongest predictors of stroke in this population. HCM patients without AF also showed increased incidence of stroke at CHA2DS2-VASc ≥1.

Conclusions

AF was independently associated with increased risks for stroke and all-cause mortality in patients with HCM. The CHA2DS2-VASc score showed poor discrimination of stroke risk in HCM patients with AF.

Stroke risk in hypertrophic cardiomyopathy patients with atrial fibrillation: a nationwide database study

Current treatment guidelines recommend anticoagulation for hypertrophic cardiomyopathy (HCM) with atrial fibrillation (AF) regardless of the CHA₂DS₂-VASc score. As aging and stroke risk factors (hypertension, diabetes mellitus) are confounders of ischemic stroke, young patients with a low stroke risk may not need anticoagulant treatment. This study aimed to determine the incidence of stroke and its risk factors in HCM patients with AF during a long-term follow-up. Using a national database, we retrospectively investigated 18,724 HCM patients from a systematic sample of 1,000,000 Taiwanese people between 1997 and 2013. The incidences of AF and stroke were estimated. Data were analyzed using Cox regression models. AF was identified in 598 patients (262 men, mean age 66.3±13.0 years) during a median follow-up of 7.0 years. The AF incidence in HCM patients was 5.83 per 1000 person-years, and the overall incidence of AF-associated stroke was 24.14 per 1000 person-years. The incidence of transient ischemic attack (TIA)/ischemic stroke varied from 20.41 to 60.55 per 1000 person-years, without proportionality to CHA₂DS₂-VASc score increase. Among patients aged <40 years, none experienced TIA/ischemic stroke. Univariate Cox regression models showed that age (p<0.001), prior TIA/ischemic stroke (p=0.02), and CHA₂DS₂-VASc score (p=0.003) were risk factors for TIA/ischemic stroke. Multivariate analysis indicated that age (hazard ratio 1.04, 95% confidence interval [CI] 1.02-1.06, p=0.001) and prior TIA/ischemic stroke (hazard ratio 2.82, 95% CI 1.27-6.25, p=0.011) were independently associated with TIA/ischemic stroke. Taiwanese patients with concomitant HCM and AF have a high stroke risk regardless of the CHA₂DS₂-VASc score. Aging is the main predictor. As the overall incidence of stroke was low in young patients, anticoagulants may not be needed in this subpopulation.

Thromboembolic Outcomes of Different Anticoagulation Strategies for Patients with Atrial Fibrillation in the Setting of Hypertrophic Cardiomyopathy: A Systematic Review

Objective

Limited data is available assessing the efficacy and safety of different anticoagulation (AC) strategies for prevention of thromboembolic events, major bleeding, and all-cause mortality in patients with hypertrophic cardiomyopathy (HCM) and atrial fibrillation (AF). In this systematic review, we conducted a literature search to examine the possible association between different AC strategies and prevention of these adverse outcomes.

Methods

Scientific databases (PubMed, EMBASE, and Scopus) were searched using relevant medical subject headings and keywords to retrieve studies published through September of 2019. Studies assessing the outcomes of interest in patients with HCM and AF receiving AC, no AC as well as direct oral anticoagulants (DOACs) and vitamin K antagonists (VKAs) were selected.

Results

This review identified 14 observational studies evaluating thromboembolic events by AC strategies in 8,479 participants with concomitant HCM and AF. The use of AC was associated with a lower pooled incidence rate of total thromboembolic events at 9.5% (112 events in 1,175 patients) compared to 22.1% with no AC (108 events in 489 patients). In addition, the use of DOACs was associated with a lower pooled incidence rate of thromboembolic events at 4.7% (169 events in 3,576 patients) compared to 8.7% with VKAs (281 events in 3,239 patients). Furthermore, the use of DOACs compared to VKAs was associated with a lower pooled incidence rate of major bleeding and all-cause mortality at 3.8% (136 events in 3,576 patients) versus 6.8% (220 events in 3,239 patients) and 4.1% (124 events in 3,008 patients) versus 16.1% (384 events in 2,380 patients), respectively.

Conclusions

AC of patients with concomitant HCM and AF was associated with a lower incidence of thromboembolic events when compared to antiplatelet therapy or no treatment. Treatment with DOACs was also associated with a lower incidence of thromboembolic events, major bleeding, and all-cause mortality when compared to VKAs.