Validity of heart failure diagnoses in administrative databases: a systematic review and meta-analysis

Artificial Intelligence-Enabled Prediction of Heart Failure Risk From Single-Lead Electrocardiograms

Importance

Despite the availability of disease-modifying therapies, scalable strategies for heart failure (HF) risk stratification remain elusive. Portable devices capable of recording single-lead electrocardiograms (ECGs) may enable large-scale community-based risk assessment.

Objective

To evaluate whether an artificial intelligence (AI) algorithm can predict HF risk from noisy single-lead ECGs.

Design, Setting, and Participants

A retrospective cohort study of individuals without HF at baseline was conducted among individuals with conventionally obtained outpatient ECGs in the integrated Yale New Haven Health System (YNHHS) and prospective population-based cohorts of the UK Biobank (UKB) and the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Data analysis was performed from September 2023 to February 2025.

Exposure

AI-ECG-defined risk of left ventricular systolic dysfunction (LVSD).

Main Outcomes and Measures

Among individuals with ECGs, lead I ECGs were isolated and a noise-adapted AI-ECG model (to simulate ECG signals from wearable devices) trained to identify LVSD was deployed. The association of the model probability with new-onset HF, defined as the first HF hospitalization, was evaluated. The discrimination of AI-ECG was compared against 2 risk scores for new-onset HF (Pooled Cohort Equations to Prevent Heart Failure [PCP-HF] and Predicting Risk of Cardiovascular Disease Events [PREVENT] equations) using the Harrel C statistic, integrated discrimination improvement, and net reclassification improvement.

Results

There were 192 667 YNHHS patients (median [IQR] age, 56 [41-69] years; 111 181 women [57.7%]), 42 141 UKB participants (median [IQR] age, 65 [59-71] years; 21 795 women [51.7%]), and 13 454 ELSA-Brasil participants (median [IQR] age, 51 [45-58] years; 7348 women [54.6%]) with baseline ECGs. A total of 3697 (1.9%) developed HF in YNHHS over a median (IQR) of 4.6 (2.8-6.6) years, 46 (0.1%) in UKB over a median (IQR) of 3.1 (2.1-4.5) years, and 31 (0.2%) in ELSA-Brasil over a median (IQR) of 4.2 (3.7-4.5) years. A positive AI-ECG screening result for LVSD was associated with a 3- to 7-fold higher risk for HF, and each 0.1 increment in the model probability was associated with a 27% to 65% higher hazard across cohorts, independent of age, sex, comorbidities, and competing risk of death. AI-ECG's discrimination for new-onset HF was 0.723 (95% CI, 0.694-0.752) in YNHHS, 0.736 (95% CI, 0.606-0.867) in UKB, and 0.828 (95% CI, 0.692-0.964) in ELSA-Brasil. Across cohorts, incorporating AI-ECG predictions alongside PCP-HF and PREVENT equations was associated with a higher Harrel C statistic (difference in addition to PCP-HF, 0.080-0.107; difference in addition to PREVENT, 0.069-0.094). AI-ECG had an integrated discrimination improvement of 0.091 to 0.205 vs PCP-HF and 0.068 to 0.192 vs PREVENT; it had a net reclassification improvement of 18.2% to 47.2% vs PCP-HF and 11.8% to 47.5% vs PREVENT.

Conclusions and Relevance

Across multinational cohorts, a noise-adapted AI-ECG model estimated HF risk using lead I ECGs, suggesting a potential HF risk-stratification strategy requiring prospective study using wearable and portable ECG devices.

Epidemiology of hospitalized heart failure in France based on national data over 10 years, 2012-2022

AIMS

We aim to describe the incidence of HF hospitalization in France in the post-pandemic era, the prevalence of HF cases and patients' characteristics, management and outcomes while focusing on sex, age and socio-economic differences and to analyse time-trends between 2012 and 2022.

METHODS AND RESULTS

Based on the French health care database providing medical information for almost the whole French population, patients hospitalized for acute decompensated HF without history of HF in the 5 years were identified by the International Classification of Diseases - 10th revision (ICD-10) codes. In 2022, the estimated prevalence of HF was 1.7% in France and has increased until the COVID-19 pandemic and decreased thereafter. The incidence of acute HF decompensation reached 201.4 per 100 000 inhabitants and has decreased since 2012 (-1% per year). A significant increase of the HF incidence was found in men aged <45 years. Women aged <65 years were less likely to be admitted in a cardiac rehabilitation (CR) unit and had higher probability of one-year mortality compared with men of the same age. One-year mortality was significantly increased in patients from the most deprived area among extreme age group only (under 65 and ≥85 years). One-year rehospitalization rates have decreased significantly, particularly in men aged <75 years. A decrease in ACE/ARBs deliveries was observed in both men and women.

CONCLUSIONS

Despite the decrease in acute HF decompensation incidence and improvements in the management, the prevalence of HF remains stable in France and prognosis remains poor.

Comparative performance of risk prediction indices for mortality or readmission following heart failure hospitalization

AIMS

Risk prediction indices used in worsening heart failure (HF) vary in complexity, performance, and the type of datasets in which they were validated. We compared the performance of seven risk prediction indices in a contemporary cohort of patients hospitalized for HF.

METHODS AND RESULTS

We assessed the performance of the Length of stay and number of Emergency department visits in the prior 6 months (LE), Length of stay, number of Emergency department visits in the prior 6 months, and admission N-Terminal prohormone of brain natriuretic peptide (NT-proBNP (LENT), Length of stay, Acuity, Charlson co-morbidity index, and number of Emergency department visits in the prior 6 months (LACE), Get With The Guidelines Heart Failure (GWTG), Readmission Risk Score (RRS), Enhanced Feedback for Effective Cardiac Treatment model (EFFECT), and Acute Decompensated Heart Failure National Registry (ADHERE) risk indices among consecutive patients hospitalized for HF and discharged alive from January 2017 to December 2019 in a network of hospitals in England. The primary composite outcome was 30-day all-cause mortality or readmission. We assessed model discrimination and overall accuracy using the C-statistic (higher values, better) and Brier score (lower values, better), respectively. Among 1206 patients in the cohort, 45.0% were female, mean (SD) age was 76.6 (11.7) years, and mean (SD) left ventricular ejection fraction was 43.0% (11.6). At 30 days, 236 (19.6%) patients were readmitted and 28 (2.3%) patients died, with 264 (21.9%) patients experiencing either readmission or death. The LENT index offered the combination of greatest risk discrimination and accuracy for the primary composite outcome (C-statistic: 0.97; 95% CI 0.96, 0.98; 0.29; Brier score: 0.05). The LE (C-statistic: 0.95; 95% CI 0.93, 0.96; Brier score: 0.06) and LACE (C-statistic: 0.90; 95% CI 0.88, 0.92; Brier score 0.09) indices had high discrimination and accuracy. Discrimination and accuracy were modest with the RRS (C-statistic: 0.65; 95% CI 0.61, 0.69; Brier score: 0.16) and EFFECT (C-statistic: 0.64; 95% CI 0.60, 0.67; Brier score: 0.16) score; and poor with the GWTG-HF (C-statistic: 0.62; 95% CI 0.58, 0.66; Brier score: 0.17) and ADHERE (C-statistic: 0.54; 95% CI 0.50, 0.57; Brier score: 0.17) scores.

CONCLUSIONS

In a study that compared the performance of seven risk prediction indices in a contemporary cohort of patients hospitalized for HF, the simple LENT index offered the greatest combination of discrimination and accuracy for the primary composite outcome of 30-day all-cause mortality or readmission. This three-variable index -using length of hospital stay, preceding emergency department visits and admission NT-proBNP level- is a practical and reliable way to assess prognosis following hospitalization for HF.

Heart failure risk stratification using artificial intelligence applied to electrocardiogram images: a multinational study

BACKGROUND AND AIMS

Current heart failure (HF) risk stratification strategies require comprehensive clinical evaluation. In this study, artificial intelligence (AI) applied to electrocardiogram (ECG) images was examined as a strategy to predict HF risk.

METHODS

Across multinational cohorts in the Yale New Haven Health System (YNHHS), UK Biobank (UKB), and Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), individuals without baseline HF were followed for the first HF hospitalization. An AI-ECG model that defines cross-sectional left ventricular systolic dysfunction from 12-lead ECG images was used, and its association with incident HF was evaluated. Discrimination was assessed using Harrell's C-statistic. Pooled cohort equations to prevent HF (PCP-HF) were used as a comparator.

RESULTS

Among 231 285 YNHHS patients, 4472 had primary HF hospitalizations over 4.5 years (inter-quartile range 2.5-6.6). In UKB and ELSA-Brasil, among 42 141 and 13 454 people, 46 and 31 developed HF over 3.1 (2.1-4.5) and 4.2 (3.7-4.5) years. A positive AI-ECG screen portended a 4- to 24-fold higher risk of new-onset HF [age-, sex-adjusted hazard ratio: YNHHS, 3.88 (95% confidence interval 3.63-4.14); UKB, 12.85 (6.87-24.02); ELSA-Brasil, 23.50 (11.09-49.81)]. The association was consistent after accounting for comorbidities and the competing risk of death. Higher probabilities were associated with progressively higher HF risk. Model discrimination was 0.718 in YNHHS, 0.769 in UKB, and 0.810 in ELSA-Brasil. In YNHHS and ELSA-Brasil, incorporating AI-ECG with PCP-HF yielded a significant improvement in discrimination over PCP-HF alone.

CONCLUSIONS

An AI model applied to a single ECG image defined the risk of future HF, representing a digital biomarker for stratifying HF risk.

Clinical and research applications of natural language processing for heart failure

Natural language processing (NLP) is a burgeoning field of machine learning/artificial intelligence that focuses on the computational processing of human language. Researchers and clinicians are using NLP methods to advance the field of medicine in general and in heart failure (HF), in particular, by processing vast amounts of previously untapped semi-structured and unstructured textual data in electronic health records. NLP has several applications to clinical research, including dramatically improving processes for cohort assembly, disease phenotyping, and outcome ascertainment, among others. NLP also has the potential to improve direct clinical care through early detection, accurate diagnosis, and evidence-based management of patients with HF. In this state-of-the-art review, we present a general overview of NLP methods and review clinical and research applications in the field of HF. We also propose several potential future directions of this emerging and rapidly evolving technological breakthrough.

A machine learning tool for identifying newly diagnosed heart failure in individuals with known diabetes in primary care

AIMS

We aimed to create a predictive model utilizing machine learning (ML) to identify new cases of congestive heart failure (CHF) in individuals with diabetes in primary health care (PHC) through the analysis of diagnostic data.

METHODS

We used a sex- and age-matched case-control design. Cases of new CHF were identified across all outpatient care settings 2015-2022 (n = 9098). We included individuals 30 years and above, by sex and age groups of 30-65 years and >65 years. The controls (five per case) were sampled from the individuals in 2015-2022 without CHF at any time between 2010 and 2022, in total 45 490. From the stochastic gradient boosting (SGB) technique model, we obtained a rank of the 10 most important factors related to newly diagnosed CHF in individuals with diabetes, with the normalized relative influence (NRI) score and a corresponding odds ratio of marginal effects (ORME). Area under curve (AUC) was calculated.

RESULTS

For women 30-65 years and >65 years, we identified 488 and 3240 new cases of CHF, respectively, and men 30-65 years and >65 years 1196 and 4174 new cases. Among the 10 most important factors in the four groups (divided by sex and lower and higher age) for newly diagnosed CHF, we found the number of visits 12 months before diagnosis (NRI 44.3%-55.9%), coronary artery disease (NRI 2.9%-7.8%), atrial fibrillation and flutter (NRI 6.6%-12.2%) and 'abnormalities of breathing' (ICD-10 code R06) (NRI 2.6%-4.4%) were predictive in all groups. For younger women, a diagnosis of COPD (NRI 2.7%) contributed to the predictive effect, while for older women, oedema (NRI 3.1%) and number of years with diabetes (NRI 3.5%) contributed to the predictive effect. For men in both age groups, chronic renal disease had predictive effect (NRI 3.9%-5.1%) The model prediction of CHF among patients with diabetes was high, AUC around 0.85 for the four groups, and with sensitivity over 0.783 and specificity over 0.708 for all four groups.

CONCLUSIONS

An SGB model using routinely collected data about diagnoses and number of visits in primary care, can accurately predict risk for diagnosis of heart failure in individuals with diabetes. Age and sex difference in predictive factors warrant further examination.

Heart Failure in Atrial Fibrillation Subtypes in Women and Men in the Tromsø Study

Background

Atrial fibrillation (AF) and heart failure (HF) often coexist and impact morbidity and mortality. There is limited knowledge on the association of AF subtypes with HF according to sex.

Objectives

The purpose of this study was to explore sex-specific associations between AF subtypes and subsequent HF, identifying HF risk factors in participants with AF, and exploring the combined impact on mortality.

Methods

14,790 women and 13,181 men from the Tromsø Study were enrolled between 1994 and 2008 and followed for incident AF and HF through 2016. Cox regression was conducted to provide HRs and 95% CIs.

Results

Those with AF had higher risk of subsequent HF in both sexes compared to those without AF. Women with permanent AF had higher relative risk of HF than men (HR: 10.52; 95% CI: 8.72-12.70, and HR: 7.65; 95% CI: 6.40-9.15, respectively). Risk factors for HF in participants with AF included smoking in all, higher diastolic blood pressure and hypertension in women, underweight, obesity, and low alcohol consumption in men. All-cause mortality was higher in women with both subtypes (paroxysmal/persistent: HR: 2.10; 95% CI: 1.78-2.48, permanent: HR: 1.40, 95% CI: 1.14-1.72) and in men with paroxysmal/persistent AF (HR: 1.66; 95% CI: 1.40-1.96). Subsequent HF increased risk of mortality in both sexes.

Conclusions

All AF subtypes were associated with increased risk of HF. Smoking was a shared risk factor, while diastolic blood pressure and hypertension were specific to women, and underweight, obesity, and low alcohol intake were specific to men. Subsequent HF increased mortality risk in all.

Risk of Atherosclerotic Cardiovascular Disease After Chronic Obstructive Pulmonary Disease Hospitalization among Primary and Secondary Prevention Older Adults

BACKGROUND

Meta-analyses have suggested that the risk of cardiovascular disease events is significantly higher after a chronic obstructive pulmonary disease (COPD) exacerbation, but the populations at highest risk have not been well characterized to date.

METHODS AND RESULTS

The authors analyzed the risk of atherosclerotic cardiovascular disease (ASCVD) hospitalizations after COPD hospitalization compared with before COPD hospitalization and patient factors associated with ASCVD hospitalizations after COPD hospitalization among 2 high-risk patient cohorts. The primary outcome was risk of an ASCVD hospitalization composite outcome (myocardial infarction, coronary artery bypass graft, percutaneous coronary intervention, stroke, transient ischemic accident) after COPD hospitalization relative to before COPD hospitalization. Additional analyses evaluated for risk factors associated with the composite ASCVD hospitalization outcome. In the high-risk primary prevention cohort, the hazard ratio (HR) estimate following adjustment for the composite ASCVD hospitalization outcome after COPD hospitalization versus before COPD hospitalization for 30 days was 0.74 (95% CI, 0.66-0.82; P≤0.0001); for 90 days, 0.69 (95% CI, 0.64-0.75; P≤0.0001); and for 1 year, 0.78 (95% CI, 0.73-0.82; P≤0.0001). In the secondary prevention cohort, the HR for 30-day hospitalization was 1.15 (95% CI, 1.05-1.26; P=0.0036); 90-day hospitalization, 1.08 (95% CI, 1.01-1.15; P=0.0178); and 1-year hospitalization, 1.07 (95% CI, 1.02-1.11; P=0.0026). Among the 19 characteristics evaluated, hyperlipidemia and history of acute ASCVD event were associated with the highest risk of ASCVD events 1 year after COPD hospitalization in the high-risk primary and secondary prevention cohorts.

CONCLUSIONS

The risk of ASCVD hospitalization was higher in patients with established ASCVD and lower among high-risk patients without established ASCVD after-COPD hospitalization relative to before hospitalization. We identified multiple risk factors for ASCVD hospitalization after COPD hospitalization.

Diagnostic accuracy of case-identification algorithms for heart failure in the general population using routinely collected health data: a systematic review

BACKGROUND

Heart failure (HF), affecting 1-4% of adults in industrialized countries, is a major public health priority. Several algorithms based on administrative health data (HAD) have been developed to detect patients with HF in a timely and inexpensive manner, in order to perform real-world studies at the population level. However, their reported diagnostic accuracy is highly variable.

OBJECTIVE

To assess the diagnostic accuracy of validated HAD-based algorithms for detecting HF, compared to clinical diagnosis, and to investigate causes of heterogeneity.

METHODS

We included all diagnostic accuracy studies that utilized HAD for the diagnosis of congestive HF in the general adult population, using clinical examination or chart review as the reference standard. A systematic search of MEDLINE (1946-2023) and Embase (1947-2023) was conducted, without restrictions. The QUADAS-2 tool was employed to assess the risk of bias and concerns regarding applicability. Due to low-quality issues of the primary studies, associated with both the index test and the reference standard definition and conduct, and to the high level of clinical heterogeneity, a quantitative synthesis was not performed. Measures of diagnostic accuracy of the included algorithms were summarized narratively and presented graphically, by population subgroups.

RESULTS

We included 24 studies (161,524 patients) and extracted 36 algorithms. Algorithm selection was based on type of administrative data and DOR. Six studies (103,018 patients, 14 algorithms) were performed in the general outpatient population, with sensitivities ranging from 24.8 to 97.3% and specificities ranging from 35.6 to 99.5%. Eight studies (14,957 patients, 10 algorithms) included hospitalized patients with sensitivities ranging from 29.0 to 96.0% and specificities ranging from 65.8 to 99.2%. The remaining studies included subgroups of the general population or hospitalized patients with cardiologic conditions and were analyzed separately. Fourteen studies had one or more domains at high risk of bias, and there were concerns regarding applicability in 9 studies.

DISCUSSION

The considerable percentage of studies with a high risk of bias, together with the high clinical heterogeneity among different studies, did not allow to generate a pooled estimate of diagnostic accuracy for HAD-based algorithms to be used in an unselected general adult population.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO CRD42023487565.

Artificial Intelligence Enabled Prediction of Heart Failure Risk from Single-lead Electrocardiograms

Importance

Despite the availability of disease-modifying therapies, scalable strategies for heart failure (HF) risk stratification remain elusive. Portable devices capable of recording single-lead electrocardiograms (ECGs) can enable large-scale community-based risk assessment.

Objective

To evaluate an artificial intelligence (AI) algorithm to predict HF risk from noisy single-lead ECGs.

Design

Multicohort study.

Setting

Retrospective cohort of individuals with outpatient ECGs in the integrated Yale New Haven Health System (YNHHS) and prospective population-based cohorts of UK Biobank (UKB) and Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).

Participants

Individuals without HF at baseline.

Exposures

AI-ECG-defined risk of left ventricular systolic dysfunction (LVSD).

Main Outcomes and Measures

Among individuals with ECGs, we isolated lead I ECGs and deployed a noise-adapted AI-ECG model trained to identify LVSD. We evaluated the association of the model probability with new-onset HF, defined as the first HF hospitalization. We compared the discrimination of AI-ECG against two risk scores for new-onset HF (PCP-HF and PREVENT equations) using Harrel's C-statistic, integrated discrimination improvement (IDI), and net reclassification improvement (NRI).

Results

There were 192,667 YNHHS patients (age 56 years [IQR, 41-69], 112,082 women [58%]), 42,141 UKB participants (65 years [59-71], 21,795 women [52%]), and 13,454 ELSA-Brasil participants (56 years [41-69], 7,348 women [55%]) with baseline ECGs. A total of 3,697 developed HF in YNHHS over 4.6 years (2.8-6.6), 46 in UKB over 3.1 years (2.1-4.5), and 31 in ELSA-Brasil over 4.2 years (3.7-4.5). A positive AI-ECG screen was associated with a 3- to 7-fold higher risk for HF, and each 0.1 increment in the model probability portended a 27-65% higher hazard across cohorts, independent of age, sex, comorbidities, and competing risk of death. AI-ECG's discrimination for new-onset HF was 0.725 in YNHHS, 0.792 in UKB, and 0.833 in ELSA-Brasil. Across cohorts, incorporating AI-ECG predictions in addition to PCP-HF and PREVENT equations resulted in improved Harrel's C-statistic (ΔPCP-HF=0.112-0.114; ΔPREVENT=0.080-0.101). AI-ECG had IDI of 0.094-0.238 and 0.090-0.192, and NRI of 15.8%-48.8% and 12.8%-36.3%, vs. PCP-HF and PREVENT, respectively.

Conclusions and Relevance

Across multinational cohorts, a noise-adapted AI model defined HF risk using lead I ECGs, suggesting a potential portable and wearable device-based HF risk-stratification strategy.

Heart failure in first- and second-generation immigrants aged 18-54 years in Sweden: A national study

PURPOSE

We aimed at analysing the risk of congestive heart failure (CHF) among first- and second-generation immigrants in younger age groups.

METHODS

All individuals aged 18-54 years, n = 3 973 454 in the first-generation study and n = 3 817 560 in the second-generation study, were included. CHF was defined as at least one registered diagnosis in the National Patient Register between 1 January 1998 and 31 December 2018. Cox regression analysis was used to estimate the relative risk [hazard ratios (HRs) with 99% confidence intervals (CIs)] of incident CHF with adjustments for age, co-morbidities and socio-demographics.

RESULTS

In the first-generation study, a total of 85 719 cases of CHF were registered, 54 369 men and 31 350 women, where fully adjusted models showed HRs for all foreign-born men of 1.12 (99% CI 1.06-1.17) and for women of 0.99 (0.92-1.05). Groups with higher risk included men from Eastern Europe, Central Europe, Africa and Asia and women from Africa and Asia, and a lower risk was found among Latin American women. In the second-generation study, a total of 88 999 cases of CHF were registered, 58 403 men and 30 596 women, where fully adjusted models showed HRs for second-generation men of 1.04 (0.99-1.09) and women of 0.97 (0.90-1.04).

CONCLUSIONS

The higher risk in some foreign-born groups needs to be paid attention to in clinical practice. The fact that almost all increased risks were attenuated and absent in second-generation immigrants suggests that lifestyle and environmental factors are more important than genetic differences in the risk of CHF.

Maternal pulmonary hypertension and cardiopulmonary outcomes during delivery hospitalization in the United States: A nationwide study from 2016-2020

BACKGROUND

Maternal pulmonary hypertension can pose substantial morbidity and mortality risks, particularly during labor and delivery. Although maternal pulmonary hypertension is conventionally considered a contraindication to pregnancy, advances in the management of pH may contribute to improving outcomes.

OBJECTIVES

In this nationwide study, we aim to characterize the prevalence of maternal pulmonary hypertension in the United States and its association with adverse cardiopulmonary outcomes during delivery hospitalizations.

STUDY DESIGN

In this cross-sectional cohort study, we analyzed delivery hospitalizations in the National Inpatient Sample from 2016 to 2020. The primary exposure was maternal pulmonary hypertension. The primary outcome was a composite of maternal cardiopulmonary morbidity events during the delivery hospitalization including: death, heart failure, intraoperative heart failure, pulmonary edema, cardiac arrest, myocardial infarction, ventricular fibrillation, respiratory failure, pneumonia, acute kidney injury, and cardiac conversion. Propensity score matching was used to estimate the association between maternal pulmonary hypertension and adverse cardiopulmonary outcomes, adjusting for sociodemographic variables and validated clinical comorbidities as covariates. Secondary outcomes included mechanical circulatory support utilization, length of stay, and total hospitalization costs.

RESULTS

Among 18,161,315 delivery hospitalizations, 4,630 patients had pulmonary hypertension, yielding a maternal pulmonary hypertension prevalence of 25 per 100,000 delivery hospitalizations with a yearly trend of increasing prevalence (odds ratio = 1.06, 95 % CI 1.01 to 1.11, P = 0.028). After propensity score matching to create well-balanced groups, 4,560 patients with pulmonary hypertension were compared to 4,560 patients without pulmonary hypertension. In this confounder-adjusted analysis, the primary composite outcome of cardiopulmonary morbidity and mortality occurred in 41.1 % of the PH group compared to 14.4 % in the no PH group (adjusted odds ratio = 4.16, 95 % CI 3.32 to 5.23, P < 0.001). Additionally, patients with PH had a higher incidence of mechanical circulatory support use (adjusted odds ratio = 9.08, 95 % CI 1.14 to 71.81, P = 0.037), longer length of stay (length of stay ratio = 2.82, 95 % CI 2.74 to 2.9, P < 0.001) and higher total hospitalization costs (total cost ratio = 1.67, 95 % CI 1.52 to 1.85, P < 0.001).

CONCLUSIONS

Maternal pulmonary hypertension is increasing in prevalence and is strongly associated with adverse cardiopulmonary outcomes in the United States, with 41.1% of pH patients experiencing a composite outcome of cardiopulmonary morbidity and mortality during delivery hospitalization. Our findings emphasize the importance of caring for patients with maternal pulmonary hypertension in a multidisciplinary setting at high-acuity centers to ensure appropriate management of cardiopulmonary complications that arise during labor and delivery.

Usefulness and caveats of real-world data for research on hypertension and its association with cardiovascular or renal disease in Japan

The role of real-world data, collected from clinical practice rather than clinical trials, has become increasingly important for investigating real-life situations, such as treatment effects. In Japan, evidence on hypertension, cardiovascular diseases, and kidney diseases using real-world data is increasing. These studies are mainly based on "the insurer-based real-world data" collected as electronic records, including data from health check-ups and medical claims such as JMDC database, DeSC database, the Japan Health Insurance Association (JHIA) database, or National Databases of Health Insurance Claims and Specific Health Checkups (NDB). Based on the insurer-based real-world data, traditional but finely stratified associations between hypertension and cardiovascular or kidney diseases can be explored. The insurer-based real-world data are also useful for pharmacoepidemiological studies that capture the distribution and trends of drug prescriptions; combined with annual health check-up data, the effectiveness of drugs can also be examined. Despite the usefulness of insurer-based real-world data collected as electronic records from a wide range of populations, we must be cautious about several points, including issues regarding population uncertainty, the validity of cardiovascular outcomes, the accuracy of blood pressure, traceability, and biases, such as indication and immortal biases. While a large sample size is considered a strength of real-world data, we must keep in mind that it does not overcome the problem of systematic error. This review discusses the usefulness and pitfalls of insurer-based real-world data in Japan through recent examples of Japanese research on hypertension and its association with cardiovascular or kidney disease.

Differences in Pharmacological Treatment of Heart Failure Among Persons with or without Major Cognitive Disorder: A Cross-Sectional Study Based on National Registries in Sweden

INTRODUCTION

Comorbidities are common among older people, and during the last decade, a strong association between heart failure (HF) and cognitive impairment has been found. As much as 40-50% of individuals with HF will also have some degree of cognitive impairment. Previous studies report an undertreatment for some cardiovascular diseases in patients with major neurocognitive disorder (NCD).

OBJECTIVE

The aim of this present study was to explore differences in pharmacological treatment of HF in individuals diagnosed with HF with or without comorbidity of major NCD.

METHODS

This study combined data from three different Swedish national registers: the Swedish National Patient Register, the Swedish registry for cognitive/dementia disorders (SveDem), and the Swedish Prescribed Drug Register. A logistic regression model including variables for age, sex, major NCD, and nursing home residency was used to analyze associations between drug use and major NCD.

RESULTS

We found a lower prevalence of filled prescriptions of renin-angiotensin system (RAS) inhibitors, β-blockers (BBs), and mineralocorticoid receptor antagonists (MRAs) among patients with major NCD. Living in a nursing home was associated with lower prevalence of RAS inhibitors, BBs, digitalis glycosides, and sodium-glucose cotransporter-2 (SGLT2) inhibitors. Females were found to have higher odds of using BBs, loop diuretics and digitalis glycosides, and lower odds of using RAS inhibitors and SGLT2 inhibitors than males.

CONCLUSION

Our findings indicate that there is possible undertreatment among individuals with HF identified in specialized care with co-occurring major NCD. Major NCD was associated with less filled prescriptions of basal pharmacological treatments such as RAS inhibitors, BBs, and MRAs. Future research needs to not only investigate this relationship further but also focus on reasons for the undertreatment of HF and other comorbidities within this group.

Risk of short-term cardiovascular disease in relation to the mode of delivery in singleton pregnancies: a retrospective cohort study

Background

Cardiovascular disease (CVD) is increasing in prevalence and affects up to 4% of pregnancies in otherwise healthy persons. The specific factors that drive the development of CVD in pregnant people are poorly characterised. This study aimed to determine whether the mode of delivery in singletons affects the risk of cardiovascular morbidity and mortality within one year in patients without prior CVD.

Methods

We designed a retrospective cohort study utilising the Nationwide Readmissions Database (NRD) to identify singleton delivery hospitalisations in the United States from Jan 1, 2010 to Nov 30, 2018. International Classification of Disease (ICD) versions 9 and 10 codes were used to identify patients with readmission for CVD within the calendar year of index delivery. Patients aged 15-54 who underwent a singleton vaginal or caesarean delivery were included. Patients with pre-existing CVD hospitalisations before or during delivery, ectopic pregnancies, or abortive outcomes were excluded. Participant data was retrieved from the NRD database. The primary outcome was hospital readmission, defined by ICD 9 and 10 codes for fatal or non-fatal CVD in the same calendar year as delivery. Cox proportional hazard regression models were used to adjust for confounders. These included maternal age, hospital bed size, hospital type, hospital teaching status, income quartile, insurance, and year of delivery. Additional sub-analyses were performed adjusting for hypertensive disorders of pregnancy and diabetes mellitus.

Findings

Of the 14,179,299 singleton deliveries, 32% (n = 4,553,492) underwent a caesarean. CVD readmissions occurred in 255.2 per 100,000 (n = 11,710) caesarean deliveries compared with 133.9 per 100,000 (n = 12,507) vaginal deliveries (rate difference [RD], 121.4, 95% confidence interval [CI], 114.8-127.9; hazard ratio [HR] adjusted for all confounders including hypertensive disorders of pregnancy and diabetes mellitus was 1.42, 95% CI 1.35-1.50). This association was highest in the first 0-29 days following delivery (HR 1.68, 95% CI 1.59-1.78). The risk of readmission for CVD persisted for one year.

Interpretation

These findings suggest that caesarean delivery of singletons is associated with a higher risk of cardiovascular morbidity in patients without pre-existing CVD. This risk was highest in the first month but remained elevated for one year after delivery. These findings add to the accumulating evidence that undergoing caesarean delivery may have long-standing health implications and support the extension of the post-partum surveillance period. Limitations of this study include the lack of adjustment for body mass index, race, and parity. We were also unable to determine the reason for the caesarean delivery.

Funding

None.

Impact of the method of calculating 30-day readmission rate after hospitalization for heart failure. Data from the VancOuver CoastAL Acute Heart Failure (VOCAL-AHF) registry

BACKGROUND

Thirty-day readmission rate after heart failure (HF) hospitalization is widely used to evaluate healthcare quality. Methodology may substantially influence estimated rates. We assessed the impact of different definitions on HF and all-cause readmission rates.

METHODS

Readmission rates were examined in 1835 patients discharged following HF hospitalization using 64 unique definitions derived from five methodological factors: (1) International Classification of Diseases-10 codes (broad vs. narrow), (2) index admission selection (single admission only first-in-year vs. random sample; or multiple admissions in year with vs. without 30-day blanking period), (3) variable denominator (number alive at discharge vs. number alive at 30 days), (4) follow-up period start (discharge date vs. day following discharge), and (5) annual reference period (calendar vs. fiscal). The impact of different factors was assessed using linear regression.

RESULTS

The calculated 30-day readmission rate for HF varied more than two-fold depending solely on the methodological approach (6.5-15.0%). All-cause admission rates exhibited similar variation (18.8-29.9%). The highest rates included all consecutive index admissions (HF 11.1-15.0%, all-cause 24.0-29.9%), and the lowest only one index admission per patient per year (HF 6.5-11.3%, all-cause 18.8-22.7%). When including multiple index admissions and compared with blanking the 30-day post-discharge, not blanking was associated with 2.3% higher readmission rates. Selecting a single admission per year with a first-in-year approach lowered readmission rates by 1.5%, while random-sampling admissions lowered estimates further by 5.2% (P < 0.001).

CONCLUSION

Calculated 30-day readmission rates varied more than two-fold by altering methods. Transparent and consistent methods are needed to ensure reproducible and comparable reporting.

Adherence to guideline-recommended care of late-onset hypertension in females versus males: A population-based cohort study

BACKGROUND

Sex-based disparities in cardiovascular outcomes may be improved with appropriate hypertension management.

OBJECTIVE

To compare the evidence-based evaluation and management of females with late-onset hypertension compared to males in the contemporary era.

METHODS

Design: Retrospective population-based cohort study.

SETTING

Ontario, Canada.

PARTICIPANTS

Residents aged ≥66 years with newly diagnosed hypertension between January 1, 2010, and December 31, 2017.

EXPOSURE

Sex (female vs. male).

OUTCOMES AND MEASURES

We used Poisson and logistic regression to estimate adjusted sex-attributable differences in the performance of guideline-recommended lab investigations. We estimated adjusted differences in time to the prescription of, and type of, first antihypertensive medication prescribed between females and males, using Cox regression.

RESULTS

Among 111,410 adults (mean age 73 years, 53% female, median follow-up 6.8 years), females underwent a similar number of guideline-recommended investigations (adjusted incidence rate ratio, 0.997 [95% confidence interval [CI] 0.99-1.002]) compared to males. Females were also as likely to complete all investigations (0.70% females, 0.77% males; adjusted odds ratio, 0.96 [95% CI 0.83-1.11]). Females were slightly less likely to be prescribed medication (adjusted hazard ratio [aHR] 0.98 [95% CI 0.96-0.99]) or, among those prescribed, less likely to be prescribed first-line medication (aHR, 0.995 [95% CI 0.994-0.997]).

CONCLUSIONS

Compared to males, females with late-onset hypertension were equally likely to complete initial investigations with comparable prescription rates. These findings suggest that there may be no clinically meaningful sex-based differences in the initial management of late-onset hypertension to explain sex-based disparities in cardiovascular outcomes.

Association between copper intake and essential hypertension: dual evidence from Mendelian randomization analysis and the NHANES database

Background

Although previous studies have identified an association between trace elements and essential hypertension, the specific trace elements involved and the mechanisms of their association remain unclear. This study aimed to elucidate the relationship between various human trace elements and essential hypertension, thereby addressing existing gaps in the research.

Methods

This study employed two-sample, multivariate, and inverse Mendelian randomization (MR) analyses to investigate the causal relationship between 15 human trace elements as exposure factors and essential hypertension as the outcome. The analysis revealed a statistically significant association between copper intake and essential hypertension. Further validation was conducted using logistic regression models based on data from the National Health and Nutrition Examination Survey (NHANES).

Results

Eighteen trace elements were initially identified through searches in the GWAS database and PubMed. After screening, 15 trace elements were selected as potential exposure factors. MR analysis, utilizing the 2021 genome-wide dataset for essential hypertension, identified copper as a risk factor, showing a positive association with hypertension. Subsequent logistic regression analyses based on NHANES data further confirmed a significant association between dietary copper intake and the risk of essential hypertension, except for the 0.80-1.08 mg/d group in model 3 (p < 0.05). Restricted cubic spline (RCS) analysis indicated a nonlinear relationship between copper intake and the risk of developing essential hypertension.

Conclusion

This study demonstrates a significant association between copper intake and the development of essential hypertension. The findings suggest that higher copper intake is linked to an increased risk of hypertension, underscoring the need to monitor copper intake levels in the prevention and management of this condition.

Validation Study of Cardiovascular International Statistical Classification of Diseases and Related Health Problems, Tenth Edition, Australian Modification (ICD-10-AM) Codes in Administrative Healthcare Databases (ANZACS-QI 77)

BACKGROUND

Administrative healthcare databases can be utilised for research. The accuracy of the International Statistical Classification of Diseases and Related Health Problems, Tenth Edition, Australian Modification (ICD-10-AM) coding of cardiovascular conditions in New Zealand is not known and requires validation.

METHOD

International Statistical Classification of Diseases and Related Health Problems, Tenth Edition, Australian Modification coded discharges for acute coronary syndrome (ACS), heart failure (HF) and atrial fibrillation (AF), in both primary and secondary diagnostic positions, were identified from four district health boards between 1 January 2019 and 31 June 2019. A sample was randomly selected for retrospective clinician review for evidence of the coded diagnosis according to contemporary diagnostic criteria. Positive predictive values (PPVs) for ICD-10-AM coding vs clinician review were calculated. This study is also known as All of New Zealand, Acute Coronary Syndrome-Quality Improvement (ANZACS-QI) 77.

RESULTS

A total of 600 cases (200 for each diagnosis, 5.0% of total identified cases) were reviewed. The PPV of ACS was 93% (95% confidence interval [CI] 89%-96%), HF was 93% (95% CI 89%-96%) and AF was 96% (95% CI 92%-98%). There were no differences in PPV between district health boards. PPV for ACS were lower in Māori vs non-Māori (72% vs 96%; p=0.004), discharge from non-Cardiology vs Cardiology services (89% vs 96%; p=0.048) and ICD-10-AM coding for unstable angina vs myocardial infarction (81% vs 95%; p=0.011). PPV for HF were higher in the primary vs secondary diagnostic position (100% vs 89%; p=0.001).

CONCLUSIONS

The PPVs of ICD-10-AM coding for ACS, HF, and AF were high in this validation study. ICD-10-AM coding can be used to identify these diagnoses in administrative databases for the purposes of healthcare evaluation and research.

Assessment of the impact of resting heart rate on the risk of major adverse cardiovascular events after ischemic stroke: a retrospective observational study

BACKGROUND

Although elevated heart rate is a risk factor for cardiovascular morbidity and mortality in healthy people, the association between resting heart rate and major cardiovascular risk in patients after acute ischemic stroke remains debated. This study evaluated the association between heart rate and major adverse cardiovascular events after ischemic stroke.

METHODS

We conducted a retrospective cohort study analyzing data from the Chang Gung Research Database for 21,655 patients with recent ischemic stroke enrolled between January 1, 2010, and September 30, 2018. Initial in-hospital heart rates were averaged and categorized into 10-beats per minute (bpm) increments. The primary outcome was the composite of hospitalization for recurrent ischemic stroke, myocardial infarction, or all-cause mortality. Secondary outcomes were hospitalization for recurrent ischemic stroke, myocardial infarction, and heart failure. Hazard ratios and 95% confidence intervals (CIs) were estimated using Cox proportional hazards models, using the heart rate < 60 bpm subgroup as the reference.

RESULTS

After a median follow-up of 3.2 years, the adjusted hazard ratios for the primary outcome were 1.13 (95% CI: 1.01 to 1.26) for heart rate 60-69 bpm, 1.35 (95% CI: 1.22 to 1.50) for heart rate 70-79 bpm, 1.64 (95% CI: 1.47 to 1.83) for heart rate 80-89 bpm, and 2.08 (95% CI: 1.85 to 2.34) for heart rate ≥ 90 bpm compared with the reference group. Heart rate ≥ 70 bpm was associated with increased risk of all secondary outcomes compared with the reference group except heart failure.  CONCLUSIONS: Heart rate is a simple measurement with important prognostic implications. In patients with ischemic stroke, initial in-hospital heart rate was associated with major adverse cardiovascular events.

Sex Differences in the Development of Anthracycline-Associated Heart Failure

BACKGROUND

Female sex is frequently cited as a risk factor for anthracycline cardiotoxicity based on pediatric data, but the role of sex in the development of cardiotoxicity has not been clearly established in adults.

OBJECTIVES

To assess the effect of female sex on the development of incident heart failure (HF) in adult patients treated with anthracyclines.

METHODS

This was a retrospective cohort study of 1525 adult patients with no prior history of HF or cardiomyopathy who were treated with anthracyclines between 1992 and 2019. The primary outcome was new HF within 5 years of the first dose of anthracyclines. The effect of sex was assessed using Cox proportional hazards and competing risk models.

RESULTS

Over a median (IQR) follow-up of 1.02 (0.30-3.01) years, 4.78% of patients developed HF (44 men and 29 women). Female sex was not associated with the primary outcome in a multivariable Cox proportional hazards model (HR 0.87; 95% CI 0.53-1.43; P = 0.58). Similar results were observed in a multivariable model accounting for the competing risk of death (HR 0.94; 95% CI 0.39-2.25; P = 0.88). Age, coronary artery disease and hematopoietic stem cell transplant were associated with the primary outcome in a multivariable Cox proportional hazards model. Age and body mass index were associated with the primary outcome in a multivariable competing risk model.

CONCLUSIONS

In this large, single-center, retrospective cohort study, female sex was not associated with incident HF in adult patients treated with anthracyclines.

CONDENSED ABSTRACT

Female sex is frequently cited as a risk factor for anthracycline cardiotoxicity based on pediatric data, but the role of sex in the development of cardiotoxicity has not been clearly established in adults. In this retrospective cohort study, we assessed the effect of female sex on the development of incident heart failure in adult patients treated with anthracyclines. Using Cox proportional hazards and competing risk regression models, we found that there was no association between female sex and heart failure after treatment with anthracyclines.

A novel approach to accurately measuring the burden of hospitalisations for cardiovascular disease in people with diabetes: A pilot study

AIM

To determine the reliability of hospital discharge codes for heart failure (HF), acute myocardial infarction (AMI) and stroke compared with adjudicated diagnosis, and to pilot a scalable approach to adjudicate records on a population-based sample.

METHODS

A population-based sample of 685 people with diabetes admitted (1274 admissions) to one of three Australian hospitals during 2018-2020 were randomly selected for this study. All medical records were reviewed and adjudicated.

RESULTS

Cardiovascular diseases were the most common primary reason for hospitalisation in people with diabetes, accounting for ~17% (215/1274) of all hospitalisations, with HF as the leading cause. ICD-10 codes substantially underestimated HF prevalence and had the lowest agreement with the adjudicated diagnosis of HF (Kappa = 0.81), compared with AMI and stroke (Kappa ≥ 0.91). While ICD-10 codes provided suboptimal sensitivity (72%) for HF, the performance was better for AMI (sensitivity 84%; specificity 100%) and stroke (sensitivity 85%; specificity 100%). A novel approach to screen possible HF cases only required adjudicating 8% (105/1274) of records, correctly identified 78/81 of HF admissions and yielded 96% sensitivity and 98% specificity.

CONCLUSIONS

While ICD-10 codes appear reliable for AMI or stroke, a more complex diagnosis such as HF benefits from a two-stage process to screen for suspected HF cases that need adjudicating. The next step is to validate this novel approach on large multi-centre studies in diabetes.

Sex-Related Disparities in Cardiovascular Outcomes Among Older Adults With Late-Onset Hypertension

BACKGROUND

It is unclear whether sex-based differences in cardiovascular outcomes exist in late-onset hypertension.

METHODS

This is a population-based cohort study in Ontario, Canada of 266 273 adults, aged ≥66 years with newly diagnosed hypertension. We determined the incidence of the primary composite cardiovascular outcome (myocardial infarction, stroke, and congestive heart failure), all-cause mortality, and cardiovascular death by sex using Cox proportional hazard models adjusted for demographic factors and comorbidities.

RESULTS

The mean age of the total cohort was 74 years, and 135 531 (51%) were female. Over a median follow-up of 6.6 (4.7-9.0) years, females experienced a lower crude incidence rate (per 1000 person-years) than males for the primary composite cardiovascular outcome (287.3 versus 311.7), death (238.4 versus 251.4), and cardiovascular death (395.7 versus 439.6), P<0.001. The risk of primary composite cardiovascular outcome was lower among females (adjusted hazard ratio, 0.75 [95% CI, 0.73-0.76]; P<0.001) than in males. This was consistent after adjusting for the competing risk of all-cause death with a subdistributional hazard ratio, 0.88 ([95% CI, 0.86-0.91]; P<0.001).

CONCLUSIONS

Females had a lower risk of cardiovascular outcomes compared with males within a population characterized by advanced age and new hypertension. Our results highlight that the severity of outcomes is influenced by sex in relation to the age at which hypertension is diagnosed. Further studies are required to identify sex-specific variations in the diagnosis and management of late-onset hypertension due to its high incidence in this group.

Diagnostic accuracy, clinical characteristics, and prognostic differences of patients with acute myocarditis according to inclusion criteria

INTRODUCTION

The diagnosis of acute myocarditis (AM) is complex due to its heterogeneity and typically is defined by either Electronic Healthcare Records (EHRs) or advanced imaging and endomyocardial biopsy, but there is no consensus. We aimed to investigate the diagnostic accuracy of these approaches for AM.

METHODS

Data on ICD 10th Revision(ICD-10) codes corresponding to AM were collected from two hospitals and compared to cardiac magnetic resonance (CMR)-confirmed or clinically suspected (CS)-AM cases with respect to diagnostic accuracy, clinical characteristics, and all-cause mortality. Next, we performed a review of published AM studies according to inclusion criteria.

RESULTS

We identified 291 unique admissions with ICD-10 codes corresponding to AM in the first three diagnostic positions. The positive predictive value of ICD-10 codes for CMR-confirmed or CS-AM was 36%, and patients with CMR-confirmed or CS-AM had a lower all-cause mortality than those with a refuted diagnosis (P = 0.019). Using an unstructured approach, patients with CMR-confirmed and CS-AM had similar demographics, comorbidity profiles and survival over a median follow-up of 52 months (P = 0.72). Our review of the literature confirmed our findings. Outcomes for patients included in studies using CMR-confirmed criteria were favourable compared to studies with endomyocardial biopsy-confirmed AM cases.

CONCLUSION

ICD-10 codes have poor accuracy in identification of AM cases and should be used with caution in clinical research. There are important differences in management and outcomes of patients according to the selection criteria used to diagnose AM. Potential selection biases must be considered when interpreting AM cohorts and requires standardization of inclusion criteria for AM studies.

A Capture-Recapture-based Ascertainment Probability Weighting Method for Effect Estimation With Under-ascertained Outcomes

Outcome under-ascertainment, characterized by the incomplete identification or reporting of cases, poses a substantial challenge in epidemiologic research. While capture-recapture methods can estimate unknown case numbers, their role in estimating exposure effects in observational studies is not well established. This paper presents an ascertainment probability weighting framework that integrates capture-recapture and propensity score weighting. We propose a nonparametric estimator of effects on binary outcomes that combines exposure propensity scores with data from two conditionally independent outcome measurements to simultaneously adjust for confounding and under-ascertainment. Demonstrating its practical application, we apply the method to estimate the relationship between health care work and coronavirus disease 2019 testing in a Swedish region. We find that ascertainment probability weighting greatly influences the estimated association compared to conventional inverse probability weighting, underscoring the importance of accounting for under-ascertainment in studies with limited outcome data coverage. We conclude with practical guidelines for the method's implementation, discussing its strengths, limitations, and suitable scenarios for application.

The influence of diagnostic subgroups, patient- and hospital characteristics for the validity of cardiovascular diagnoses-Data from a Norwegian hospital trust

BACKGROUND

Cardiovascular discharge diagnoses may serve as endpoints in epidemiological studies if they have a high validity. Aim was to study if diagnoses-specific characteristics like type, sub-categories, and position of cardiovascular diagnoses affected diagnostic accuracy.

METHODS

Patients (n = 7,164) with a discharge diagnosis of acute myocardial infarction, heart failure or cerebrovascular disease were included. Data were presented as positive predictive values (PPV) and sensitivity.

RESULTS

PPV was high (≥88%) for acute myocardial infarction (n = 2,189) (except for outpatients). For heart failure (n = 4,026) PPV was 67% overall, but higher (>99%) when etiology or echocardiography was included. For hemorrhagic (n = 257) and ischemic (n = 1,034) strokes PPVs were 87% and 80%, respectively, with sensitivity of 79% and 75%. Transient ischemic attacks (n = 926) had PPV 56%, but sensitivity 86%. Primary diagnoses showed higher validity than subsequent diagnoses and inpatient diagnoses were more valid than outpatient diagnoses (except for transient ischemic attack). The diagnoses of acute myocardial infarction and heart failure where most valid when placed at cardiology units, while ischemic stroke when discharged from an internal medicine unit.

CONCLUSIONS

The diagnoses of acute myocardial infarction and stroke had excellent validity when placed during hospital stays. Similarly, heart failure diagnoses had excellent validity when echocardiography was performed before placing the diagnosis, while overall the diagnoses of heart failure and transient ischemic attack were less valid. In conclusion, the results indicate that cardiovascular diagnoses based on objective findings such as acute myocardial infarction and stroke have excellent validity and may be used as endpoints in clinical epidemiological studies with less rigid validation.

Scalable Risk Stratification for Heart Failure Using Artificial Intelligence applied to 12-lead Electrocardiographic Images: A Multinational Study

Background

Current risk stratification strategies for heart failure (HF) risk require either specific blood-based biomarkers or comprehensive clinical evaluation. In this study, we evaluated the use of artificial intelligence (AI) applied to images of electrocardiograms (ECGs) to predict HF risk.

Methods

Across multinational longitudinal cohorts in the integrated Yale New Haven Health System (YNHHS) and in population-based UK Biobank (UKB) and Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), we identified individuals without HF at baseline. Incident HF was defined based on the first occurrence of an HF hospitalization. We evaluated an AI-ECG model that defines the cross-sectional probability of left ventricular dysfunction from a single image of a 12-lead ECG and its association with incident HF. We accounted for the competing risk of death using the Fine-Gray subdistribution model and evaluated the discrimination using Harrel's c-statistic. The pooled cohort equations to prevent HF (PCP-HF) were used as a comparator for estimating incident HF risk.

Results

Among 231,285 individuals at YNHHS, 4472 had a primary HF hospitalization over 4.5 years (IQR 2.5-6.6) of follow-up. In UKB and ELSA-Brasil, among 42,741 and 13,454 people, 46 and 31 developed HF over a follow-up of 3.1 (2.1-4.5) and 4.2 (3.7-4.5) years, respectively. A positive AI-ECG screen portended a 4-fold higher risk of incident HF among YNHHS patients (age-, sex-adjusted HR [aHR] 3.88 [95% CI, 3.63-4.14]). In UKB and ELSA-Brasil, a positive-screen ECG portended 13- and 24-fold higher hazard of incident HF, respectively (aHR: UKBB, 12.85 [6.87-24.02]; ELSA-Brasil, 23.50 [11.09-49.81]). The association was consistent after accounting for comorbidities and the competing risk of death. Higher model output probabilities were progressively associated with a higher risk for HF. The model's discrimination for incident HF was 0.718 in YNHHS, 0.769 in UKB, and 0.810 in ELSA-Brasil. Across cohorts, incorporating model probability with PCP-HF yielded a significant improvement in discrimination over PCP-HF alone.

Conclusions

An AI model applied to images of 12-lead ECGs can identify those at elevated risk of HF across multinational cohorts. As a digital biomarker of HF risk that requires just an ECG image, this AI-ECG approach can enable scalable and efficient screening for HF risk.

Hospital Readmissions for Fluid Overload among Individuals with Diabetes and Diabetic Kidney Disease: Risk Factors and Multivariable Prediction Models

AIMS

Hospital readmissions due to recurrent fluid overload in diabetes and diabetic kidney disease can be avoided with evidence-based interventions. We aimed to identify at-risk patients who can benefit from these interventions by developing risk prediction models for readmissions for fluid overload in people living with diabetes and diabetic kidney disease.

METHODS

This was a single-center retrospective cohort study of 1,531 adults with diabetes and diabetic kidney disease hospitalized for fluid overload, congestive heart failure, pulmonary edema, and generalized edema between 2015 and 2017. The multivariable regression models for 30-day and 90-day readmission for fluid overload were compared with the LACE score for discrimination, calibration, sensitivity, specificity, and net reclassification index (NRI).

RESULTS

Readmissions for fluid overload within 30 days and 90 days occurred in 8.6% and 17.2% of patients with diabetes, and 8.2% and 18.3% of patients with diabetic kidney disease, respectively. After adjusting for demographics, comorbidities, clinical parameters, and medications, a history of alcoholism (HR 3.85, 95% CI: 1.41-10.55) and prior hospitalization for fluid overload (HR 2.50, 95% CI: 1.26-4.96) were independently associated with 30-day readmission in patients with diabetic kidney disease, as well as in individuals with diabetes. Additionally, current smoking, absence of hypertension, and high-dose intravenous furosemide were also associated with 30-day readmission in individuals with diabetes. Prior hospitalization for fluid overload (HR 2.43, 95% CI: 1.50-3.94), cardiovascular disease (HR 1.44, 95% CI: 1.03-2.02), eGFR ≤45 mL/min/1.73 m2 (HR 1.39, 95% CI: 1.003-1.93) was independently associated with 90-day readmissions in individuals with diabetic kidney disease. Additionally, thiazide prescription at discharge reduced 90-day readmission in diabetic kidney disease, while the need for high-dose intravenous furosemide predicted 90-day readmission in diabetes. The clinical and clinico-psychological models for 90-day readmission in individuals with diabetes and diabetic kidney disease had better discrimination and calibration than the LACE score. The NRI for the clinico-psychosocial models to predict 30- and 90-day readmissions in diabetes was 22.4% and 28.9%, respectively. The NRI for the clinico-psychosocial models to predict 30- and 90-day readmissions in diabetic kidney disease was 5.6% and 38.9%, respectively.

CONCLUSION

The risk models can potentially be used to identify patients at risk of readmission for fluid overload for evidence-based interventions, such as patient education or transitional care programs to reduce preventable hospitalizations.

Accuracy of heart failure ascertainment using routinely collected healthcare data: a systematic review and meta-analysis

BACKGROUND

Ascertainment of heart failure (HF) hospitalizations in cardiovascular trials is costly and complex, involving processes that could be streamlined by using routinely collected healthcare data (RCD). The utility of coded RCD for HF outcome ascertainment in randomized trials requires assessment. We systematically reviewed studies assessing RCD-based HF outcome ascertainment against "gold standard" (GS) methods to study the feasibility of using such methods in clinical trials.

METHODS

Studies assessing International Classification of Disease (ICD) coded RCD-based HF outcome ascertainment against GS methods and reporting at least one agreement statistic were identified by searching MEDLINE and Embase from inception to May 2021. Data on study characteristics, details of RCD and GS data sources and definitions, and test statistics were reviewed. Summary sensitivities and specificities for studies ascertaining acute and prevalent HF were estimated using a bivariate random effects meta-analysis. Heterogeneity was evaluated using I2 statistics and hierarchical summary receiver operating characteristic (HSROC) curves.

RESULTS

A total of 58 studies of 48,643 GS-adjudicated HF events were included in this review. Strategies used to improve case identification included the use of broader coding definitions, combining multiple data sources, and using machine learning algorithms to search free text data, but these methods were not always successful and at times reduced specificity in individual studies. Meta-analysis of 17 acute HF studies showed that RCD algorithms have high specificity (96.2%, 95% confidence interval [CI] 91.5-98.3), but lacked sensitivity (63.5%, 95% CI 51.3-74.1) with similar results for 21 prevalent HF studies. There was considerable heterogeneity between studies.

CONCLUSIONS

RCD can correctly identify HF outcomes but may miss approximately one-third of events. Methods used to improve case identification should also focus on minimizing false positives.

Early repeat hospitalization for fluid overload in individuals with cardiovascular disease and risks: a retrospective cohort study

AIMS

Fluid overload is a common manifestation of cardiovascular and kidney disease and a leading cause of hospitalizations. To identify patients at risk of recurrent severe fluid overload, we evaluated the incidence and risk factors associated with early repeat hospitalization for fluid overload among individuals with cardiovascular disease and risks.

METHODS

Single-center retrospective cohort study of 3423 consecutive adults with an index hospitalization for fluid overload between January 2015 and December 2017 and had cardiovascular risks (older age, diabetes mellitus, hypertension, dyslipidemia, kidney disease, known cardiovascular disease), but excluded if lost to follow-up or eGFR < 15 ml/min/1.73 m2. The outcome was early repeat hospitalization for fluid overload within 30 days of discharge.

RESULTS

The mean age was 73.9 ± 11.6 years and eGFR was 54.1 ± 24.6 ml/min/1.73 m2 at index hospitalization. Early repeat hospitalization for fluid overload occurred in 291 patients (8.5%). After adjusting for demographics, comorbidities, clinical parameters during index hospitalization and medications at discharge, cardiovascular disease (adjusted odds ratio, OR 1.66, 95% CI 1.27-2.17), prior hospitalization for fluid overload within 3 months (OR 2.52, 95% CI 1.17-5.44), prior hospitalization for any cause in within 6 months (OR 1.33, 95% CI 1.02-1.73) and intravenous furosemide use (OR 1.58, 95% CI 1.10-2.28) were associated with early repeat hospitalization for fluid overload. Higher systolic BP on admission (OR 0.992, 95% 0.986-0.998) and diuretic at discharge (OR 0.50, 95% CI 0.26-0.98) reduced early hospitalization for fluid overload.

CONCLUSION

Patients at-risk of early repeat hospitalization for fluid overload may be identified using these risk factors for targeted interventions.

Dynamic Prognosis Prediction for Patients on DAPT After Drug-Eluting Stent Implantation: Model Development and Validation

BACKGROUND

The rapid evolution of artificial intelligence (AI) in conjunction with recent updates in dual antiplatelet therapy (DAPT) management guidelines emphasizes the necessity for innovative models to predict ischemic or bleeding events after drug-eluting stent implantation. Leveraging AI for dynamic prediction has the potential to revolutionize risk stratification and provide personalized decision support for DAPT management.

METHODS AND RESULTS

We developed and validated a new AI-based pipeline using retrospective data of drug-eluting stent-treated patients, sourced from the Cerner Health Facts data set (n=98 236) and Optum's de-identified Clinformatics Data Mart Database (n=9978). The 36 months following drug-eluting stent implantation were designated as our primary forecasting interval, further segmented into 6 sequential prediction windows. We evaluated 5 distinct AI algorithms for their precision in predicting ischemic and bleeding risks. Model discriminative accuracy was assessed using the area under the receiver operating characteristic curve, among other metrics. The weighted light gradient boosting machine stood out as the preeminent model, thus earning its place as our AI-DAPT model. The AI-DAPT demonstrated peak accuracy in the 30 to 36 months window, charting an area under the receiver operating characteristic curve of 90% [95% CI, 88%-92%] for ischemia and 84% [95% CI, 82%-87%] for bleeding predictions.

CONCLUSIONS

Our AI-DAPT excels in formulating iterative, refined dynamic predictions by assimilating ongoing updates from patients' clinical profiles, holding value as a novel smart clinical tool to facilitate optimal DAPT duration management with high accuracy and adaptability.

Widening ethnic inequities in heart failure incidence in New Zealand

OBJECTIVE

Ethnic inequities in heart failure (HF) have been documented in several countries. This study describes New Zealand (NZ) trends in incident HF hospitalisation by ethnicity between 2006 and 2018.

METHODS

Incident HF hospitalisations in ≥20-year-old subjects were identified through International Classification of Diseases, 10th Revision-coded national hospitalisation records. Incidence was calculated for different ethnic, sex and age groups and were age standardised. Trends were estimated with joinpoint regression.

RESULTS

Of 116 113 incident HF hospitalisations, 12.8% were Māori, 5.7% Pacific people, 3.0% Asians and 78.6% Europeans/others. 64% of Māori and Pacific patients were aged <70 years, compared with 37% of Asian and 19% of European/others. In 2018, incidence rate ratios compared with European/others were 6.0 (95% CI 4.9 to 7.3), 7.5 (95% CI 6.0 to 9.4) and 0.5 (95% CI 0.3 to 0.8) for Māori, Pacific people and Asians aged 20-49 years; 3.7 (95% CI 3.4 to 4.0), 3.6 (95% CI 3.2 to 4.1) and 0.5 (95% CI 0.4 to 0.6) for Māori, Pacific people and Asians aged 50-69 years; and 1.5 (95% CI 1.4 to 1.6), 1.5 (95% CI 1.3 to 1.7) and 0.5 (95% CI 0.5 to 0.6) for Māori, Pacific people and Asians aged ≥70 years. Between 2006 and 2018, ethnicity-specific rates diverged in ≥70-year-old subjects due to a decline in European/others (annual percentage change (APC) -2.0%, 95% CI -2.5% to -1.6%) and Asians (APC -3.3%, 95% CI -4.4% to -2.1%), but rates remained unchanged for Māori and Pacific people. In contrast, regardless of ethnicity, rates either increased or remained unchanged in <70-year-old subjects.

CONCLUSION

Ethnic inequities in incident HF hospitalisation have widened in NZ over the past 13 years. Urgent action is required to address the predisposing factors that lead to development of HF in Maori and Pacific people.

Association Between Use of Fluoroquinolones and Risk of Mitral or Aortic Valve Regurgitation: A Nationwide Cohort Study

Biological plausibility suggests that fluoroquinolones may lead to mitral valve regurgitation or aortic valve regurgitation (MR/AR) through a collagen degradation pathway. However, available real-world studies were limited and yielded inconsistent findings. We estimated the risk of MR/AR associated with fluoroquinolones compared with other antibiotics with similar indications in a population-based cohort study. We identified adult patients who initiated fluoroquinolones or comparison antibiotics from the nationwide Taiwanese claims database. Patients were followed for up to 60 days after cohort entry. Cox regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of MR/AR comparing fluoroquinolones to comparison antibiotics after 1:1 propensity score (PS) matching. All analyses were conducted by type of fluoroquinolone (fluoroquinolones as a class, respiratory fluoroquinolones, and non-respiratory fluoroquinolones) and comparison antibiotic (amoxicillin/clavulanate or ampicillin/sulbactam, extended-spectrum cephalosporins). Among 6,649,284 eligible patients, the crude incidence rates of MR/AR ranged from 1.44 to 4.99 per 1,000 person-years across different types of fluoroquinolones and comparison antibiotics. However, fluoroquinolone use was not associated with an increased risk in each pairwise PS-matched comparison. HRs were 1.00 (95% CI, 0.89-1.11) for fluoroquinolones as a class, 0.96 (95% CI, 0.83-1.12) for respiratory fluoroquinolones, and 0.87 (95% CI, 0.75-1.01) for non-respiratory fluoroquinolones, compared with amoxicillin/clavulanate or ampicillin/sulbactam. Results were similar when fluoroquinolones were compared with extended-spectrum cephalosporins (HRs of 0.96, 95% CI, 0.82-1.12, HR, 1.05, 95% CI, 0.86-1.28, and HR, 0.88, 95% CI, 0.75-1.03, respectively). This large-scale cohort study did not find a higher risk of MR/AR with different types of fluoroquinolones in the adult population.

Retrospective comparison of traditional and artificial intelligence-based heart failure phenotyping in a US health system to enable real-world evidence

OBJECTIVE

Quantitatively evaluate the quality of data underlying real-world evidence (RWE) in heart failure (HF).

DESIGN

Retrospective comparison of accuracy in identifying patients with HF and phenotypic information was made using traditional (ie, structured query language applied to structured electronic health record (EHR) data) and advanced (ie, artificial intelligence (AI) applied to unstructured EHR data) RWE approaches. The performance of each approach was measured by the harmonic mean of precision and recall (F1 score) using manual annotation of medical records as a reference standard.

SETTING

EHR data from a large academic healthcare system in North America between 2015 and 2019, with an expected catchment of approximately 5 00 000 patients.

POPULATION

4288 encounters for 1155 patients aged 18-85 years, with 472 patients identified as having HF.

OUTCOME MEASURES

HF and associated concepts, such as comorbidities, left ventricular ejection fraction, and selected medications.

RESULTS

The average F1 scores across 19 HF-specific concepts were 49.0% and 94.1% for the traditional and advanced approaches, respectively (p<0.001 for all concepts with available data). The absolute difference in F1 score between approaches was 45.1% (98.1% relative increase in F1 score using the advanced approach). The advanced approach achieved superior F1 scores for HF presence, phenotype and associated comorbidities. Some phenotypes, such as HF with preserved ejection fraction, revealed dramatic differences in extraction accuracy based on technology applied, with a 4.9% F1 score when using natural language processing (NLP) alone and a 91.0% F1 score when using NLP plus AI-based inference.

CONCLUSIONS

A traditional RWE generation approach resulted in low data quality in patients with HF. While an advanced approach demonstrated high accuracy, the results varied dramatically based on extraction techniques. For future studies, advanced approaches and accuracy measurement may be required to ensure data are fit-for-purpose.

Association Between End-Stage Liver Disease and Incident Heart Failure in an Integrated Health System

BACKGROUND

End-stage liver disease (ESLD) and heart failure (HF) often coexist and are associated with significant morbidity and mortality. However, the true incidence of HF among patients with ESLD remains understudied.

OBJECTIVE

This study aims to evaluate the association between ESLD and incident HF in a real-world clinical cohort.

DESIGN AND PARTICIPANTS

A retrospective electronic health records database analysis of individuals with ESLD and frequency-matched controls without ESLD in a large integrated health system.

MAIN MEASURES

The primary outcome was incident HF, which was defined by the International Classification of Disease codes and manually adjudicated by physician reviewers. The Kaplan-Meier method was used to estimate the cumulative incidence of HF. Multivariate proportional hazards models adjusted for shared metabolic factors (diabetes, hypertension, chronic kidney disease, coronary heart disease, body mass index) were used to compare the risk of HF in patients with and without ESLD.

KEY RESULTS

Of 5004 patients (2502 with ESLD and 2502 without ESLD), the median (Q1-Q3) age was 57.0 (55.0-65.0) years, 59% were male, and 18% had diabetes. Over a median (Q1-Q3) follow-up of 2.3 (0.6-6.0) years, 121 incident HF cases occurred. Risk for incident HF was significantly higher for patients with ESLD compared with the non-ESLD group (adjusted HR: 4.67; 95% CI: 2.82-7.75; p < 0.001), with the majority of the ESLD group (70.7%) having HF with preserved ejection fraction (ejection fraction ≥ 50%).

CONCLUSION

ESLD was significantly associated with a higher risk of incident HF, independent of shared metabolic risk factors, with the predominant phenotype being HF with preserved ejection fraction.

Sodium-Glucose Cotransporter-2 Inhibitors After Acute Myocardial Infarction in Patients With Type 2 Diabetes: A Population-Based Investigation

Background Whether the early use of sodium-glucose cotransporter-2 (SGLT2) inhibitors have cardioprotective effects following acute myocardial infarction is unknown. Thus, we aimed to evaluate the association between the early initiation of SGLT2 inhibitors and cardiac event rates in patients with diabetes with acute myocardial infarction undergoing percutaneous coronary intervention. Methods and Results Based on the National Health Insurance claims data in South Korea, patients who received percutaneous coronary intervention for acute myocardial infarction between 2014 and 2018 were analyzed. Patients given SGLT2 inhibitors or other glucose-lowering drugs were matched based on a propensity score. The primary end point was a composite of all-cause mortality and hospitalizations for heart failure. Major adverse cardiac events (a composite of all-cause death, nonfatal myocardial infarction, and ischemic stroke) were compared as the secondary end point. After 1:2 propensity score matching, the SGLT2 inhibitors group (938 patients) and the no use of SGLT2 inhibitors group (1876 patients) were compared. During a median follow-up of 2.1 years, the early use of SGLT2 inhibitors was associated with lower risks of both the primary end point (9.8% versus 13.9%; adjusted hazard ratio [HR], 0.68 [95% CI, 0.54-0.87]; P=0.002) and secondary end point (9.1% versus 11.6%; adjusted HR, 0.77 [95% CI, 0.60-0.99]; P=0.04). All-cause mortality and hospitalizations for heart failure were also significantly lower in early users of SGLT2 inhibitors. Conclusions The early use of SGLT2 inhibitors in patients with diabetes treated with percutaneous coronary intervention for acute myocardial infarction was associated with a significantly lower risk of cardiovascular events, including all-cause mortality, hospitalizations for heart failure, and major adverse cardiac events.

Preserved Ratio Impaired Spirometry in Relationship to Cardiovascular Outcomes: A Large Prospective Cohort Study

BACKGROUND

Preserved ratio impaired spirometry (PRISm) findings are a heterogeneous condition characterized by a normal FEV₁ to FVC ratio with underlying impairment of pulmonary function. Data relating to the association of baseline and trajectories of PRISm findings with diverse cardiovascular outcomes are sparse.

RESEARCH QUESTION

How do baseline and trajectories of PRISm findings impact subsequent cardiovascular events?

STUDY DESIGN AND METHODS

In the UK Biobank cohort study, we included participants free of cardiovascular disease (CVD) with spirometry (FEV₁ and FVC values) at baseline (2006-2010). Participants with baseline spirometry and follow-up spirometry (2014-2020) were included in the lung function trajectory analysis. Cox proportional hazards multivariate regression was performed to evaluate the outcomes of major adverse cardiovascular events (MACEs), incident myocardial infarction (MI), stroke, heart failure (HF), and CVD mortality in association with lung function.

RESULTS

For baseline analysis (329,954 participants), the multivariate adjusted hazard ratios (HRs) for participants had PRISm findings (vs normal spirometry findings) were 1.26 (95% CI, 1.17-1.35) for MACE, 1.12 (95% CI, 1.01-1.25) for MI, 1.88 (95% CI, 1.72-2.05) for HF, 1.26 (95% CI, 1.13-1.40) for stroke, and 1.55 (95% CI, 1.37-1.76) for CVD mortality, respectively. A total of 22,781 participants underwent follow-up spirometry after an average of 8.9 years. Trajectory analysis showed that persistent PRISm findings (HR, 1.96; 95% CI, 1.24-3.09) and airflow obstruction (HR, 1.43; 95% CI, 1.00-2.04) was associated with a higher incidence of MACE vs consistently normal lung function. Compared with persistent PRISm findings, changing from PRISm to normal spirometry findings was associated with a lower incidence of MACE (HR, 0.42; 95% CI, 0.19-0.99).

INTERPRETATION

Individuals with baseline or persistent PRISm findings were at a higher risk of diverse cardiovascular outcomes even after adjusting for a wide range of confounding factors. However, individuals who transitioned from PRISm to normal findings showed a similar cardiovascular risk as those with normal lung function.

Identifying encephalopathy in patients admitted to an intensive care unit: Going beyond structured information using natural language processing

Background

Encephalopathy is a severe co-morbid condition in critically ill patients that includes different clinical constellation of neurological symptoms. However, even for the most recognised form, delirium, this medical condition is rarely recorded in structured fields of electronic health records precluding large and unbiased retrospective studies. We aimed to identify patients with encephalopathy using a machine learning-based approach over clinical notes in electronic health records.

Methods

We used a list of ICD-9 codes and clinical concepts related to encephalopathy to define a cohort of patients from the MIMIC-III dataset. Clinical notes were annotated with MedCAT and vectorized with a bag-of-word approach or word embedding using clinical concepts normalised to standard nomenclatures as features. Machine learning algorithms (support vector machines and random forest) trained with clinical notes from patients who had a diagnosis of encephalopathy (defined by ICD-9 codes) were used to classify patients with clinical concepts related to encephalopathy in their clinical notes but without any ICD-9 relevant code. A random selection of 50 patients were reviewed by a clinical expert for model validation.

Results

Among 46,520 different patients, 7.5% had encephalopathy related ICD-9 codes in all their admissions (group 1, definite encephalopathy), 45% clinical concepts related to encephalopathy only in their clinical notes (group 2, possible encephalopathy) and 38% did not have encephalopathy related concepts neither in structured nor in clinical notes (group 3, non-encephalopathy). Length of stay, mortality rate or number of co-morbid conditions were higher in groups 1 and 2 compared to group 3. The best model to classify patients from group 2 as patients with encephalopathy (SVM using embeddings) had F1 of 85% and predicted 31% patients from group 2 as having encephalopathy with a probability >90%. Validation on new cases found a precision ranging from 92% to 98% depending on the criteria considered.

Conclusions

Natural language processing techniques can leverage relevant clinical information that might help to identify patients with under-recognised clinical disorders such as encephalopathy. In the MIMIC dataset, this approach identifies with high probability thousands of patients that did not have a formal diagnosis in the structured information of the EHR.

Feasibility and accuracy of linking a heart failure registry to the national claims database using indirect identifiers

BACKGROUND

Heart failure (HF) registries include rich data on patient inclusion characteristics, but follow-up information is often incomplete. Medicoadministrative databases may provide less clinical information than registries, e.g. on left ventricular ejection fraction (LVEF), but long-term data are exhaustive and reliable. The combination of the two types of database is therefore appealing, but the feasibility and accuracy of such linking are largely unexplored.

AIMS

To assess the feasibility and accuracy of linking an HF registry (FRESH; FREnch Survey on Heart Failure) with the French National Healthcare System database (SNDS).

METHODS

A probabilistic algorithm was developed to link and match patient data included in the FRESH HF registry with anonymized records from the SNDS, which include: hospitalizations and diagnostic codes; all care-related reimbursements by national health system; and deaths. Consistency was assessed between deaths recorded in the registry and in the SNDS. A comparison between the two databases was carried out on several identifiable clinical characteristics (history of HF hospitalization, diabetes, atrial fibrillation, chronic bronchopneumopathy, severe renal failure and stroke) and on events during 1-year follow-up after inclusion.

RESULTS

Of 2719 patients included in the FRESH registry (1049 during decompensation; 1670 during outpatient follow-up), 1885 could be matched with a high accuracy of 94.3% for deaths. Mortality curves were superimposable, including curves according to type of HF and LVEF. The rates of missing data in the FRESH registry were 2.3-8.4% for clinical characteristics and 17.5% for hospitalizations during follow-up. The discrepancy rate for clinical characteristics was 3-13%. Hospitalization rates were significantly higher in the SNDS than in the registry cohort.

CONCLUSIONS

The anonymous matching of an HF research cohort with a national health database is feasible, with a significant proportion of patients being accurately matched, and facilitates combination of clinical data and a reduced rate of losses to follow-up.

Clinical characteristics and mortality of patients with heart failure in Southern Sweden from 2013 to 2019: a population-based cohort study

OBJECTIVES

To describe clinical characteristics and prognosis related to heart failure (HF) phenotypes in a community-based population by applying a novel algorithm to obtain ejection fractions (EF) from electronic medical records.

DESIGN

Retrospective population-based cohort study.

SETTING

Data were collected for all patients with HF in Southwest Sweden. The region consists of three acute care hospitals, 40 inpatient wards, 2 emergency departments, 30 outpatient specialty clinics and 48 primary healthcare.

PARTICIPANTS

8902 patients had an HF diagnosis based on the International Classification of Diseases, Tenth Revision during the study period. Patients <18 years as well as patients declining to participate were excluded resulting in a study population of 8775 patients.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome measure was distribution of HF phenotypes by echocardiography. The secondary outcome measures were 1 year all-cause mortality and HR for all-cause mortality using Cox regression models.

RESULTS

Out of 8775 patients with HF, 5023 (57%) had a conclusive echocardiography distributed into HF with reduced EF (35%), HF with mildly reduced EF (27%) and HF with preserved EF (38%). A total of 43% of the cohort did not have a conclusive echocardiography, and therefore no defined phenotype (HF-NDP). One-year all-cause mortality was 42% within the HF-NDP group and 30% among those with a conclusive EF. The HR of all-cause mortality in the HF-NDP group was 1.27 (95% CI 1.17 to 1.37) when compared with the confirmed EF group. There was no significant difference in survival within the HF phenotypes.

CONCLUSIONS

This population-based study showed a distribution of HF phenotypes that varies from those in selected HF registries, with fewer patients with HF with reduced EF and more patients with HF with preserved EF. Furthermore, 1-year all-cause mortality was significantly higher among patients with HF who had not undergone a conclusive echocardiography at diagnosis, highlighting the importance of correct diagnostic procedure to improve treatment strategies and outcomes.

Adherence and persistence to pharmacotherapy in patients with heart failure: a nationwide cohort study, 2014-2020

AIMS

We aimed to study initiation, adherence, and long-term persistence to beta-blockers (BB), renin-angiotensin system inhibitors (RASi), and mineralocorticoid receptor antagonists (MRA) in a nationwide cohort of patients with heart failure (HF).

METHODS

Patients aged 18-80 years in Norway with a first diagnosis of HF from 2014 until 2020 that survived ≥30 days were identified from the Norwegian Patient Registry and linked to the Norwegian Prescription Database. We collected information about BB, RASi [angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers (ARB), and angiotensin receptor-neprilysin inhibitors (ARNI)], and MRA. Dual HF therapy was defined as taking at least two out of three drug classes, whereas triple HF therapy was defined as taking all three. Initiation (time to initiation) and persistence (time to discontinuation using a grace period of 30 days) of HF drugs was calculated by the Kaplan-Meier method, followed to outcome of interest, death, or December 2020. One-year adherence was measured as proportion of days covered (PDC) using a cut-off at 80%. For adherence and persistence measurements, we allowed for maximum 60 days of stockpiling and switching within drug groups. We performed sensitivity analyses to test the robustness of our findings.

RESULTS

Out of 54 899 patients included in the cohort, 75%, 69%, and 21% initiated a BB, RASi, and MRA, respectively, whereas 13% did not receive any. Dual and triple HF therapy was prescribed to 61% and 16%, respectively. The proportion of adherent patients during the first year following initiation was 83%, 81%, 84%, and 61% for BB, RASi, ARNI, and MRA, whereas 42% and 5% were adherent to dual and triple HF therapy, respectively. From 2 to 5 years following initiation, persistence decreased from 58% to 38%, 57% to 37%, and 31% to 15% for BB, RASi, and MRA, respectively. Within the RASi group, persistence was higher for ARNI than for ACEI and ARB. There were no major changes in either initiation or adherence of the drug classes from 2014 to 2019, except for an increase in initiation and adherence of MRA.

CONCLUSIONS

We found low adherence to dual and triple HF therapies in this nationwide cohort study of newly diagnosed HF patients. Efforts are needed to increase adherence and persistence to HF therapies into clinical practice, emphasizing maintenance of multiple drug therapies in patients with such an indication.

Administrative Data in Cardiovascular Research-A Comparison of Polish National Health Fund and CRAFT Registry Data

(1) Background: Administrative data allows for time- and cost-efficient acquisition of large volumes of individual patient data invaluable for evaluation of the prevalence of diseases and clinical outcomes. The aim of the study was to evaluate the accuracy of data collected from the Polish National Health Fund (NHF), from a researcher's perspective, in regard to a cohort of atrial fibrillation patients. (2) Methods: NHF data regarding atrial fibrillation and common cardiovascular comorbidities was compared with the data collected manually from the individual patients' health records (IHR) collected in the retrospective CRAFT registry (NCT02987062). (3) Results: Data from the NHF underestimated the proportion of patients with AF (NHF = 83% vs. IHR = 100%) while overestimating the proportion of patients with other cardiovascular comorbidities in the cohort. Significantly higher CHA2DS2VASc (Median, [Q1-Q3]) (NHF: 1, [0-2]; vs. IHR: 1, [0-1]; p < 0.001) and HAS-BLED (Median, [Q1-Q3]) (NHF: 4, [2-6] vs. IHR: 3, [2-5]; p < 0.001) scores were calculated according to NHF in comparison to IHR data, respectively. (4) Conclusions: Clinical researchers should be aware that significant differences between IHR and billing data in cardiovascular research can be observed which should be acknowledged while drawing conclusions from administrative data-based cohorts. Natural Language Processing of IHR could further increase administrative data quality in the future.

Systematic approach to outcome assessment from coded electronic healthcare records in the DaRe2THINK NHS-embedded randomized trial. EUROPEAN HEART JOURNAL. DIGITAL HEALTH 2022;3:426-436. [PMID: 36712153 PMCID: PMC9708037 DOI: 10.1093/ehjdh/ztac046]

Aims

Improving the efficiency of clinical trials is key to their continued importance in directing evidence-based patient care. Digital innovations, in particular the use of electronic healthcare records (EHRs), allow for large-scale screening and follow up of participants. However, it is critical these developments are accompanied by robust and transparent methods that can support high-quality and high clinical value research.

Methods and results

The DaRe2THINK trial includes a series of novel processes, including nationwide pseudonymized pre screening of the primary-care EHR across England, digital enrolment, remote e-consent, and 'no-visit' follow up by linking all primary- and secondary-care health data with patient-reported outcomes. DaRe2THINK is a pragmatic, healthcare-embedded randomized trial testing whether earlier use of direct oral anticoagulants in patients with prior or current atrial fibrillation can prevent thromboembolic events and cognitive decline (www.birmingham.ac.uk/dare2think). This study outlines the systematic approach and methodology employed to define patient information and outcome events. This includes transparency on all medical code lists and phenotypes used in the trial across a variety of national data sources, including Clinical Practice Research Datalink Aurum (primary care), Hospital Episode Statistics (secondary care), and the Office for National Statistics (mortality).

Conclusion

Co-designed by a patient and public involvement team, DaRe2THINK presents an opportunity to transform the approach to randomized trials in the setting of routine healthcare, providing high-quality evidence generation in populations representative of the community at risk.

Evaluation of a Beta-Blocker-Edema-Loop Diuretic Prescribing Cascade: A Prescription Sequence Symmetry Analysis

BACKGROUND

Drug-related adverse events associated with antihypertensive therapy may result in subsequent prescribing of other potentially harmful medications, known as prescribing cascades. The aim of this study was to assess the magnitude and characteristics of a beta-blocker-edema-loop diuretic prescribing cascade.

METHODS

A prescription sequence symmetry analysis was used to assess loop diuretic initiation before and after initiation of beta-blockers among patients 20 years or older without heart failure, atrial fibrillation, other arrythmias, or use of calcium channel blocker within a U.S. private insurance claims database (2005-2018). The temporality of loop diuretic initiation relative to a beta-blocker or negative control (renin-angiotensin system blocker) initiation was tabulated. Secular trend-adjusted sequence ratios (aSRs) with 95% confidence intervals (CIs) compared the initiation of loop diuretic 90 days before and after initiation of beta-blockers.

RESULTS

Among 988,675 beta-blocker initiators, 9,489 patients initiated a new loop diuretic prescription 90 days after and 5,245 patients before beta-blocker initiation, resulting in an aSR of 1.78 (95% CI, 1.72-1.84). An estimated 1.72 beta-blocker initiators per 100 patient-years experienced the prescribing cascade in the first 90 days. The aSR was disproportionately higher among older adults (aSR 1.97), men (aSR 2.25), and patients who initiated metoprolol tartrate (aSR 2.48), labetalol (aSR 2.18), or metoprolol succinate (aSR 2.11). Negative control results (aSR 1.09, 95% CI, 1.05-1.13) generally corroborated our findings, but suggested modest within-person time-varying confounding.

CONCLUSIONS

We observed excess use of loop diuretics following beta-blocker initiation that was only partially explained by secular trends or hypertension progression.

International Classification of Diseases Codes are Useful in Identifying Cirrhosis in Administrative Databases

BACKGROUND

Health administrative databases are essential to define patient populations, make socioeconomic predictions, and facilitate medical research and healthcare planning. The accuracy of this data is dependent on valid codes/coding algorithms.

AIMS

The aim of this study was to systematically identify and summarize the validity of International Classification of Diseases (ICD) codes for identifying patients with cirrhosis in administrative data.

METHODS

Electronic databases, MEDLINE (via Ovid), EMBASE (via Ovid), the Web of Science, and CINAHL (via EBSCOhost), were searched for validation studies which compared ICD codes related to cirrhosis to a clinical reference standard, and reported statistical measures of performance.

RESULTS

Fourteen studies were included in the review. There was a large variation in the algorithms used to validate ICD codes to diagnose cirrhosis. Despite the variation, the positive predictive value (PPV) was greater than 84% and the specificity was greater than 75% in the majority of the studies. The negative predictive value (NPV) was lower, but still was associated with values greater than 70% in the majority of studies. Sensitivity data varied significantly with values ranging from 0.27 to 99%.

CONCLUSIONS

Evaluated ICD codes for cirrhosis, including codes for chronic liver disease, cirrhosis-specific codes, and cirrhosis-related complications, have demonstrated variable sensitivity and reasonable specificity for the identification of cirrhosis. Additional research is needed to maximize the identification of persons with cirrhosis to avoid underestimating the burden of disease.

Long-Term Beta-Blocker Therapy in Patients With Stable Coronary Artery Disease After Percutaneous Coronary Intervention

Background

It is unclear whether beta-blocker treatment is advantageous in patients with stable coronary artery disease (CAD) who underwent percutaneous coronary intervention (PCI). We evaluated the clinical impact of long-term beta-blocker maintenance in patients with stable CAD after PCI with drug-eluting stent (DES).

Methods

From a nationwide cohort database, we identified the stable CAD patients without current or prior history of myocardial infarction or heart failure who underwent DES implantation. An intention-to-treat principle was used to analyze the impact of beta-blocker treatment on long-term outcomes of major adverse cardiovascular events (MACE) composed of cardiovascular death, myocardial infarction, and hospitalization with heart failure.

Results

After stabilized inverse probability of treatment weighting, a total of 78,380 patients with stable CAD was enrolled; 45,746 patients with and 32,634 without beta-blocker treatment. At 5 years after PCI with a 6-month quarantine period, the adjusted incidence of MACE was significantly higher in patients treated with beta-blockers [10.0 vs. 9.1%; hazard ratio (HR) 1.11, 95% CI 1.06–1.16, p &lt; 0.001] in an intention-to-treat analysis. There was no significant difference in all-cause death between patients treated with and without beta-blockers (8.1 vs. 8.2%; HR 0.99, 95% CI 0.94–1.04, p = 0.62). Statistical analysis with a time-varying Cox regression and rank-preserving structure failure time model revealed similar results to the intention-to-treat analysis.

Conclusions

Among patients with stable CAD undergoing DES implantation, long-term maintenance with beta-blocker treatment might not be associated with clinical outcome improvement.

Trial Registration

ClinicalTrial.gov (NCT04715594).

Risks of Comorbidities in Patients With Palmoplantar Pustulosis vs Patients With Psoriasis Vulgaris or Pompholyx in Korea

Importance

Palmoplantar pustulosis (PPP) has been reported to be accompanied by systemic conditions. However, the risks of comorbidities in patients with PPP have rarely been evaluated.

Objective

To assess the risks of comorbidities in patients with PPP compared with patients with psoriasis vulgaris or pompholyx.

Design, Setting, and Participants

This nationwide population-based cross-sectional study used data from the Korean National Health Insurance database and the National Health Screening Program collected from January 1, 2010, to December 31, 2019. Data were analyzed from July 1, 2020, to October 31, 2021. Korean patients diagnosed with PPP, psoriasis vulgaris, or pompholyx who visited a dermatologist between January 1, 2010, and December 31, 2019, were enrolled.

Exposures

Presence of PPP.

Main Outcomes and Measures

The risks of comorbidities among patients with PPP vs patients with psoriasis vulgaris or pompholyx were evaluated using a multivariable logistic regression model.

Results

A total of 37 399 patients with PPP (mean [SD] age, 48.98 [17.20] years; 51.7% female), 332 279 patients with psoriasis vulgaris (mean [SD] age, 47.29 [18.34] years; 58.7% male), and 365 415 patients with pompholyx (mean [SD] age, 40.92 [17.63] years; 57.4% female) were included in the analyses. Compared with patients with pompholyx, those with PPP had significantly higher risks of developing psoriasis vulgaris (adjusted odds ratio [aOR], 72.96; 95% CI, 68.19-78.05; P < .001), psoriatic arthritis (aOR, 8.06; 95% CI, 6.55-9.92; P < .001), ankylosing spondylitis (aOR, 1.91; 95% CI, 1.61-2.27; P < .001), type 1 diabetes (aOR, 1.33; 95% CI, 1.16-1.52; P < .001), type 2 diabetes (aOR, 1.33; 95% CI, 1.29-1.38; P < .001), Graves disease (aOR, 1.25; 95% CI, 1.11-1.42; P < .001), Crohn disease (aOR, 1.63; 95% CI, 1.11-2.40; P = .01), and vitiligo (aOR, 1.87; 95% CI, 1.65-2.12; P < .001) after adjusting for demographic covariates. The risks of ankylosing spondylitis (aOR, 1.37; 95% CI, 1.16-1.62; P < .001) and Graves disease (aOR, 1.40; 95% CI, 1.23-1.58; P < .001) were significantly higher among patients with PPP vs psoriasis vulgaris. However, the risks of psoriatic arthritis (aOR, 0.54; 95% CI, 0.47-0.63; P < .001), systemic lupus erythematosus (aOR, 0.67; 95% CI, 0.46-0.97; P = .04), Sjögren syndrome (aOR, 0.70; 95% CI, 0.50-0.96; P = .03), systemic sclerosis (aOR, 0.29; 95% CI, 0.11-0.77; P = .01), vitiligo (aOR, 0.53; 95% CI, 0.47-0.60; P < .001), and alopecia areata (aOR, 0.88; 95% CI, 0.81-0.95; P = .001) were significantly lower among those with PPP vs psoriasis vulgaris.

Conclusions and Relevance

The results of this cross-sectional study suggest that patients with PPP have an overlapping comorbidity profile with patients with psoriasis vulgaris but not patients with pompholyx. However, the risks of comorbidities among patients with PPP may be substantially different from those among patients with psoriasis vulgaris.