Artificial intelligence and heart failure: A state-of-the-art review.

Heart failure (HF) is a heterogeneous syndrome affecting more than 60 million individuals globally. Despite recent advancements in understanding of the pathophysiology of HF, many issues remain including residual risk despite therapy, understanding the pathophysiology and phenotypes of patients with HF and preserved ejection fraction, and the challenges related to integrating a large amount of disparate information available for risk stratification and management of these patients. Risk prediction algorithms based on artificial intelligence (AI) may have superior predictive ability compared to traditional methods in certain instances. AI algorithms can play a pivotal role in the evolution of HF care by facilitating clinical decision making to overcome various challenges such as allocation of treatment to patients who are at highest risk or are more likely to benefit from therapies, prediction of adverse outcomes, and early identification of patients with subclinical disease or worsening HF. With the ability to integrate and synthesize large amounts of data with multidimensional interactions, AI algorithms can supply information with which physicians can improve their ability to make timely and better decisions. In this review, we provide an overview of the AI algorithms that have been developed for establishing early diagnosis of HF, phenotyping HF with preserved ejection fraction, and stratifying HF disease severity. This review also discusses the challenges in clinical deployment of AI algorithms in HF, and the potential path forward for developing future novel learning-based algorithms to improve HF care.

An Individual Patient-Level Meta-Analysis of Ischemic Versus Nonischemic Cardiomyopathy and Trajectory of Decongestion in Patients With Acute Decompensated Heart Failure.

Data are limited regarding the impact of ischemic cardiomyopathy (ICM) or non-ICM (NICM) on the trajectory of in-hospital decongestion among patients with acute decompensated heart failure (ADHF). Therefore, we aimed to assess the course of decongestion among patients admitted for ADHF by history of ICM and NICM. Patients included in the DOSE (Diuretic strategies in patients with acute decompensated heart failure), ROSE (ROSE acute heart failure randomized trial), and Ultrafiltration in decompensated heart failure with cardiorenal syndrome (CARRESS-HF) trials of patients with ADHF were categorized into ICM and NICM based on history. Among 762 patients included in our meta-analysis, 433 (56.8%) had a history of ICM. Patients with ICM were older (70.8 vs 63.9 years; p ≤0.001) and had higher rates of co-morbidities. After covariate adjustment, there was no significant differences between NICM and ICM regarding net fluid loss (4,952 vs 4,384 ml, p = 0.81) or mean change in serum N-terminal pro-brain natriuretic peptide (-2,162 vs -1,809 pg/ml, p = 0.092). Mean change in weight showed modest improvement in favor of patients with NICM, but this did not meet statistical significance (-8.24 vs -7.70 pounds, p = 0.068). After adjustment, there was no significant difference in the risk of 60-day composite all-cause mortality or hospitalization for HF among those with ICM versus NICM. Among patients with left ventricular ejection fraction <40%, NICM was associated with higher scoring on global sense of well-being (global visual analog scale; +25.5 vs +19.1, p = 0.023) and improvement in serum creatinine (-0.031 mg/100 ml vs +0.042 mg/100 ml, p = 0.009) at 72 hours. Among patients with left ventricular ejection fraction >40%, NICM was associated with decreased scoring on global visual analog scale at 72 hours (+15.7 vs +21.2, p = 0.049). In conclusion, more than half of the patients admitted for ADHF had ICM. History of ICM was not independently associated with a difference in course of decongestion, self-assessment of well-being and dyspnea, or short-term clinical outcomes.

Meta-Analysis Evaluating Risk of Hyperkalemia Stratified by Baseline MRA Usage in Patients with Heart Failure Receiving SGLT2 Inhibitors.

BACKGROUND: Both mineralocorticoid receptor antagonists (MRAs) and sodium-glucose co-transporter type 2 inhibitors (SGLT2is) have demonstrated beneficial reductions in cardiovascular outcomes. However, the risk of precipitating hyperkalemia with their concomitant usage remains unclear. METHODS: MEDLINE and Cochrane were searched from inception through March 2022. Randomized controlled trials on patients with heart failure (HF) evaluating the effect of SGLT2is on clinical outcomes between MRA users and non-users were considered for inclusion. Outcomes of interest were mild and moderate/severe hyperkalemia, for which hazard ratios (HR) were pooled using a random effects model. RESULTS: From the 972 articles retrieved from the initial search, three RCTs (n = 14,462 patients) were included in our meta-analysis. Pooled analysis demonstrated no significant difference in the incidence of mild hyperkalemia between MRA users (HR 0.82 [0.70-0.97]) and non-users (HR 0.95 [0.77-1.17]) (P-interaction = 0.28). The risk of severe hyperkalemia was significantly decreased in MRA users (HR 0.59 [0.44-0.78]; p = 0.0002; I2 = 0%) but not in non-users (HR 0.76 [0.56-1.02]; p = 0.07; I2 = 0%) (P-interaction = 0.22). Sensitivity analysis including patients with HF with reduced ejection fraction (HFrEF) revealed similar results across all subgroups, but no significant reduction in the incidence of mild hyperkalemia (HR 0.89 [0.76-1.04] was noted in MRA users with HFrEF. CONCLUSION: MRAs reduced the risk of mild or moderate/severe hyperkalemia, when added to SGLT2is. Future clinical trials should target scrupulous assessment of the risk of mild and moderate/severe hyperkalemia when used concomitantly with MRAs.

Inconsistent Outcome Reporting in Heart Failure Randomized Controlled Trials.

BACKGROUND: Randomized controlled trials (RCTs) may report outcomes different from those prespecified on trial-registration websites, protocols and statistical analysis plans (SAPs). This study sought to investigate the prevalence and characteristics of heart failure (HF) RCTs that report outcomes different from those prespecified. METHODS AND RESULTS: MEDLINE via PubMed was searched to include phase II-IV HF RCTs in 9 high-impact journals from 2010 to 2020. Outcomes reported in trial publications were compared with prespecified outcomes in protocols, registration websites and SAPs. We used the χ2 or Fisher exact test to analyze correlations between trial characteristics and inconsistencies. Among 216 trials, 32 inconsistencies were observed in 28 trials (13.0%). Among 32 inconsistencies, 2 (6.3%) pertained to omission of prespecified primary outcomes, 4 (12.5%) to omission of prespecified secondary outcomes, 2 (6.3%) to changing prespecified primary outcomes to secondary outcomes, and 2 (6.3%) to changing prespecified secondary outcomes to primary outcomes. Of the inconsistencies, 3 (9.4%) pertained to addition of new primary outcomes, 17 (53.1%) to addition of new secondary outcomes, and 2 (6.3%,) to changes in the timing of assessment of primary outcomes. The majority of the inconsistencies favored statistically significant findings; 78 (36.1%) were registered retrospectively. Single-center recruitment was associated with outcome inconsistencies (ß = -0.14; 95% CI, -0.22 - -0.01; P = 0.035). CONCLUSIONS: More than 1 in 10 trials reported outcomes inconsistent with those specified in trial registration websites, SAPs and protocols. An action plan is warranted to minimize selective reporting and improve transparency.

Effectiveness of endovascular repair versus open surgery for the treatment of thoracoabdominal aneurysm: A systematic review and meta analysis.

Background: Thoracoabdominal aortic aneurysms (TAAAs) are associated with significant comorbidities. The aim of our study is to compare the outcomes of open repair versus endovascular repair of TAAAs. Methods: A thorough literature search was conducted on MEDLINE, Embase, and Cochrane Central databases. The analysis included observational studies comparing the outcomes of surgical vs endovascular aneurysm repair (EVAR) of TAAA. Mortality, spinal cord ischemia (SCI), renal failure, stroke, paraplegia, and respiratory and cardiac problems were all included in the studies. The results were provided as relative risks (RRs) with 95% confidence intervals (CIs). These were then aggregated using an inverse variance weighted random-effects model, and the pooled analysis was displayed using forest plots. Results: This meta-analysis compromising of twelve studies revealed significant results, favoring endovascular repair versus open surgery for all-cause mortality (HR = 1.91; 95% CI: 1.68-2.18; P < 0.00001), SCI (HR = 1.62; 95% CI: 1.18-2.21; P = 0.003), respiratory complications (HR = 2.22; 95% CI: 1.78-2.77; P < 0.00001), and cardiac complications (HR = 1.66; 95% CI: 1.38-2.00; P < 0.00001). Upon subgroup analysis based on propensity matched, results were consistent and significant for the outcomes of all-cause mortality, cardiac complications, and respiratory complications. For the propensity unmatched subgroup, the incidence of all-cause mortality, SCI, respiratory complications, and cardiac complications were lower among endovascular repair cohort. Conclusion: Current evidence supports the use of endovascular repair over open surgery. However, there is a need to conduct dedicated randomized controlled trials to effectively compare and determine the benefits and risk of both strategies.

Utilization of Implantable Cardioverter-Defibrillators in Patients With Heart Transplant (from National Inpatient Sample Database).

Heart transplant (HT) recipients represent a unique and vulnerable population in whom medium and long-term outcomes are significantly affected by the risk of arrhythmias and sudden cardiac death. The use of implantable cardioverter-defibrillators (ICDs) in this population remains debated. A retrospective analysis of the National Inpatient Sample data between 2009 and 2018 was conducted. Hospitalization data on patients who underwent HT, or who had a preexisting HT, and who received a new ICD were included (excluding the preexisting ICD). Outcomes assessed included inpatient mortality, length of stay, and inflation-adjusted costs. We explored temporal trends in ICD placement and mean length of stay, and predictors of ICD placement. Between 2009 and 2018, 22,673 hospitalizations were recorded for HT, during which patients either received a concurrent new ICD placement (n = 70 [0.31%]) or no new ICD placement (n = 22,603 [99.7%]). During the same period, 146,555 admissions were recorded in patients with a history of HT. ICD placement in patients with a preexisting HT was associated with significantly higher inflation-adjusted costs ($55,680.7 vs $17,219.2; p <0.001). Predictors of ICD placement in preexisting patients with HT included cardiac arrest during hospitalization (odds ratio [OR]:14.3 [3.5 to 58.6]), drug abuse (OR:6.0 [1.3 to 27.1]), and previous PCI (OR:6.0 [2.1 to 17.3]). In conclusion, ICD placement in patients with HT history was associated with significantly higher inflation-adjusted costs. In patients with HT history, factors predicting ICD placement included cardiac arrest at hospitalization, previous PCI, and drug abuse.

Effects of COVID-19 pandemic and lockdown on lifestyle and mental health of students: A retrospective study from Karachi, Pakistan.

Introduction: Due to the COVID-19 pandemic, many countries imposed lockdowns on their citizens in an attempt to contain the disease. Pakistan is one of these countries. A government mandated lockdown can have mitigating psychological effects on young adults, out of which a large fraction is made up of students. This study aims to investigate the correlations between changes in sleep pattern, perception of time, and digital media usage. Furthermore, it explores the impact of these changes on the mental health of students of different educational levels. Methods: This cross-sectional study was conducted via a web-based questionnaire, from March 24 to April 26, 2020. The survey was targeted at students and 251 responses were obtained. It was a 5-section long questionnaire. The first section inquired about demographics of participants. Each of the other 4 sections was devoted to changes in sleep pattern, perception of time flow, digital media usage and mental health status of students. Close-ended questions with multiple choice responses, dichotomous, interval and 4-point Likert scales were used in the construction of the survey questionnaire. Chi2 T-tests multinomial and binary logistic regression were used as primary statistical tests. All data were analysed using Statistical Package for Social Science (SPSS) version 23.0 (IBM Corp., Armonk, NY). Results: Out of 251 adolescents that participated in our study, the majority (70.2%) were females. The mean age of the participants was 19.40 ± 1.62 years. Two-thirds of the respondents did not have much trouble falling asleep (66.5%). The analysis found no significant association between longer sleep periods and procrastination level (P = 0.054). Nearly three-fourths (72.9%) of our participants felt that getting through quarantine would have been more difficult if they did not have any electronic gadgets. Of these, a majority (85.8%) had a general feeling of tiredness and lacked motivation (P = 0.023). Additionally, a large number of students (69.7%) had reported that time is seemingly moving faster. A significant relationship between increased usage of electronic items and longer sleep periods was also noted (P = 0.005). With respect to the level of education, statistically significant values were noted for alarm use both before and after quarantine began (P = 0.021 and P = 0.004, respectively). Further analysis showed that there was a significant difference in the median difference of time spent on social media before the outbreak (3.0 ± 32.46) and time spent on social media after the outbreak (6.0 ± 3.52) in a single day (P = 0.000). Conclusions: Our research has revealed that due to the lockdown imposed by the government in response to COVID-19, the sleeping patterns of the students was affected the most. Our findings show that the increase in use of social media applications led to a widespread increase in the length of sleep, worsening of sleep habits (people sleeping at much later hours than usual), and a general feeling of tiredness. A general lack of recollection regarding what day of the week it was, as well as a change in the perceived flow of time were also notable. All these findings indicate the decline in mental health of students due to the lockdown. Promoting better sleep routines, minimising the use of digital media, and encouragement of students to take up more hobbies could collectively improve the health and mood of students in self-quarantine.

Introduction: En raison de la pandémie de COVID-19, de nombreux pays ont imposé des verrouillages à leurs citoyens pour tenter de contenir la maladie. Le Pakistan est l'un de ces pays. Un verrouillage mandaté par le gouvernement peut avoir des effets psychologiques atténuants sur les jeunes adultes, dont une grande partie est composée d'étudiants. Cette étude vise à étudier les corrélations entre les changements dans les habitudes de sommeil, la perception du temps et l'utilisation des médias numériques. De plus, il explore l'impact de ces changements sur la santé mentale des élèves de différents niveaux d'éducation. Méthodes: Cette étude transversale a été menée via un questionnaire en ligne, du 24 mars au 26 avril 2020. L'enquête visait les étudiants et 251 réponses ont été obtenues. C'était un questionnaire de 5 sections. La première section a posé des questions sur la démographie des participants. Chacune des 4 autres sections était consacrée aux changements dans les habitudes de sommeil, à la perception de l'écoulement du temps, à l'utilisation des médias numériques et à l'état de santé mentale des élèves. Des questions fermées avec des réponses à choix multiples, des échelles dichotomiques, d'intervalle et de Likert à 4 points ont été utilisées dans la construction du questionnaire d'enquête. Le chi carré, les tests T multinomiaux et la régression logistique binaire ont été utilisés comme tests statistiques primaires. Toutes les données ont été analysées en utilisant Statistical Package for Social Science (SPSS) version 23.0 (IBM Corp., Armonk, NY). Résultats: Sur 251 adolescents qui ont participé à notre étude, la majorité (70,2 %) étaient des femmes. L'âge moyen des participants était de 19,40 ± 1,62 ans. Les deux tiers des répondants n'avaient pas beaucoup de mal à s'endormir (66,5 %). L'analyse n'a trouvé aucune association significative entre des périodes de sommeil plus longues et le niveau de procrastination (p = 0,054). Près des trois quarts (72,9 %) de nos participants ont estimé que passer la quarantaine aurait été plus difficile s'ils n'avaient pas de gadgets électroniques. Parmi ceux-ci, une majorité (85,8 %) avait une sensation générale de fatigue et manquait de motivation (p = 0,023). De plus, un grand nombre d'étudiants (69,7 %) ont déclaré que le temps passe apparemment plus vite. Une relation significative entre une utilisation accrue des articles électroniques et des périodes de sommeil plus longues a également été notée (p = 0,005). En ce qui concerne le niveau d'éducation, des valeurs statistiquement significatives ont été notées pour l'utilisation des alarmes avant et après le début de la quarantaine (p = 0,021 et p = 0,004, respectivement). Une analyse plus approfondie a montré qu'il y avait une différence significative dans la différence médiane entre le temps passé sur les réseaux sociaux avant l'épidémie (3,0 ± 32,46) et le temps passé sur les réseaux sociaux après l'épidémie (6,0 ± 3,52) en une seule journée (p = 0,000). Conclusions: Notre recherche a révélé qu'en raison du verrouillage imposé par le gouvernement en réponse au COVID-19, les habitudes de sommeil des étudiants étaient les plus affectées. Nos résultats montrent que l'augmentation de l'utilisation des applications de médias sociaux a conduit à une augmentation généralisée de la durée du sommeil, à une aggravation des habitudes de sommeil (personnes qui dorment beaucoup plus tard que d'habitude) et à une sensation générale de fatigue. Un manque général de souvenir du jour de la semaine, ainsi qu'un changement dans l'écoulement perçu du temps, étaient également notables. Tous ces résultats indiquent le déclin de la santé mentale des étudiants en raison du verrouillage. La promotion de meilleures habitudes de sommeil, la minimisation de l'utilisation des médias numériques et l'encouragement des étudiants à adopter plus de passe-temps pourraient collectivement améliorer la santé et l'humeur des étudiants en quarantaine.

Meta-analysis evaluating the impact of chili-pepper intake on all-cause and cardiovascular mortality: A systematic review.

BACKGROUND: Dietetics today occupy a significant place in the field of research, helping to discover cardiovascular benefits of healthy diets and consumption of organic foods such as fruits, vegetables, legumes, nuts, and whole grains. One of the components of vegetable-based diet is chili pepper (CP) which has been found to affect all-cause mortality. METHODS: MEDLINE, EMBASE, Scopus, EBSCO, and Cochrane (Wiley) Central Register of Controlled Trials were searched from inception till January 9, 2020, identifying all relevant studies using keywords and truncations. Studies were included if (1) they were observational or randomized in nature (2) included patients consuming CP and (3) evaluated direct comparison between regular and rarely/never CP consumption. RESULTS: Our preliminary search yielded 6976 articles. Post exclusion and after full-text screening, four potential observational studies with a population of 570,762. Pooled analysis found reduced all-cause mortality in CP consumers compared to nonconsumers with a risk ratio (RR) of 0.75 [95% CI: 0.64-0.88; p = 0.0004; I 2 = 97%]. The RR for CVD, cancer related and CVA deaths were 0.74 [95% CI: 0.62-0.88; p = 0.0006, I 2 = 66%], 0.77 [95% CI: 0.71-0.84; p = 0.0001; I 2 = 49%] and 0.76 [95% CI: 0.36-1.60; p = 0.47; I2 = 93%], respectively. CONCLUSION: Statistically significant results of our analysis put forward a rationale indicating an association between lower risk of all-cause, cardiovascular and cancer related deaths and CP consumption.

Effect of iron supplementation in patients with heart failure and iron deficiency: A systematic review and meta-analysis.

BACKGROUND: The effectiveness of oral and intravenous iron supplementation in reducing the risk of mortality and hospitalizations in HF patients with iron deficiency is not well-established. METHODS: A thorough literature search was conducted across 2 electronic databases (Medline and Cochrane Central) from inception through March 2021. RCTs assessing the impact of iron supplementation on clinical outcomes in iron deficient HF patients were considered for inclusion. Primary end-points included all-cause mortality and HF hospitalization. Evaluations were reported as odds ratios (ORs) or risk ratios (RRs) with 95% confidence intervals (CI) and analysis was performed using a random effects model. I2 index was used to assess heterogeneity. RESULTS: From the 2599 articles retrieved from initial search, 10 potentially relevant studies (n = 2187 patients) were included in the final analysis. Both oral (OR: 0.93; 95% CI: 0.08-11.30; p = 0.951) and intravenous (OR: 0.97; 95% CI: 0.73-1.29; p = 0.840) iron supplementation did not significantly reduce all-cause mortality. However, intravenous iron supplementation significantly decreased the rates of overall (OR: 0.52; 95% CI: 0.33-0.81; p = 0.004) and HF (OR: 0.42; 95% CI: 0.22-0.80; p = 0.009) hospitalizations. In addition, intravenous ferric carboxymaltose therapy significantly reduced the time to first HF hospitalization or cardiovascular mortality (RR = 0.70; 95% CI = 0.50-1.00; p = 0.048), but had no effect on time to first cardiovascular death (RR: 0.94; 95% CI: 0.70-1.25; p = 0.655). CONCLUSION: Oral or intravenous iron supplementation did not reduce mortality in iron deficient HF patients. However, intravenous iron supplementation was associated with a significant decrease in overall and HF hospitalizations.