Citations

Institution: Mayo Clinic Clear

Showing 10 of 108152 citations.

Intraoperative Radiation Therapy for Recurrent Cervical and Endometrial Cancer: Predicting Morbidity and Mortality in a Contemporary Cohort
Howlett, Lindsay N. , Fadadu, Priyal , Grcevich, Leah, Fought, Angela J. , McGree, Michaela E. , Giannini, Andrea , Butler, Kristina , Marnholtz, Amanda A., Haddock, Michael G. , Garda, Allison E. , Dowdy, Sean C. , Kumar, Amanika , Fadadu, Priyal P , Grcevich, Leah O, Butler, Kristina A., Garda, Allison E , Kumar, Amanika , Tortorella, Lucia , Langstraat, Carrie

Cancers | October 28, 2024 | Volume: 16 | Issue: 21 | Pages: 3628

10.3390/cancers16213628 | PMID 39518067 Get It

Background/Objectives: Our objective was to describe the use of intraoperative radiation therapy (IORT) for the treatment of recurrent/persistent cervical or endometrial cancer and assess predictors of postoperative complications and 3-year mortality. Methods: In this multi-site retrospective study, data were abstracted for recurrent/persistent endometrial or cervical cancer patients who underwent IORT from June 2004 to May 2021. Complications were graded on the six-point Accordion scale. Variables associated with complications were analyzed with univariate logistic regression, while variables associated with death within 3 years were analyzed with Cox proportional hazards modeling. Survival was analyzed with the Kaplan–Meier method. Results: Eighty patients had planned IORT for recurrent/persistent endometrial (n = 35) or cervical cancer (n = 45). The mean age of the cohort was 56.8 years (SD = 13.7), and the median disease-free interval from primary disease to recurrence was 20.0 months (IQR 10.0–63.1). The overall survival at 3 years was 48.6% (95% CI: 38.3–61.6%) with a median survival of 2.8 years. Within 30 days postoperative, 16 patients (20.1%) had grade 3–5 complications and one death (1.3%) occurred. Factors associated with grade 3+ complication included ECOG PS 2–3 (OR 18.00, p = 0.04), neoadjuvant chemotherapy and/or immunotherapy (OR 6.98, p < 0.01), and pelvic sidewall involvement (OR 8.80, p = 0.04). Factors associated with death within 3 years of surgery included ECOG PS 2–3 (HR 8.97, p < 0.01), neoadjuvant chemotherapy and/or immunotherapy (HR 2.34, p = 0.03), whether exenteration was performed (HR 2.64, p = 0.01), and positive resection margin (HR 3.37, p < 0.01). Conclusions: In well-selected patients, IORT is a feasible and safe option for the treatment of recurrent/persistent gynecologic malignancy with an appreciable survival benefit.
Rotational atherectomy versus orbital atherectomy for the treatment of calcified coronary lesions: a systematic review and meta-analysis of 81,873 patients
Amin, A. M., Tanashat, Mohammad , Sharaf, I, Altobaishat, Obieda , Abuelazm, Mohamed , Amin, A M, Turkmani, Mustafa , Manasrah, AlMothana , Abouzid, Mohamed

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2365 Get It

Abstract Background Previous studies have suggested that pre-treatment of coronary artery calcification with rotational atherectomy (RA) prior to stent deployment can enhance procedural success, but this does not necessarily translate into improved clinical outcomes. Recently, the Orbital Atherectomy (OA) system has been introduced into clinical practice to enhance lesion preparation. Purpose Our study aims to evaluate the safety and efficacy of both RA and OA as pre-treatment strategies before stent deployment in patients with coronary artery calcification. Methods We conducted a comprehensive systematic review and meta-analysis of randomized controlled trials (RCTs) and observational studies. These were identified through systematic PubMed, Web of Science, Scopus, Embase, and Cochrane searches until January 2024. Dichotomous data were pooled using risk ratio (RR), and continuous data were pooled using mean difference (MD), both with a 95% confidence interval (CI), using Stata version 17. Results Our analysis included eleven studies encompassing 81,873 patients. No significant difference was observed between RA and OA in terms of major adverse cardiovascular events (MACE) (RR: 0.81 with 95% CI [0.64, 1.02], P= 0.07). However, OA was significantly associated with a decreased incidence of all-cause mortality (RR: 0.60 with 95% CI [0.52, 0.70], P&lt; 0.01), target vessel revascularization (TVR) (RR: 0.43 with 95% CI [0.22, 0.82], P= 0.01), coronary artery bypass grafting (CABG) (RR: 0.78 with 95% CI [0.62, 0.97], P= 0.03), and fluoroscopy time (MD: -3.93 with 95% CI [-7.39, -0.48], P= 0.03). There was no significant difference between RA and OA in cardiac death (RR: 2.65 with 95% CI [0.94, 7.51], P= 0.07), target lesion revascularization (TLR) (RR: 0.57 with 95% CI [0.23, 1.41], P= 0.22), myocardial infarction (RR: 0.79 with 95% CI [0.62, 1.01], P= 0.06), stent thrombosis (RR: 0.33 with 95% CI [0.09, 1.23], P= 0.10), stroke (RR: 0.76 with 95% CI [0.57, 1.01], P= 0.06), re-admission (RR: 0.76 with 95% CI [0.48, 1.19], P= 0.23), vessel closure (RR: 0.78 with 95% CI [0.18, 3.33], P= 0.74), procedural duration (MD: -5.43 with 95% CI [-21.71, 10.84], P= 0.51), length of hospital stay (MD: 0.21 with 95% CI [-0.47, 0.90], P= 0.54), and contrast amount (MD: -3.68 with 95% CI [-9.37, 2.01], P= 0.21). Conclusion Both RA and OA as pre-treatment strategies before stent deployment demonstrate similar outcomes for numerous cardiovascular events. However, OA is associated with lower rates of all-cause mortality, TVR, CABG, and fluoroscopy time.
Clinical, echocardiographic, and pathologic characteristics of patients with pathology-proven nonbacterial thrombotic endocarditis
El‐Am, Edward , Ahmad, Ali , Abbasi, Mehdi , Akiki, Elias , Bois, Melanie C. , Maleszewski, Joseph J. , Kurmann, Reto , Klarich, Kyle W.

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.3040 Get It

Abstract Background Non-bacterial thrombotic endocarditis (NBTE) is a non-infectious condition characterized by thrombotic and/or inflammatory depositions involving the cardiac valves. It is a challenging diagnosis of exclusion. Purpose This study aimed to better characterize patients with pathologically proven NBTE and assess its clinical, echocardiographic characteristics and prognostic implications. Methods We retrospectively identified patients with pathology-proven NBTE at a single center. Patients with positive preoperative blood cultures, a history of active or treated infective endocarditis, or identification of valvular organisms by special stains were excluded from the study. Patients with pathology reports of valve perforation or chordal rupture were also excluded because of the high likelihood of an infectious etiology for these lesions. Results A total of 53 patients with NBTE (median age 57 years; 61% females) were identified. In this cohort, 22% had a history of malignancy, 30% had connective tissue disease, 32% had coronary artery disease, and 11% prior history of NBTE. Neurological events accounted for 18 (34%) of the presenting symptoms (TIA n=17, CVA n=9), while new valvular dysfunction in 7 (13%) and fever in 8 (15%) of patients. Five patients (9%) had NBTE involving more than one valve. NBTE involved the mitral valve in 64%, aortic valve in 38% prosthetic/mechanical valve in 11%, and tricuspid valve in 5%. NBTE was identified at the time of autopsy in 34% of patients. NBTE recurred in 3 patients during follow-up. Excluding patients diagnosed at the time of autopsy, 9% of patients died within 1 year of NBTE diagnosis. Conclusion NBTE was diagnosed pathologically in the presence of hyper-inflammatory states due to critical illness or immunological diseases. It was predominantly on the left side of the heart and was associated with embolic phenomena, either clinically or on imaging. Patients diagnosed with antemortem with NBTE have a high mortality in the first year (9%).
Cardiopulmonary exercise test findings are not the same in distinct types of cardiac amyloidosis
Ortega-Aviles, Laura M., Grogan, Martha , Allison, Thomas G. , Santos-Patarroyo, Sebastián , Ortega-Aviles, L M

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2964 Get It

Abstract Background Cardiopulmonary exercise testing (CPET) is a widely used tool for risk stratification and prognosis in patients with chronic heart failure. In amyloid patients, cardiac amyloid deposits are associated with poor prognosis, increased risk for arrhythmias, sudden death, and subsequent exercise limitation due to restrictive cardiomyopathy. The type of amyloidosis and its affectation of the heart are different, and little is known about the CPET parameters according to the type of cardiac amyloidosis. Purpose In this study, we aim to compare CPET data in different types of amyloidosis to determine if CPET variables evaluated could have a significant difference that could impact the follow-up or prognosis of the patients based on the type of amyloidosis. Methods We conducted a retrospective cohort study of 194 patients with a diagnosis of cardiac amyloidosis referred for CPET from February 2010 to February 2024. Patients with at least one CPET data were included in our analysis. The type of amyloidosis, comorbidities, and echocardiogram data were retrieved from medical records. Patients were divided into 3 different groups according to the type of amyloidosis: AL Amyloidosis(AL), Variant TTR Amyloidosis (vTRR), and Wild Type TTR Amyloidosis (wtTTR). Individuals with CPET studies after heart transplant were excluded from the analysis. Fisher’s exact test or chi-square was used to compare categorical variables. ANOVA and Tukey-Kramer (TK) were used to compare continuous variables. Adjustment using General Linear Models (GLM) procedure was used to identify any confounder that could affect the results. A p-value &lt;0.05 was considered significant. Results Out of the 194 patients, 66 had AL Amyloidosis, 40 had vTTR Amyloidosis, and 88 had wtTTR Amyloidosis. Most of the patients were male, and the male predominance was higher in the wtTRR group. There were 41 (62.1%) males in the AL group, 35 (87.5%) in the vTRR group, and 83 (94.3%) in the wtTTR group (p&lt;0.0001). Comorbidities were balanced among the groups. Additional demographic and comorbidities data are shown in Table 1. From the evaluated CPET variables, VE/VCO2 nadir and slope were significantly higher in the wtTTR group when compared to the other groups, as well as Peak VE and Peak TV(Figure A). However, after adjusting for sex, BMI, and age, VE/VCO2 nadir was still significantly higher in the wtTRR group (p=0.003) compared to the other groups, as well as Peak VE, which was higher again in vTTR patients (p=0.001), and Peak TV (Table 2). Conclusion Ventilation variables evaluated in CPET showed alterations in the wtTTR group. These findings suggest that wtTRR amyloid patients could have a severe problem in ventilation that could be of significant relevance in future prognosis.
Relationship between obesity and hypertrophic or dilated cardiomyopathy: the role of sex
Panagiotopoulos, Ιoannis , Lazaros, Georgios, Tsioufis, Konstantinos , Antonopoulos, Alexios S. , Terentes-Printzios, Dimitrios G. , Omer, Mohamed , Mentias, Amgad , Elgendy, Islam Y. , Vlachopoulos, Charalambos V.

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2888 Get It

Abstract Background Evidence suggests an association between obesity and the risk for cardiomyopathy development, however robust evidence is still lacking. In this study we sought to explore the relationship of obesity with hypertrophic (HCM) and dilated cardiomyopathy (DCM) and possible interactions with sex using large scale epidemiological real-world data. Methods and Results We analyzed data from the Nationwide Inpatient Sample (NIS) of U.S. hospitalizations for the years 2015-2019. There were a total of n=46,934 admissions with a diagnosis of HCM and n=170,924 with DCM. There was a significant interaction between cardiomyopathies’ diagnosis with sex and age subgroups; the rates of both DCM and HCM increased with age (p&lt;0.001 for both); DCM diagnosis was significantly higher in males compared with females (0.85% vs. 0.35%, p&lt;0.001). After adjustment for age, sex, race, and presence of arterial hypertension there was a significant stepwise positive association between obesity and the population rates of both cardiomyopathy subtypes. For hospitalized patients with a body mass index (BMI) ≥30kg/m2 there was an odds ratio (OR) of 1.68 (95%CI: 1.55-1.81, p&lt;0.001) for HCM andOR=1.82 (95%CI: 1.79-1.84, p&lt;0.001) for DCM. More importantly, the positive relationship between a cardiomyopathy diagnosis (HCM or DCM) with increasing BMI was driven by the male sex (p&lt;0.001 for both) and it was non-significant in females. Conclusions The findings from this nationwide observational analysis support a sexual dimorphism in the relationship between obesity and hypertrophic or dilated cardiomyopathy, which should be further investigated.Cardiomyopathy-obesity
Exercise stress echocardiography detects patients with paroxysmal/persistent atrial fibrillation
Vazquez, J P, Begidova, Irina , Padang, Ratnasari , Merli, Elisa , Kobal, Sergio , Picano, Eugenio , Zagatina, A , Ciampi, Quirino , Boshchenko, A , Rodríguez‐Zanella, Hugo , Ágoston, Gergely , Varga, Albert , Wierzbowska‐Drabik, Karina , Arbucci, Rosina , Pellikka, Patricia A.

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.302 Get It

Abstract Background Atrial fibrillation (AF) is the most prevalent atrial tachyarrhythmia, affecting an estimated 2-4% of adults worldwide. AF is a well-recognized and treatable risk factor for stroke, but it often remains asymptomatic or subclinical, leading to underdiagnosis. This study aimed to identify distinguishing signs during exercise stress echocardiography (ESE) among patients in sinus rhythm (SR), with and without history of paroxysmal/persistent AF (PAF). Methods An international prospective cohort comprising 1035 consecutive patients was initially considered from 12 cardiology institutions. These patients were referred for clinically-driven ESE as part of the Stress Echo 2020-2030 study network. All patients underwent comprehensive ESE with the ABCDE protocol, which included assessment of B-lines, left ventricle contractile reserve (LVCR), ECG-based heart rate reserve (HRR), LV ejection fraction (EF), LV end-diastolic volume, E/e’, pulmonary artery systolic pressure estimation, tricuspid annular plane sytolic excursion (TAPSE), left atrium volume index (LAVi) and left atrium reservoir phase strain (LASr) at rest and with exercise. Results After excluding 89 patients who were in AF during the ESE, 946 patients in SR were included as the main study cohort (101 with and 845 without a history of PAF). A scoring system was then constructed using all independent variables that differentiate the two groups (those with and without history of PAF). Presence of systemic hypertension, rest LAVI&gt;31ml/m2, rest |LASr|&lt; 21%, stress LAVI &gt;29 ml/m2, stress |LASr| &lt;24%, stress E/e’ &gt;7, ∆TAPSE &lt;4.5 mm, HRR&lt;1.87, and LVCR&lt;1.6 were identified as the discriminators between patients with and without PAF. The composite score, summing all 9 items (with a total score range from 0 to 9), yielded a score of &gt;4 as the best balance between sensitivity (87%) and specificity (60%) in predicting those with PAF. The receiver operating characteristic curve analysis of the scoring model showed an AUC of 0.79 (p-value &lt;0.0001), Figure. Conclusion ESE evaluation can be a valuable adjunct for identifying patients in SR with history of PAF. This multifaceted approach enhances the potential for more accurate detection and clinical management of individuals in SR at risk for AF, who could be considered for aggressive monitoring programs to prevent the development of new onset AF and subsequent increase in cardiovascular risk.
Association of weekday-to-weekend sleep duration pattern with myocardial infarction: analysis of the NHANES 2017- 2020
Kulthamrongsri, Kritpong, Nuchpramool, Prachawanee , Buahung, Ratchapol, Sodsri, Tulaton , Numwong, Terasut , Kulthamrongsri, Narathorn , Yinadsawaphan, Thanaboon , Wattanachayakul, Phuuwadith , Treewaree, Sukrit , Danpanichkul, Pojsakorn , Tantisattamo, Ekamol

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2666 Get It

Abstract Background Short sleep is linked to increased acute myocardial infarction (MI) risk.However, the association of weekend-to-weekday sleep patterns with MI in a nationally representative population remains unknown. The goal is to investigate how weekday and weekend sleep durations affect myocardial infarction (MI) in the general population, and if longer weekend sleep can offset insufficient weekday sleep's effects. Methods A cross-sectional study utilized data from the 2017-2020 NHANES to examine the relationship between sleep patterns and MI. Participants self-reported their average sleep durations on weekdays (1) and weekends (2), along with the calculated weekend-to-weekday sleep duration ratio (SDR) (3). Logistic regression models, adjusting for various factors, including demographics and comorbidities, were employed to analyze the association. Those with SDR &gt; 1 had more weekend sleep, while an SDR = 1 indicated equal sleep durations. Results The study involved 9,205 adult participants, with a mean ± SD age of 68.8 ± 10.8 years and 30.7% being female. Among them, 4.65% (428 patients) had a history of MI. The mean ± SD of weekday and weekend sleep hours was 7.6 ± 1.7 and 8.3 ± 1.8 hours, respectively. For weekday sleep duration, after univariate analysis, individuals compared to those in Q1 were statistically significant 0.72 times less likely in Q2 and 1.4 times more likely in Q4 to experience MI . While MI was 0.76 times less likely to occur among participants in Q3, it was not statistically significant (95% CI Q2: 0.54, 0.96; Q4: 1.05, 1.76; Q3:0.57, 1.00). After adjusting, participants only in Q3 were statistically significant 0.71 times less likely to develop MI.(Q3: 95% CI 0.51, 0.99). For weekend sleep duration, the univariate analysis showed that individuals in Q2, Q3, and Q4 were significantly less likely to experience MI (OR = 0.63, 0.67, and 0.69; 95% CIs: 0.48-0.82, 0.51-0.87, 0.53-0.90, respectively) After adjusting, only participants in Q2 were statistically significant less likely to develop MI (OR = 0.71; 95% CI 0.52 - 0.96). For the weekend-to-weekday sleep duration ratio (SDR), after univariate analysis, individuals were statistically significant 0.54 times less likely, and 1.5 times more likely to experience MI in those whose SDR is more than 1, and those whose SDR equals 1, respectively (95% CI SDR &gt; 1: 1.14, 1.98; SDR = 1: 0.39, 0.72). After adjusting, participants with SDR more than 1 were statistically significant 0.69 times less likely to develop MI. Conclusions Moderate sleep durations on weekdays (in Q3), weekends (in Q2), along with SDR more than 1, were negatively linked with MI in a general population. This implies that sleep durations on both weekends and weekdays have comparable associations with MI. Additionally, full recovery sleep on weekends following sleep-deprived weekdays, might reduce MI risk. Further studies are needed to explore impact of weekend and weekday sleep patterns on MI.Histogram and Margins plotTable of Odd ratios
The association between cardiovascular health and physiologic age as determined by artificial intelligence-enabled electrocardiography
Aviles, Laura Ortega, Medina-Inojosa, Betsy, Medina-Inojosa, J R, Medina‐Inojosa, Betsy J. , Medina-Inojosa, J R, Attia, Zachi I. , López-Jiménez, Francisco

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2691 Get It

Abstract Background The concept of ideal cardiovascular health (CVH) was developed by the American Heart Association (AHA) as a composite of seven health factors and behaviors labeled as the Life’s Simple 7 (LS7). Ideal CVH has been associated with complex measures of cardiovascular aging. We hypothesized that physiologic aging, as determined by a simple, routinely used tool such as the electrocardiogram (ECG) enabled by an artificial intelligence algorithm to predict an individual’s physiologic age (ECG age), will be associated with poor CVH as defined by an adapted version of the AHA LS7. Methods We conducted an observational retrospective study of consecutive patients aged ≥18 years who sought care between June 2019 and March 2022, self-reported behavioral history through a validated intake questionnaire in clinical outpatient visits, had a clinically indicated ECG within one year, and provided research authorization. We applied a previously developed AI-ECG convolutional neural network to derive the ECG-Age. We excluded patients with missing data, those with body mass index (BMI) &lt;18.5 kg/m2, previous history of left-bundle branch block, atherosclerotic cardiovascular disease events, atrial fibrillation, heart failure, and non-skin cancer. An adapted version of the LS7 was used to define CVH, which includes anthropometrics (e.g., BMI), behavioral history (e.g., smoking, physical activity, and diet), vital signs (e.g., blood pressure), and laboratory values (e.g., total cholesterol, and fasting glucose). CVH was stratified as ideal, intermediate, and poor if the LS7 score had values of 9-14, 7-8, and 0-6, respectively, constructed by adding 2,1 or 0 points to each metric that met each category criteria, respectively. Accelerated aging was defined as having a Delta-Age ≥1SD above the mean, whereas Delta-age was defined as ECG minus chronological age. Logistic regression models tested the association between CVH categories and positive Delta-Age while adjusting for chronological age and sex. Results We included 114,214 individuals (53.3% females, 86.9% Caucasian). Mean chronological age was 57.7±14 years and mean AI-ECG- Age was 58 ± 12.44 years (R2 0.80, p-value &lt;0.001). Mean Delta-Age was 0.28 ±8.8 years. LS7 average score was 8.3 ± 2.23; 46.8% had ideal, 31.5% intermediate, and 21.7% had poor CVH (Figure A). Compared to those with ideal CVH, those with poor CVH were more than 50% more likely to have Delta-Age ≥1SD (OR 1.61, CI 95% 1.54-1.69, p-value: &lt;0.001), independent of chronological age and sex. (Figure B). Conclusion Poor CVH, defined by the AHA LS7, is linked with biological aging as determined by AI-enabled ECG, highlighting the importance of promoting CVH in adults as a strategy to decelerate biological aging and reduce its associated mortality.
Racial, ethnic, and sex disparities in the utilization and outcomes of tricuspid valve surgery
Alkhouli, Mohammad, Ismayl, Mahmoud , Ahmed, H, Goldsweig, Andrew M. , Guerrero, Mayra

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2496 Get It

Abstract Background Data on racial/ethnic and sex disparities in the utilization and outcomes of tricuspid valve surgery (TVS) in the United States are scarce. Purpose We aimed to evaluate the impact of race/ethnicity and sex on the utilization and outcomes of TVS. Methods We analyzed the National Inpatient Sample database from 2016 to 2020 to identify hospitalizations for tricuspid valve surgery. Racial/ethnic and sex disparities in tricuspid valve surgery outcomes were determined using logistic regression models. Results Between 2016 and 2020, 19,395 hospitalizations for tricuspid valve surgery were identified. The utilization rates (defined as the number of tricuspid valve surgery performed per 100,000 US hospitalizations) were lower in Black and Hispanic patients compared with White patients for surgical tricuspid valve repair (331 versus 493 versus 634, p&lt;0.01) and surgical tricuspid valve replacement (312 versus 601 versus 728, p&lt;0.01). Similarly, the utilization rates were lower for women compared with men for surgical tricuspid valve repair (1,021 versus 1,364, p&lt;0.01) and surgical tricuspid valve replacement (930 versus 1,316, p&lt;0.01). Compared to White men undergoing tricuspid valve surgery, all women had lower odds of acute kidney injury (aOR 0.65, 95% CI 0.55-0.78) and higher odds of blood transfusion (aOR 1.30, 95% CI 1.07-1.59), and Black men had higher odds of blood transfusion (aOR 1.59, 95% CI 1.08-2.35). In-hospital mortality and other surgical complications were similar between all groups (all p&gt;0.05). Conclusions Significant racial/ethnic and sex disparities exist in the utilization of tricuspid valve surgery in the United States. Further studies are needed to understand the reasons for these disparities and to identify effective strategies for their mitigation.
Unlocking the future: machine learning models for predicting in-hospital mortality after transcatheter aortic valve replacement: a single-arm meta-analysis
Mangat, M M Manvir, Palicherla, Anirudh , Ellauzi, Rama , Kalathil, Ruth Ann Mathew, Pusapati, Suma , Dufani, Jalal, Aboeata, Ahmed, Anavekar, Nandan S., Ahmed, H, Ismayl, Mahmoud , Goldsweig, Andrew M.

European Heart Journal | October 28, 2024 | Volume: 45 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehae666.2455 Get It

Abstract Background Transcatheter aortic valve replacement (TAVR) continues to be increasingly performed. While machine learning has demonstrated utility in predicting adverse outcomes in various cardiovascular procedures, its effectiveness in predicting in-hospital mortality in TAVR remains uncertain. Purpose We aimed to examine the accuracy of machine learning in predicting in-hospital mortality in TAVR. Methods We searched PubMed and Cochrane databases up to February 2024 for observational studies assessing the diagnostic accuracy of machine learning for predicting in-hospital mortality in TAVR. Extracted data included study characteristics, patient demographics, type of predictive model used, and area under the receiver operating characteristic curves (AUC). Predictive accuracy was assessed by pooling AUC of machine-learning models across studies. AUC values ranging from 0.80 to 0.90, 0.70 to 0.80, 0.60 to 0.70, and 0.50 to 0.60 corresponded to excellent, good, fair, and poor discrimination ability, respectively (1). We used a random effects model to estimate the overall effect size with a 95% confidence interval (CI). Heterogeneity was evaluated with the I2 test.We conducted subgroup analyses based on the type of machine-learning model utilized to assess for any potential differences in predictive accuracy. Results A total of three studies were identified, comprising a study population of 70,714 patients. Across studies, eight machine learning models were utilized, consisting of three artificial neural network models, one logistic regression model, one naive Bayes model, two random forest models, and one support vector machine model. Pooled AUC showed excellent accuracy of machine learning in predicting in-hospital mortality in TAVR (0.92; 95% CI 0.90 - 0.95). Subgroup analyses revealed similar yet excellent accuracy across all types of machine learning models utilized: artificial neural network (0.92; 95% CI 0.86-0.99), logistic regression (0.92; 95% CI 0.89-0.95), naive Bayes (0.90; 95% CI 0.88-0.92), random forest (0.94; 95% CI 0.87-1.00), and support vector machine (0.94; 95% CI 0.91-0.96). Conclusion Our study suggests that utilizing machine learning can accurately predict in-hospital mortality in TAVR. Operators of TAVR procedures may benefit from incorporating machine learning prediction models in their clinical practice. Further exploration of machine learning on outcomes in TAVR is needed to verify our findings.