Electrocardiographic Predictors of Heart Failure With Reduced Versus Preserved Ejection Fraction: The Multi‐Ethnic Study of Atherosclerosis

O’Neal, Wesley T.; Mazur, Matylda; Bertoni, Alain G.; Bluemke, David A.; Al‐Mallah, Mouaz H.; Lima, João A.C.; Kitzman, Dalane W.; Soliman, Elsayed Z.

doi:10.1161/jaha.117.006023

Cited by 59 publications

(58 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Other investigators have evaluated the diagnostic implications of individual ECG characteristics for identifying low EF. In general, these studies have demonstrated hazard ratios (HR) in the range of 1.5 to 2.5 for low EF for each individual feature (with the exception of left bundle branch block (LBBB) which is associated with a HR of 3.93) . However, when applied individually, these features are of limited utility and, perhaps most importantly, they need to be assessed manually.…”

Section: Discussionmentioning

confidence: 99%

“…14 associated with a HR of 3.93). 16 However, when applied individually, these features are of limited utility and, perhaps most importantly, they need to be assessed manually. The current study leverages state of the art computing technology, a very large data set, nonlinear models, and features derived using convolution layers which allow the network to "see" features that a human cannot.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Prospective validation of a deep learning electrocardiogram algorithm for the detection of left ventricular systolic dysfunction

Attia

Kapa

Yao

et al. 2019

Cardiovasc electrophysiol

124

View full text Add to dashboard Cite

Objectives We sought to validate a deep learning algorithm designed to predict an ejection fraction (EF) less than or equal to 35% based on the 12‐lead electrocardiogram (ECG) in a large prospective cohort. Background Patients undergoing routine ECG may have undetected left ventricular (LV) dysfunction that warrants further echocardiographic assessment. However, identification of these patients can be challenging. Methods We applied the algorithm to all ECGs interpreted by the Mayo Clinic ECG laboratory in September 2018. The performance of the algorithm was tested among patients with recent echocardiographic assessments of LV function. We also applied the algorithm in patients with no recent echocardiographic assessments of LV function to determine the rate of new “positive screens.” Results Among 16 056 adult patients who underwent routine ECG, 8600 (age 67.1 ± 15.2 years, 45.6% male), had a transthoracic echocardiogram (TTE) and 3874 patients had a TTE and ECG less than 1 month apart. Among these patients, the algorithm was able to detect an EF less than or equal to 35% with 86.8% specificity, 82.5% sensitivity, and 86.5% accuracy, (area under the curve, 0.918). Among 474 “false‐positives screens,” 189 (39.8%) had an EF of 36% to 50%. Among patients with no prior TTE, the algorithm identified 3.5% of the patients with suspected EF less than or equal to 35%. Exploratory analysis suggests false positives could be reduced by assessing NT‐pro‐BNP after the initial “positive screen.” Conclusions A deep learning algorithm detected depressed LV function with good accuracy in routine practice. Further studies are needed to validate the algorithm in patients with no prior echocardiogram and to assess the impact on echocardiography utilization, cost, and clinical outcomes.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Prospective validation of a deep learning electrocardiogram algorithm for the detection of left ventricular systolic dysfunction

Attia

Kapa

Yao

et al. 2019

Cardiovasc electrophysiol

124

View full text Add to dashboard Cite

show abstract

“…There is limited literature which has investigated the clinical relevance of ECG changes in HFpEF 21,22 . The study carried out by Cenkerova et al demonstrated that HFpEF patients who died during the course of FU, had a tendency to have a prolonged QTc interval (427 ± 42 ms vs. 454 ± 42 ms, P = 0.058) 21 .…”

Section: Discussionmentioning

confidence: 99%

Clinical predictors of all‐cause mortality in patients presenting to specialist heart failure clinic with raised NT‐proBNP and no heart failure

et al. 2020

View full text Add to dashboard Cite

Aims Clinical outcomes for patients suspected of having heart failure (HF) who do not meet the diagnostic criteria of any type of HF by echocardiography remain unknown. The aim of this study was to investigate the clinical predictors of all-cause mortality in patients with suspected HF, a raised N-terminal pro-b-type natriuretic peptide (NTproBNP) and who do not meet the diagnostic criteria of any type of HF by echocardiography. Methods and results Relevant data were taken from the Sheffield HEArt Failure (SHEAF) registry (222349P4). The inclusion criteria were presence of symptoms raising suspicion of HF, NTproBNP > 400 pg/mL, and preserved left ventricular function. Exclusion criteria were any type of HF by echocardiography. The outcome was defined as all-cause mortality. Cox proportional-hazards regression model was used to investigate the association between the survival time of patients and clinical variables; 1031 patients were identified with NTproBNP > 400 pg/mL but who did not have echocardiographic evidence of HF. All-cause mortality was 21.5% (222 deaths) over the mean follow-up (FU) period of 6 ± 2 years. NTproBNP was similar in patients who were alive or dead (P = 0.96). However, age (HR 1, P < 0.01), chronic kidney disease (CKD, HR 1.2, P < 0.01), chronic pulmonary obstructive disease (COPD, HR 1.6, P < 0.01), dementia (HR 5.9, P < 0.01), male gender (HR 1.4, P < 0.01), first-degree atrioventricular block (HR 2.1, P < 0.01), left axis deviation (HR 1.6, P = 0.04), and diabetes (HR 1.4, P = 0.03) were associated with all-cause mortality. In multivariate regression, age, gender, CKD stage, COPD, and dementia were independently associated with mortality. In patients with NTproBNP > 627 pg/mL, NYHA class predicted death (II, 19.6%; III, 27.4%; IV, 66.7%; P < 0.01). Conclusions Patients with no HF on echocardiography but raised NTproBNP suffer excess mortality particularly in the presence of certain clinical variables. Age, male gender, worsening CKD stage, presence of COPD, and dementia are independently associated with all-cause mortality in these patients. An NTproBNP > 627 pg/mL coupled with NYHA class could identify patients at greatest risk of death.

show abstract

“…In addition, different cutoffs for left ventricular ejection fraction (LVEF) were used to define HeFNEF: ≥40% (Cenkerova, Dubrava, Pokorna, Kaluzay, & Jurkovicova, 2016;Danciu et al, 2006;Hendry, Krisdinarti, & Erika, 2016), >40% (Hawkins et al, 2007;Olsson et al, 2006), ≥45% (Adabag et al, 2014;Donal et al, 2014;Joseph et al, 2016;Komajda et al, 2011;Nikolaidou et al, 2017;Shah et al, 2013), >45% (Ho et al, 2013;Lee et al, 2009;Park et al, 2013;Zile et al, 2011), ≥50% (Gigliotti et al, 2017;Gijsberts et al, 2016;Hummel, Skorcz, & Koelling, 2009;Khan et al, 2007;Lund et al, 2013;Martinez Santos et al, 2016;Masoudi et al, 2003;Menet et al, 2014;O'Neal et al, 2017;Pascual-Figal et al, 2017;Peyster, Norman, & Domanski, 2004;Senni et al, 1998;Shenkman et al, 2002;Yap et al, 2015), >50% (Eicher et al, 2012;Oskouie, Prenner, Shah, & Sauer, 2017;Sanchis et al, 2015;Selvaraj et al, 2014;Shah et al, 2015), and ≥55% (Varadarajan & Pai, 2003). The following methods were used to measure ejection fraction: echocardiography, nuclear scintigraphy, and contrast ventriculography.…”

Section: Studiesmentioning

confidence: 99%

Electrocardiographic characteristics in patients with heart failure and normal ejection fraction: A systematic review and meta‐analysis

Nikolaidou

Samuel

Marincowitz

et al. 2019

Noninvasive Electrocardiol

View full text Add to dashboard Cite

Background Little is known about ECG abnormalities in patients with heart failure and normal ejection fraction (HeFNEF) and how they relate to different etiologies or outcomes. Methods and Results We searched the literature for peer‐reviewed studies describing ECG abnormalities in HeFNEF other than heart rhythm alone. Thirty five studies were identified and 32,006 participants. ECG abnormalities reported in patients with HeFNEF include atrial fibrillation (prevalence 12%–46%), long PR interval (11%–20%), left ventricular hypertrophy (LVH, 10%–30%), pathological Q waves (11%–18%), RBBB (6%–16%), LBBB (0%–8%), and long JTc (3%–4%). Atrial fibrillation is more common in patients with HeFNEF compared to those with heart failure and reduced ejection fraction (HeFREF). In contrast, long PR interval, LVH, Q waves, LBBB, and long JTc are more common in patients with HeFREF. A pooled effect estimate analysis showed that QRS duration ≥120 ms, although uncommon (13%–19%), is associated with worse outcomes in patients with HeFNEF. Conclusions There is high variability in the prevalence of ECG abnormalities in patients with HeFNEF. Atrial fibrillation is more common in patients with HeFNEF compared to those with HeFREF. QRS duration ≥120 ms is associated with worse outcomes in patients with HeFNEF. Further studies are needed to address whether ECG abnormalities correlate with different phenotypes in HeFNEF.

show abstract

Electrocardiographic Predictors of Heart Failure With Reduced Versus Preserved Ejection Fraction: The Multi‐Ethnic Study of Atherosclerosis

Cited by 59 publications

References 38 publications

Prospective validation of a deep learning electrocardiogram algorithm for the detection of left ventricular systolic dysfunction

Prospective validation of a deep learning electrocardiogram algorithm for the detection of left ventricular systolic dysfunction

Clinical predictors of all‐cause mortality in patients presenting to specialist heart failure clinic with raised NT‐proBNP and no heart failure

Electrocardiographic characteristics in patients with heart failure and normal ejection fraction: A systematic review and meta‐analysis

Contact Info

Product

Resources

About