The utility of routine clinical 12-lead ECG in assessing eligibility for subcutaneous implantable cardioverter defibrillator

Thomas, J. A. C.; Alday, Erick Andres Perez; Hamilton, Christopher; Kabir, Muammar; Park, Eugene A.; Tereshchenko, Larisa G.

doi:10.1016/j.compbiomed.2018.05.002

Cited by 11 publications

(19 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Out-of-sample validation of the lasso and elastic net prediction models in the previous non-ACHD study population 12 yielded high sensitivity of the pre-selected in this study threshold (≥ - 0.5): 100% for the elastic net model, and 77% for lasso model. Validation ROC AUC in a non-ACHD population was unsatisfactory for all models: specifically for lasso (ROC AUC 0.554; 95%CI 0.355-0.754), elastic net (ROC AUC 0.548; 95%CI 0.340-0.756), and ridge model (ROC AUC 0.477; 95%CI 0.282-0.671).…”

Section: Resultsmentioning

confidence: 73%

“…The results of this study underscore the need to improve S-ICD technology further to increase the number of eligible ACHD patients. Our previous study 12 showed remarkable (3-fold) improvement in S-ICD eligibility after ECG filtering. In this study, high QRS and T voltage was the main reason for S-ICD ineligibility and turning S-ICD ECG filtering feature ON can increase the number of eligible ACHD patients.…”

Section: Discussionmentioning

confidence: 86%

“…For out-of-sample validation of the machine learning models, we used the data of our previous S-ICD eligibility study, 12 which enrolled a widely generalizable sample of the OHSU outpatient population, with a broad range of age (18–81 y), body mass index (BMI; 19–53 kg/m 2 ), QRS duration (66–150 ms), and left ventricular ejection fraction (LVEF; 37–77%).…”

Section: Methodsmentioning

confidence: 99%

“…We applied a machine learning technique to develop a prediction of left-sided and right-sided S-ICD eligibility. Supine ECG datasets served for machine learning (training and testing), whereas standing ECG datasets, and the data of our previous S-ICD eligibility study 12 served for validation. We compared logistic regression, lasso, elastic net, and ridge models in four machine learning steps (Figure 2).…”

Section: Methodsmentioning

confidence: 99%

“…We conducted this study with several goals: (1) assess left-and right-sided S-ICD eligibility in ACHD patients, (2) validate the previous 12 left-sided S-ICD eligibility prediction tool, (3) use machine learning to predict right-and left-sided S-ICD eligibility in ACHD patients, and (4) develop and validate transformation matrices to transform 12-lead ECG to S-ICD 3-lead ECG, and vice versa.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Eligibility for Subcutaneous Implantable Cardioverter Defibrillator in Congenital Heart Disease

Wang

Javadekar

Rajagopalan

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Background: The goals of this study were: assess left-and right-sided subcutaneous implantable cardioverter-defibrillator (SICD) eligibility in adult congenital heart disease (ACHD) patients, use machine learning to predict SICD eligibility in ACHD patients, and transform 12-lead ECG to S-ICD 3-lead ECG, and vice versa. Methods: ACHD outpatients (n=101; age 42±14 y; 52% female; 85% white; left ventricular ejection fraction (LVEF) 56±9%) were enrolled in a prospective study. Supine and standing 12-lead ECG was recorded simultaneously with a right- and left-sided S-ICD 3-lead ECG. Peak-to-peak QRS and T amplitudes, RR, PR, QT, QTc, QRS intervals, Tmax, and R/Tmax (31 predictor variables) were tested. Model selection, training, and testing were performed using supine ECG datasets. Validation was performed using standing ECG datasets and out-of-sample non-ACHD population (n=68; age 54±16 y; 54% female; 94% white; LVEF 61±8%). Results: A 40% of participants were ineligible for S-ICD. Tetralogy of Fallot patients passed right-sided screening (57%) more often than left-sided (21%; McNemar's χ2 P=0.025). The ridge model demonstrated the best cross-validation function. Validation of the ridge models was satisfactory for standing left-sided [ROC AUC 0.687 (95%CI 0.582-0.791)] and right-sided [ROC AUC 0.655(95%CI 0.549-0.762)] SICD eligibility prediction. Out-of-sample validation in the non-ACHD population yielded a 100% sensitivity of the pre-selected threshold for the elastic net model. Validation of the transformation matrices showed satisfactory agreement (<0.1 mV difference). Conclusion: Nearly half of the contemporary ACHD population is ineligible for S-ICD. Machine-learning prediction of SICD eligibility can be used for screening of SICD candidates. Clinical Trial Registration: URL: www.clinicaltrials.gov Unique identifier: NCT03209726

show abstract

Section: Resultsmentioning

confidence: 73%

Section: Discussionmentioning

confidence: 86%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Eligibility for Subcutaneous Implantable Cardioverter Defibrillator in Congenital Heart Disease

Wang

Javadekar

Rajagopalan

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

To the editor: S‐ICD eligibility in adult congenital heart disease

2021

Self Cite

View full text Add to dashboard Cite

We read with concern the recent article by Zormpas et al., 'Eligibility for subcutaneous implantable cardioverterdefibrillator in adults with congenital heart disease'. 1 The authors reported a high (83%) rate of eligibility for subcutaneous implantable cardioverter-defibrillator (S-ICD) in the adult congenital heart disease (ACHD) population, which is in sharp contrast to several previous ACHD studies: by Wang et al. 2 (60% eligibility), by Alonso et al. 3 in mixed ACHD (75% eligibility) and tetralogy of Fallot (77% eligibility), 4 and by Garside et al. 5 (75% eligibility; left-sided only).One discrepancy is between the reported data in figures and what is discussed in the results section. Figure 1 in the manuscript 1 shows that 22% of patients failed the left-sided automated screening test and 25% of patients failed the right-sided automated test. The authors stated that only 'two patients were found eligible in the left parasternal but not in the right parasternal position, while all patients found eligible in the right parasternal position were also eligible in the left parasternal position'. Mathematically, it is not possible to achieve the reported 83% rate of eligibility per written findings, but, instead, figure 1 provides results more consistent with previous studies. [2][3][4] The authors 1 did not discuss possible reasons for disagreement of their findings with previous studies. It is essential to discuss the limitations of the study in order for the reader to gain perspective of the study.The methods included the use of an automated Boston-Scientific screening test. However, details about the Boston-Scientific programmer settings of the test and whether or not SMART Pass S-ICD function (ECG filtering settings) 6 was considered were not included. Furthermore, it is unusual for the automated screening test to yield 'no results.' Further details of the automated test with 'no results' would be helpful for troubleshooting and meaningful interpretation. It would be necessary if the authors compared their results with a previous comprehensive evaluation of the automatic screening tool. 7 An important limitation of the study was that a single investigator performed the ECG-based screening test, and therefore, inter-rater and intra-rater agreement has not been investigated. There was also no mention of investigator

show abstract

Author reply: S‐ICD eligibilities in adults with congenital heart disease

et al. 2021

View full text Add to dashboard Cite

We thank Wang et al. for their interest in our article on S-ICD eligibility in adults with congenital heart disease (ACHD). 1 Wang et al. emphasize that our results are in contrast to several previous ACHD studies, [2][3][4][5] including one recent study from their group. 2 Carefully reviewing the references, our results regarding S-ICD eligibility in ACHD patients (83%) are obviously in line with previous published data. In the studies mentioned by Wang et al., S-ICD eligibility was 75%, 3 77%, 4 and 75.4%. 5 Further studies not mentioned by Wang et al.showed even higher eligibility rates of 93.5% 6 and 86.7%. 7 More precisely, the result of Wang et al. 2 reporting only 60% S-ICD eligibility is exceptionally lower compared to the above-mentioned studies. Our eligibility rate lies even closer to the aforementioned mean of the published data than the eligibility rate from Wang et al. 2 (Figure 1). The references added by Wang et al. therefore nicely extend the well-described underlying evidence in the field.Wang et al. highlight a discrepancy between the reported data in the figures and the results section in a previous version of our manuscript, which has already been corrected

show abstract

The utility of routine clinical 12-lead ECG in assessing eligibility for subcutaneous implantable cardioverter defibrillator

Abstract: Routine clinical 12-lead ECG can be used to predict S-ICD eligibility. ECG filtering may improve S-ICD eligibility.

Cited by 11 publications

References 20 publications

Eligibility for Subcutaneous Implantable Cardioverter Defibrillator in Congenital Heart Disease

Eligibility for Subcutaneous Implantable Cardioverter Defibrillator in Congenital Heart Disease

To the editor: S‐ICD eligibility in adult congenital heart disease

Author reply: S‐ICD eligibilities in adults with congenital heart disease

Contact Info

Product

Resources

About