Use of epicardial electrocardiograms for detecting cardiac allograft rejection

Non-invasive detection of cardiac allograft rejection among heart transplant recipients using an electrocardiogram based deep learning model

Aims

Current non-invasive screening methods for cardiac allograft rejection have shown limited discrimination and are yet to be broadly integrated into heart transplant care. Given electrocardiogram (ECG) changes have been reported with severe cardiac allograft rejection, this study aimed to develop a deep-learning model, a form of artificial intelligence, to detect allograft rejection using the 12-lead ECG (AI-ECG).

Methods and results

Heart transplant recipients were identified across three Mayo Clinic sites between 1998 and 2021. Twelve-lead digital ECG data and endomyocardial biopsy results were extracted from medical records. Allograft rejection was defined as moderate or severe acute cellular rejection (ACR) based on International Society for Heart and Lung Transplantation guidelines. The extracted data (7590 unique ECG-biopsy pairs, belonging to 1427 patients) was partitioned into training (80%), validation (10%), and test sets (10%) such that each patient was included in only one partition. Model performance metrics were based on the test set (n = 140 patients; 758 ECG-biopsy pairs). The AI-ECG detected ACR with an area under the receiver operating curve (AUC) of 0.84 [95% confidence interval (CI): 0.78-0.90] and 95% (19/20; 95% CI: 75-100%) sensitivity. A prospective proof-of-concept screening study (n = 56; 97 ECG-biopsy pairs) showed the AI-ECG detected ACR with AUC = 0.78 (95% CI: 0.61-0.96) and 100% (2/2; 95% CI: 16-100%) sensitivity.

Conclusion

An AI-ECG model is effective for detection of moderate-to-severe ACR in heart transplant recipients. Our findings could improve transplant care by providing a rapid, non-invasive, and potentially remote screening option for cardiac allograft function.

Novel implantable device to detect cardiac allograft rejection

BACKGROUND

Allograft rejection remains the nemesis of solid organ transplantation. Soul Mate is a novel implantable wireless data transmission system that analyzes 9 intramyocardial electrogram parameters recorded from 4 or 6 configurations of 2 or 3 epicardial leads to detect allograft rejection. This study determined the ability of the Soul Mate to detect early rejection of transplanted hearts.

METHODS AND RESULTS

Five dogs underwent heterotopic cervical heart transplantation and simultaneous implantation of the Soul Mate's Cardiac Rejection Monitoring Device. Dogs were initially immunosuppressed, but subsequent drug discontinuation allowed allograft rejection to appear. Allograft biopsies were performed at regular intervals to determine rejection grade, which was compared to a calculated rejection score determined as percent change from baseline of values for each intramyocardial electrogram. There was significant correlation between the biopsy results and the evolution of 5 parameters. The strongest correlation (r=0.939; P<0.001) was obtained using the "general median" parameter from 4 configurations, assessed 1 day before the biopsy, with a sensitivity of 85.7% and a specificity of 100% compared to the myocardial biopsy results.

CONCLUSIONS

The Soul Mate allograft rejection monitoring system accurately detected transplanted heart rejection in a canine model noninvasively with continuous sampling. This proof-of-concept study suggests that the Soul Mate could be used to more intensely and more frequently monitor cardiac allografts for rejection.

Noninvasive diagnosis of cardiac allograft rejection in an orthotopic canine model

BACKGROUND

A decline in the R wave voltage obtained from surface electrocardiograms once correlated with cardiac allograft rejection. With cyclosporine therapy, however, these electrocardiographic findings became inconsistent, occurring only during severe rejection episodes. Despite cyclosporine use, intramyocardial unipolar peak-to-peak amplitudes obtained from plunge electrodes are reported to be highly sensitive and specific for diagnosing rejection. These reports are based on heterotopic grafts, which atrophy over time, making long-term voltage changes during rejection difficult to interpret. The purpose of this study was to use analysis of unipolar peak-to-peak amplitudes as a prospective monitoring tool for diagnosing orthotopic cardiac allograft rejection.

METHODS

Ten adult mongrel dogs underwent orthotopic heart transplantation with the attachment of four intramyocardial leads. The unipolar peak-to-peak amplitudes were measured daily and compared with endomyocardial biopsy results.

RESULTS

We found that intramyocardial unipolar peak-to-peak amplitude analysis had a sensitivity and a specificity of 100% for diagnosing rejection. We also found that as the number of myocardial leads increased, the sensitivity of detecting rejection also increased.

CONCLUSIONS

We conclude that unipolar peak-to-peak amplitude analysis is an accurate noninvasive means for early detection of cardiac allograft rejection in an orthotopic model. Its success should allow less frequent, more selective use of endomyocardial biopsy.

Evoked epimyocardial electrogram for rejection diagnosis after heart transplantation

An endomyocardial electrogram (ECG) was reported to be a sensitive and practicable method for rejection monitoring after heart transplantation. Long-term follow up was limited, however, by variations of signals. The repolarization part of ECG signals vary with changes of heart rate. Both can be avoided by using pacemaker-induced signals. For stimulation and sensing of the ventricular-evoked response, a new type of electrode with fractal surface structure was used. Twenty patients undergoing heart transplantation were evaluated. Amplitudes of the depolarization and repolarization part of the ventricular-evoked response signals were analyzed and related to the degree of acute rejection according to histological findings from endomyocardial biopsy. Signals were transferred by Internet and analyzed automatically. In the case of focal moderate rejection (grade 2, International Society for Heart Transplantation grading) and higher degrees of rejection, a significant amplitude decrease was found. This sensitive non-invasive method for rejection monitoring with a high level of reliability provides the possibility of reducing the number of endomyocardial biopsies.

Noninvasive diagnosis of cardiac allograft rejection: the effect of procainamide

The surface electrocardiogram (ECG) has been used as a noninvasive technique for the diagnosis of cardiac allograft rejection. Alteration in conduction, R-wave amplitude, and rhythm have been associated with rejection. These ECG findings are modulated by the myocyte sodium channel, but are inconsistent and occur only during severe rejection episodes. The purpose of this study was to (1) characterize changes in cardiac electrophysiology during allograft rejection using the highly sensitive intramyocardial electrocardiogram and (2) determine whether pharmacological sodium channel blockade with procainamide enhances subtle ECG changes. Nine mongrel dogs underwent heterotopic heart transplantation in which four intramyocardial leads (one anteriorly and posteriorly on each ventricle) were attached. Leads exited to a subcutaneously placed ECG block which was transcutaneously accessed posttransplant to record direct intramyocardial electrocardiograms. Six animals were treated with procainamide, while three were not and served as controls. Daily measurements included the QRS, QT, and QTc intervals and the R-wave amplitude. Endomyocardial biopsies were performed weekly and also when significant decline in ECG amplitude occurred. Detailed ECG interval analysis failed to establish any correlation between conduction and rejection, even in the procainamide-treated group. Intramyocardial amplitude analysis, however, had a sensitivity of 100% and a specificity of 86% for the diagnosis of rejection. The results indicate that intramyocardial ECG interval analysis is not predictive of rejection even when prolonging conduction with procainamide. Amplitude analysis, however, remains an accurate noninvasive means for the early detection of cardiac allograft rejection and should allow more selective use of endomyocardial biopsy.

Bipolar intramyocardial electrogram from an implanted telemetric pacemaker for the diagnosis of cardiac allograft rejection

Noninvasive detection of acute cardiac allograft rejection remains a challenge. Analysis of the epicardial electrogram transmitted by unipolar telemetric pacemaker can be of help in the detection of rejection with myocytolysis but is hampered by extracardiac factors. Instead, the contribution of a bipolar pacemaker for this purpose was studied. The bipolar peak-to-peak amplitude of epicardial electrograms (BPPA) from 25 patients implanted with a bipolar pacemaker at the time of heart transplantation was measured at the time of endomyocardial biopsies. BPPA was expressed as a percent of the baseline value. A voltage drop of more than 10% was considered an indication of rejection. Of 118 biopsies, 80 were free of rejection and 38 showed mild to moderate rejection (Grade 1A = 12; Grade 1B = 17; Grade 2 or 3 = 9). The mean value of BPPA was less for grade 2 biopsies (86.4 +/- 17%) than for biopsies with no or mild rejection (101.3 +/- 14.3% for Grade 0, 101.4 +/- 13.8% for Grade 1A, and 98.6 +/- 18% in Grade 1; P < 0.05). Diagnostic concordance between BPPA measurement and biopsy results increased with the histological severity of rejection (Grade 1A = 1/12, Grade 1B = 5/17, Grade > 2 = 7/9). Acute rejection was diagnosed with a sensitivity of 34% for Grade 1A, 46% for Grade 1B, and 78% for rejection episodes with myocytolysis (Grade > or = 2). Specificity remained approximately 90% for all histologic grades.(ABSTRACT TRUNCATED AT 250 WORDS)