Electrocardiographic abnormality classification in peripartum cardiomyopathy and its prognostic relevance regarding left ventricular recovery

T. Gausepohl (Hannover)1, T. Pfeffer (Hannover)1, T. Reinke (Hannover)2, J. Müller-Leisse (Hannover)2, H. A. K. Hillmann (Hannover)2, D. Berliner (Hannover)2, D. Hilfiker-Kleiner (Hannover)3, M. Ricke-Hoch (Hannover)1, J. Bauersachs (Hannover)2, D. Duncker (Hannover)1
1Medizinische Hochschule Hannover Klinik für Kardiologie und Angiologie Hannover, Deutschland; 2Medizinische Hochschule Hannover Kardiologie und Angiologie Hannover, Deutschland; 3Medizinische Hochschule Hannover Präsidium Hannover, Deutschland
Background: Peripartum cardiomyopathy (PPCM) is a rare but life-threatening cause of heart failure and a significant contributor to maternal mortality and morbidity. Although many patients show recovery of left ventricular (LV) function, a substantial proportion experiences persistent cardiac dysfunction. As part of the initial diagnostic workup, the 12-lead electrocardiogram (ECG) may serve as a widely available and cost-effective tool for early risk stratification. This study aimed to evaluate the prevalence of ECG abnormalities in a German PPCM cohort and assess the prognostic value of two criteria for ECG classification: the criteria proposed by the Investigations of Pregnancy Associated Cardiomyopathy (IPAC) study group and the Minnesota ECG classification (MECGA) with regard to LV recovery at 6 months.
 
Methods and results: We retrospectively analyzed data from 71 PPCM patients treated at Hannover Medical School with baseline ECGs available. Follow-up ECGs at 1 and 6 months after diagnosis were available in 56 and 57 patients respectively. ECGs were classified as (ab)normal according to IPAC and MECGA criteria, and correlated with LV recovery status at 6 months (defined as full recovery: LVEF ≥50%; partial recovery: LVEF 36–49%; no recovery: LVEF ≤35% or adverse outcome such as LVAD, transplantation or death). Logistic regression was used to evaluate the predictive value of ECG findings. 
Mean baseline LVEF was 29% ± 11 and improved to 48% ± 10 after 6 months. At diagnosis, 28% of ECGs were classified as abnormal according to the IPAC criteria and 49% when the MECGA criteria were applied. The most frequent abnormalities were T-wave changes (59%), left bundle branch block (14%), and signs of LV hypertrophy.
An abnormal ECG at baseline significantly predicted incomplete recovery at 6 months or at which other time point?: IPAC (Odd’s ratio [OR] 4.07, interquartile range [IQR] 1.07-20.18), MECGA (OR 4.34, IQR 1.39-14.79), and combined IPAC+MECGA (OR 6.00, IQR 1.87-21.28). In the ECG at 6 months, only MECGA and combined criteria remained significantly associated with incomplete recovery (MECGA: OR 8.50, IQR 2.17-43.72; IPAC+MECGA: OR 7.11, IQR 1.94-31.22). All patients with no recovery met abnormal MECGA criteria at both timepoints, while some patients displayed a normal ECG according to IPAC criteria despite incomplete LV recovery at 6 months.
 
In conclusion, ECG abnormalities are highly prevalent among PPCM patients and provide valuable prognostic information. The MECGA criteria, particularly due to inclusion of T-wave and hypertrophy markers, outperformed IPAC criteria in predicting LV non-recovery. Given its accessibility and low cost, ECG remains an important tool in PPCM management, especially in low-resource settings.