Predictive factors of recovery in patients admitted with seemingly reversible causes of symptomatic atrioventricular block

Abstract Introduction Atrioventricular block (AVB) can be associated with secondary causes, some of which are potentially reversible. Permanent Pacemaker (PPM) implantation aims at ameliorating symptoms an preventing sudden cardiac death but is not recommended in transient causes that can be corrected and prevented. Purpose Our aim was to assess predictors of recovery in patients admitted for symptomatic AVB with potential reversible causes. Methods We performed a retrospective analysis between February 2011 and November 2022 of all patients admitted with symptomatic second (including 2:1 and high-grade), third degree and atrial fibrillation with AVB and a potential reversible cause in a single expert centre. AVB secondary to Acute Coronary Syndromes and patients who required PPM to tolerate bradycardia inducing drugs were excluded from analysis. Medical records were analysed for demographics, clinical data and outcomes. Results 135 patients were analysed. The mean age at implantation was 78±9 years with a male preponderance (57%). The rhythm of presentation was complete AVB for most of the cohort (85,2%). A prior ECG was obtained for most patients (70.4%), with atrioventricular (AV) conduction abnormalities identified in 22% of patients and intraventricular (IV) conduction abnormalities identified in 51% of patients. Bradycardic drug therapy was the most identified reversible cause (88,1%). Significant hyperkalaemia (categorized as above 5,5 mmol/L) was identified in 23% of patients. In 24 patients (17,8%) there was recovery of rhythm with correction of the reversible cause and a PPM was not implanted. The independent predictors of recovery of rhythm were: a prior of chronic kidney disease (CKD) in dialysis (OR 7,8; CI 95% 1,2-49,5 p<0,05), moderate to severe hyperkalaemia (OR 7; CI 1,7 – 28,6 95%, p<0,05), high dose bradycardic drug therapy (OR 3,9; CI 95% 1,5-10,4, p<0,05), dual therapy with bradycardic drugs, (OR 15,7; CI 95% 1,6-158,4, p<0,05), presence of both bradycardic drugs and hyperkalaemia (OR 4,7; CI 95% 1,5-14,2, p<0,05) and a presentation rhythm different from complete AVB (OR 0,2; CI 95% 0,1- 0,7 , p<0,05). The age, sex, and prior AV or IV conduction abnormality were non-significant. Through a method of logistic regression, a proportionate score was developed composed of moderate to severe hyperkalaemia (3,5 points), high dose bradycardic drug therapy (1 point) and presentation rhythm different from complete AVB (1 point) with a high predictive power for the occurrence of recovery (OR 2,5; CI 95% 1,6-3,9, p<0,05; r2 0,24) (figure 1) and a very good discriminative capacity, with the ROC curve analysis (figure 2) demonstrating an AUC of 0,80. Conclusions In patients with symptomatic AVB and a potentially reversible cause, several factors were identified as predictive of recovery. A simple 3-point score was developed with a high predictive power and a very good discriminative capacity for recovery.