Abstract
Funding Acknowledgements
Type of funding sources: None.
Introduction
Current guidelines indicate that pacing methods that maintain physiologic ventricular activation (biventricular pacing or His-bundle pacing) should be chosen over right ventricular pacing (Vp) among patients with EF 36-50% who are expected to Vp >40% of the time. There are no guidelines to help predict which patients will receive a high burden of Vp and this is left to operator
opinion. We sought to ascertain whether operator opinion is an accurate predictor of high burden of Vp.
Methods
This was a single-centre single-blinded observational study of patients who received pacemaker implant for treatment of bradycardia between April 2015 and 2019 and had at least 12 month follow-up data on record. Patients’ demographic, clinical, electrocardiographic and echocardiographic data were reviewed in a blinded fashion by a senior implanting physician, who estimated whether the Vp at 12 months would exceed 40%. The Vp at approximately 12 months was obtained from the pacing records and compared with the prediction.
Results
Some 982 patients underwent pacemaker implantation during the study period, 698 for conduction system disease (CSD), 267 for sinus node disease (SND) and 17 for other conditions. Overall, 856 had valid follow-up data. Of these, 543 (63.4%) were predicted to Vp >40% , and 527(61.6%) were documented as having Vp >40%. The sensitivity and specificity of operator prediction were 93.2% and 84.2%, with positive and negative predictive values of 90.4% and 88.5%. Table 1 illustrates analyses of different populations by clinical parameter. In sub-group analysis, complete heart block and PR > 300ms were significant factors for accurate prediction of Vp > 40%, however clinical features, such as syncope, were poor discriminators.
Conclusion
In this single-centre study, among patients receiving pacemaker implant for treatment of bradycardia, operator prediction of the burden of Vp >40% has an acceptable degree of accuracy. Sub-group analysis suggests that certain clinical parameters could make this prediction easier. Table 1. Comparison of operator opinion SND CSD CHB SND+PR < 160 PR > 300 Syncope Non-syncope n 698 267 216 84 60 409 344 Sensitivity 44.4% 97.7% 100% 6.3% 100% 86.4% 87.9% Specificity 98.3% 62.0% 45.2% 97.1% 0%* 89.9% 79.6% PPV 87.0% 90.6% 91.6% 33.3% 98.3% 94.2% 92.2% NPV 87.9% 87.9% 100% 81.5% - 77.6% 70.5% * only 1 patient did not RV pace >40% - this was not predicted. SND – sinus node disease; CSD – conduction system disease; CHB – complete heart block, PPV – positive predictive value; NPV – negative predictive value