Utility of electrophysiological studies to predict arrhythmic events

Table 2 Diagnostic accuracy of the electrophysiological study and the left ventricular ejection fraction for the primary and secondary endpoint

Predictor variable	Sensitivity	Specificity	PPV	NPV	AUROC (95%CI)
Primary endpoint
Sustained VT during EPS
All study participants (n = 265)	69.4%	57.9%	27.2%	89.3%	0.636 (0.563-0.709)
Subgroup of study participants with LVEF < 35% (n = 106)	66.7%	48.1%	30.5%	80.9%	0.574 (0.468-0.680)
Subgroup of study participants with LVEF ≥ 35% (n = 159)	72.7%	63.5%	24.2%	93.5%	0.681 (0.578-0.785)
LVEF < 35%	55.1%	63.4%	25.5%	86.1%	0.593 (0.515-0.670)
Secondary endpoint
Sustained VT during EPS
All study participants (n = 265)	66.7%	55.6%	17.6%	92.1%	0.611 (0.524-0.699)
Subgroup of study participants with LVEF < 35% (n = 106)	61.1%	45.5%	18.6%	85.1%	0.533 (0.406-0.660)
Subgroup of study participants with LVEF ≥ 35% (n = 159)	73.3%	61.8%	16.7%	95.7%	0.676 (0.553-0.798)
LVEF < 35%	54.5%	62.1%	17.0%	90.6%	0.583 (0.491-0.675)

AUROC: Area under the receiver operating characteristic curve; EPS: Electrophysiological study; NPV: Negative predictive value; PPV: Positive predictive value; VT: Ventricular tachycardia; LVEF: Left ventricular ejection fraction.