Device-dependent variability of activated clotting time in left-sided electrophysiology procedures
V Spahiu, I Zeljkovic, H Nilius, S Duetschler, T Kueffer, B Gerguri, V Gjergjizi, C Herrera-Siklody, N Kozhuharov, H Servatius, H Tanner, L Roten, M Nagler, T Reichlin, B KovacsAbstract
Background
Current guidelines recommend maintaining an activated clotting time (ACT) > 300 s during left-sided electrophysiology (EP) procedures to minimize thromboembolic risk. However, it remains largely unknown whether different routinely used ACT devices and cartridge types measure the same biological signal.
Purpose
To quantify inter-device and intra-device (cartridge) variability of ACT measurements during left-sided EP procedures under "real-world" anticoagulation conditions.
Methods
We prospectively compared ACT values obtained simultaneously from the same whole blood samples during left-sided EP procedures using three different commercially available ACT devices (Optical, OP; Electromechanical, EM; Electrochemical, EC) and cartridge (high- and low range, HR; LR) types. We included 185 consecutive patients undergoing left atrial (n=146) or left ventricular (n=23) ablations. All patients received a heparin bolus (150-200 IU/kg) to maintain an ACT of 300-350 s as measured by the reference ACT device (EM, HR cartridge). ACT was measured 5 minutes after heparin administration or every 30 minutes. For statistical analysis, Pearson’s correlation coefficients were calculated among the three ACT devies and for HR and LR separately. In addition, Passing–Bablok regression (PB) models were fitted to evaluate the presence of systematic bias between assays.
Results
Correlations between the test systems were moderate, ranging from 0.52 (EM vs. OP) to 0.58 (EM vs. EC) with the HR cartridges (Figure 1, panels B and D). Similar results were observed for the LR cartridges, with correlations of 0.55 (EM vs. OP) and 0.49 (EM vs. EC)(Figure 2, panels B and D). PB analysis demonstrated statistically significant bias in both intercept and slope across all comparisons (Figures 1 and 2, panels B and D). The absolute difference between measured ACT ranged from –115 to 130 s with the HR cartridges (Figure 1 panels A and C) and from –40 to 630 s with the LR cartridges (Figure 2 panels A and C). In particular, comparisons involving the EM device + LR cartridges showed pronounced systematic bias as demonstrated on Figure 2, panels B and D.
Conclusion
ACT results differ substantially between devices and cartridge types, indicating that different devices do not measure the same thing. Applying uniform guideline thresholds (e.g., ≥300 s) without considering device-specific variability may therefore lead to unintended under- or over-anticoagulation during left-sided EP procedures. Defining device-specific therapeutic ACT ranges—or applying validated correction factors—could improve consistency and patient safety.