Predictors of preprocedural direct oral anticoagulant levels in patients having an elective surgery or procedure

Abstract

The Perioperative Anticoagulation Use for Surgery Evaluation (PAUSE) study prospectively evaluated a prespecified periprocedural-interruption strategy of direct oral anticoagulants (DOACs) among patients with atrial fibrillation. Logistic regression analyses were performed to identify clinical parameters associated with residual DOAC levels ≥30 ng/mL or ≥50 ng/mL. Patients undergoing low-bleed-risk procedures were more likely to have residual levels of ≥30 ng/mL and ≥50 ng/mL. For low-risk procedures, age ≥75 years, female sex, a creatinine clearance (CrCl) <50 mL/min, and an interruption of <36 hours were associated with a greater likelihood of levels ≥30 ng/mL, whereas age ≥75 years, female sex, a CrCl of <50 mL/min, and standard DOAC dosing were associated with levels ≥50 ng/mL. For high-risk procedures, weight of <70 kg, CrCl <50 mL/min, and standard DOAC dosing were associated with residual levels ≥30 ng/mL, whereas female sex was associated with levels ≥50 ng/mL. For low-risk procedures, apixaban was associated with a higher likelihood of levels ≥30 ng/mL as compared with dabigatran (P = .0019) and of levels ≥50 ng/mL when compared with rivaroxaban (P = .0003). For high-risk procedures, apixaban was marginally associated with a higher likelihood of residual levels ≥30 ng/mL when compared with dabigatran (P = .05), whereas rivaroxaban was associated with a higher likelihood of levels ≥30 ng/mL as compared with apixaban. Further study is required to determine whether adjustments to perioperative plans based on these clinical parameters could result in a lower risk of residual DOAC levels. The PAUSE trial was registered at www.clinicaltrials.gov as #NCT2228798.

Trial registration: ClinicalTrials.gov NCT02228798.

Graphical abstract

Figure 1.

Residual DOAC levels according to preprocedural DOAC-interruption interval in hours. Patients undergoing low- and high-bleeding-risk procedures had their DOAC stopped 1 and 2 days prior to the procedure, respectively. Patients undergoing low-bleeding-risk procedures are indicated in black, whereas patients undergoing high-bleeding-risk procedures are indicated in blue. (A) The median preoperative-interruption intervals for low- and high-bleeding-risk procedures for apixaban were 39.4 hours (IQR, 37.5, 41.5) and 63.8 hours (IQR, 61.0, 67.0), respectively. (B) The median preoperative-interruption intervals for dabigatran for low- and high-bleeding-risk procedures were 40.5 (IQR, 38.4, 44.7) and 63.8 (IQR, 61.7, 74.3), respectively. (C) The median preoperative-interruption intervals for rivaroxaban for low- and high-bleeding-risk procedures were 48.0 (IQR, 40.8, 51.0) and 72.0 (IQR, 66.1, 75.0), respectively.

Figure 2.

Residual DOAC levels according to interruption interval. The proportion of patients with residual preprocedural levels ≥30 ng/mL and ≥50 ng/mL declines with increasing interruption duration. (A) Apixaban. (B) Dabigatran. (C) Rivaroxaban.  Apixaban was associated with a higher likelihood of levels ≥30 ng/mL compared with dabigatran and of levels ≥50 ng/mL when compared with rivaroxaban for low-risk procedures with shorter interruption intervals. Rivaroxaban was associated with a higher likelihood of levels ≥30 ng/mL as compared with apixaban for high-risk procedures with longer interruption intervals.