Bridging for an isolated subtherapeutic INR: an evaluation of clinical practice patterns, outcomes, and costs from an anticoagulation clinic.

No formal recommendations support bridging patients taking warfarin for a subtherapeutic international normalized ratio (INR). This study aimed to: (1) characterize practices at one anticoagulation clinic, (2) evaluate adverse events, and (3) compare cost of bridging versus withholding bridging for subtherapeutic INR. A retrospective chart review of 320 patients having 546 isolated subtherapeutic INR episodes included patients with an INR below their therapeutic range, preceded by two INRs within or above range. Bridged episodes required more frequent follow-up visits to achieve therapeutic INR (2.5 ± 1.0 vs. 2.2 ± 0.6; P = 0.097), but fewer days until the INR returned to therapeutic range (6.8 ± 5.0 vs. 18.9 ± 16.0; P < 0.0001). The strongest predictor of bridging was the magnitude the INR fell below the therapeutic range, where those with a severely-low INR were 30-fold more likely to be bridged (P < 0.0001), and moderately-low INR episodes were 6-fold more likely to be bridged compared with mildly-low INR (P < 0.0001). Those at high thromboembolic risk were more likely to be bridged than at low-risk (OR 3.39 [1.50-7.68]; P = 0.0034). Increasing age reduced the likelihood of being bridged (OR 0.97 [0.95-0.99]; P = 0.0118). Adverse events were infrequent in both the bridged and non-bridged; thrombosis (2.0 vs. 0.7%), major bleeding (2.0 vs. 1.3%), minor bleeding (4.1 vs. 3.1%) and bruising (18.4 vs. 3.6%). Incremental cost difference of bridging was significantly greater for total cost ($967.13) and its components, direct medical ($951.32), transportation ($2.73) and productivity cost ($13.08). It is unclear if bridging for an isolated subtherapeutic INR reduces thrombosis risk, but it is associated with higher costs.