Results of Octaplex for reversal of warfarin anticoagulation in patients with hip fracture

Results

We identified 33 patients who met our inclusion criteria (Fig. 1). Their demographic and comorbidity data are shown in Table 1. Fracture type and treatment are shown in Table 2. All patients were taking warfarin preoperatively: 26 (82%) for atrial fibrillation and 6 (18%) for previous DVT and/or PE. It is important to note that many of these patients had multiple pre-existing comorbidities, particularly coronary artery disease (55%), congestive heart failure (CHF; 45%) and chronic obstructive pulmonary disease (39%). The mean Charlson Comorbidity Index (CCI) score was 2.82.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1

Selection of patients included in the study.

The median time from Octaplex administration to a measured INR of 1.4 or lower was just over 1 hour. This is indicative of the time required for Octaplex infusion and the process of obtaining a repeated coagulation study — theoretically, the onset of action of Octaplex may be as fast as 10 minutes.15 Table 3 and Table 4 show the results of Octaplex on the INR and time to surgery for our population. A single dose of Octaplex was effective in correcting the INR to 1.4 or lower in 29 patients (88%), with an average dose of 1470 units of Octaplex. Of the 4 remaining patients, 2 required a second dose of Octaplex; 1 underwent surgery with an INR of 1.5 and 1 passed away before surgery. All patients received oral or intravenous vitamin K before surgery, with a mean dose of 19.5 mg (range 5–55 mg) given. Two patients had Octaplex administered after they were brought to the operating room to minimize delays to surgery; therefore, zero hours passed from Octaplex administration to the start of surgery. The median delay from time of admission to surgery was 22 hours.

Spinal anesthesia was used in 20 patients (63%), without any cases of postoperative epidural hematoma. Intraoperative estimated blood loss (EBL) was less than 300 mL in nearly all patients, except for an EBL of 400 mL for a patient undergoing hemiarthroplasty and an EBL of 800 mL in a patient undergoing cephalomedullary fixation of a subtrochanteric fracture that required open reduction. The overall mean EBL was 244 ± 150.5 mL. Transfusion of packed red blood cells was required intraoperatively or postoperatively in 11 patients (33%), with a range of 0–3 units of blood administered. The median LOS was 11 (range 2–62) days before discharge home or to a rehabilitation facility.

Five out of 33 patients (15.2%) died within 30 days of Octaplex administration or hip fracture surgery, with an overall 30-day survival of 84.8% ± 6.2%. Two patients died before their hip fracture surgery after receiving Octaplex. The first developed electrocardiogram changes consistent with an inferior MI immediately after administration of Octaplex and died shortly thereafter. The second sustained an intraoperative cardiac arrest and died during administration of a spinal anesthetic 6 hours after she received Octaplex. Three patients died within 30 days after hip fracture surgery. Two deaths occurred at 3 and 21 days from a presumed fatal MI or arrhythmia. The final death occurred after the patient experienced a respiratory arrest at their care home 28 days postoperatively. Of these 5 patients who died within 30 days, 4 received both Octaplex and FFP preoperatively.

Ten patients (31%) preoperatively received FFP in addition to Octaplex, and the survival of this subgroup was significantly worse. Seven patients received Octaplex after FFP because of insufficient INR reversal. One patient received Octaplex and FFP simultaneously, and 2 patients received FFP after Octaplex despite a post-Octaplex INR of 1.4 or lower, all for undocumented reasons. In patients who received Octaplex but no FFP, 30-day survival was 95.7% ± 4.3%. However, in patients who received both Octaplex and FFP, 30-day survival was 60.0% ± 15.5%. This difference was statistically significant using a log-rank comparison (p = 0.002). The survival curves are shown in Figure 2. However, these 2 groups are dissimilar in age and comorbidities, as patients who received both Octaplex and FFP were older (mean age 84.1 v. 79.4, p = 0.09) and had more comorbidities (mean CCI score 3.90 v. 2.35, p = 0.049) than those who received only Octaplex.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2

Kaplan–Meier survival of subgroups who received Octaplex only or Octaplex + fresh frozen plasma (FFP).

Aside from the deaths described previously, no other cardiac events were noted postoperatively. No patients experienced CHF between the time of Octaplex administration and their hip fracture surgery. Five patients experienced CHF postoperatively within 30 days of Octaplex administration. Three of these patients were treated with furosemide during their postoperative hospital stay, and 2 required readmission for management of CHF at 20 and 29 days postoperatively. The only venous thromboembolic complication identified was a single case of DVT diagnosed at 7 days postoperatively. No strokes, PEs, or arterial thrombi were identified within 30 days after Octaplex administration.

Three patients experienced orthopedic complications within 30 days of surgery. Two patients had persistent drainage from their surgical wounds after resumption of anticoagulation therapy and were treated with negative pressure wound therapy. This was successful in 1 patient, but the other was readmitted to hospital 22 days postoperatively for a deep infection that required surgical irrigation and débridement followed by home intravenous antibiotic therapy. The organisms identified were Propionium acnes and Staphylococcus aureus (coagulase negative). The third orthopedic complication involved a patient who had a ground-level fall 29 days after discharge and sustained a periprosthetic fracture around the tip of his long cephalomedullary nail. His fracture was stabilized with a distal femoral locking plate after his anticoagulation was reversed without the use of Octaplex.