Activated 4-Factor Prothrombin Complex Concentrate as a Hemostatic Adjunct for Neonatal Cardiac Surgery: A Propensity Score-Matched Cohort Study

Abstract

BACKGROUND: Prothrombin complex concentrates are an emerging "off-label" therapy to augment hemostasis after cardiopulmonary bypass (CPB), but data supporting their use for neonatal cardiac surgery are limited. METHODS: We retrospectively reviewed neonates undergoing open heart surgery with firsttime sternotomy between May 2014 and December 2018 from a hospital electronic health record database. Neonates who received activated 4-factor prothrombin complex concentrate (a4FPCC) after CPB were propensity score matched (PSM) to neonates who did not receive a4FPCC (control group). The primary efficacy outcome was total volume (mL/kg) of blood products transfused after CPB, including the first 24 hours on the cardiovascular intensive care unit (CVICU). The primary safety outcome was the incidence of 7-and 30-day postoperative thromboembolism. Secondary outcomes included 24 hours postoperative chest tube output, time to extubation, duration of CVICU stay, duration of hospital stay, 30-day mortality, and incidence of acute kidney injury on postoperative day 3. We used linear regression modeling on PSM data for the primary efficacy outcome. For the primary safety outcome, we tested for differences using McNemar test on PSM data. For secondary outcomes, we used linear regression, Fisher exact test, or survival analyses as appropriate, with false discovery rate-adjusted P values. RESULTS: A total of 165 neonates were included in the final data analysis: 86 in the control group and 79 in the a4FPCC group. After PSM, there were 43 patients in the control group and 43 in the a4FPCC group. We found a statistically significant difference in mean total blood products transfused for the a4FPCC group (47.5 mL/kg) compared with the control group (63.7 mL/kg) for PSM patients (adjusted difference, 15.3; 95% CI, 29.4-1.3; P = .032). We did not find a statistically significant difference in 7-or 30-day thromboembolic rate, postoperative chest tube output, time to extubation, incidence of postoperative acute kidney injury (AKI), or 30-day mortality between the groups. The a4FPCC group had a significantly longer length of intensive care unit stay (32.9 vs 13.3 days; adjusted P = .049) and hospital stay (44.6 vs 24.1 days; adjusted P = .049) compared with the control group. CONCLUSIONS: We found that the use of a4FPCC as a hemostatic adjunct for post-CPB bleeding in neonatal cardiac surgery was associated with a decrease in mean total blood products transfused after CPB without an increased rate of 7-or 30-day postoperative thromboembolism. Our findings suggest that a4FPCCs can be considered as part of a hemostasis pathway for refractory bleeding in neonatal cardiac surgery.

Comparison of Prothrombin Complex Concentrate with Activated Factor VII Use for Bleeding Following Cardiopulmonary Bypass in Children

“…16,17 Another recent study by Navaratnam et al found that the use of four-factor PCC as a hemostatic adjunct for post-CPB bleeding in neonatal cardiac surgery was associated with a decrease in mean total blood products transfused after CPB without an increased rate of 7-or 30-day postoperative thromboembolism. 18 A study in adult patients, which compared outcomes following rFVIIa administration with PCC administration after CPB, showed that PCC led to significantly less chest tube output, fresh frozen plasma transfusion requirements, and platelet transfusion requirements compared with rFVIIa. 19 The purpose of the current study was to compare the efficacy and safety profiles of rFVIIa with PCC for control of bleeding in pediatric patients undergoing CPB for congenital heart disease.…”

FEIBA Use in Neonatal Cardiac Surgery: A Risky Business That Needs Further Investigation

“…In this issue of Anesthesia & Analgesia , Navaratnam et al 4 from Stanford University report their experience with the use of factor VIII inhibitor-bypassing agent (FEIBA) in neonates undergoing cardiac surgery with CPB. The authors performed a retrospective review of 165 neonates who underwent cardiac surgery between May 2014 and December 2018.…”

In Response

“…Because the PSM reduced our initial sample size by 50%, we also used inverse probability of treatment weighting (IPTW) on the primary efficacy outcome as a sensitivity analysis so that we could use more of the data and determine whether the differences in blood product usage remained in the weighted total cohort. In Table 3, 2 we showed unadjusted means and standardized differences along with IPTW estimates of treatment differences for the entire cohort of 165.…”

“…We are not in favor of post hoc power calculations as requested by Kahntroff et al 1,3,4 However, we agree that explicit estimates of the required size for potential randomized trials would be a useful addition to help guide future prospective trial design. As one possible approach, we calculated the following study size estimates using summary statistics from the matched cohort for the 2 primary outcomes: total blood products (a4FPCC mean 47.5 and SD 26.4, control mean 63.7 and SD 41.7, from Table 2 2 ) and postoperative thromboembolism rates (7 days: a4FPCC 2/43 and control 3/43; 30 days: a4FPCC 7/43 and control 4/43, from Table 4 2 ). For each, we estimated N (where N = total number of subjects required) for 80% and 90% power and for level 0.05 and level 0.025 (assuming alpha-splitting for 2 primary outcomes).…”