Importance: The conventional method of heparin and protamine management during cardiopulmonary bypass (CPB) is based on total body weight (TBW) which fails to account for the heterogenous response to unfractionated heparin (UFH) in each individual patient. Although inaccurate heparin dosing and reversal may lead to adverse postoperative outcomes, the literature is inconclusive on whether individualized anticoagulation management based on real-time blood heparin concentration (HC) improves post-CBP outcomes.

Objective: To conduct a meta-analysis comparing post-CPB outcomes in patients receiving conventional versus individualized anticoagulation management.

Data Sources: This systematically searched the databases: Medline, EMBASE, PubMed, CINHL and Google Scholar for full text papers published up until March 2023.

Study Selection: We recruited randomized controlled trials (RCTs) and prospective studies comparing the outcomes of dosing heparin and/or protamine for adults undergoing CPB based on measured HC versus TBW. Papers were excluded if they involved pediatric patients and patients of extracorporeal membrane oxygenation (EMCO) or studied non-CPB cardiovascular interventions.

Data Extraction and Synthesis: The literature search was conducted following the PRISMA guidelines. Data was extracted independently by multiple observers. Random effects meta-analyses were conducted to compare the outcome profiles of these two interventions. 

Main Outcomes and Measures: Primary endpoints include postoperative blood loss, transfusion requirements, heparin rebound, and reoperation for bleeding. The hypothesis and endpoints were formulated prior to data collection.

Results: We enrolled 25 articles including 25 RCTs and 4 prospective cohort studies involving 3810 patients.  Compared to TBW based heparin dosing, patients in HC group had significantly higher total heparin dose (MD=65.61mg, 95% CI:32-99.21mg), significantly lower protamine to heparin ratio (MD = 0.29, 95% CI: 0.18-0.41), significant higher early postoperative platelet counts (MD = 8.83, 95% CI: 2.07-15.69), and significantly reduced total postoperative blood loss (MD = 51.13 ml, 95% CI: 8.79-93.49 ml).  The random-effects meta-regression indicated that the changes in total protamine reversal dose (PD p= 0.0467) and total heparin dose(THD) or combined with length of aortic cross clamping time (p= 0.04) were associated with increasing postoperative blood loss.  There were no significant differences in CPB duration, aortic cross clamping time, postoperative red blood cell transfusion, incidences of heparin rebound, or chest wound re-exploration for bleeding.

Conclusions and Relevance: Compared to TBW based heparin dosing, individualized HC based management may beneficially preserve platelet count and function while reducing postoperative blood loss. The significant correlation between PD and THD with post operative blood loss suggests that precise dosing of protamine and heparin may reduce postoperative bleeding. Dosing protamine based on residual heparin may be mandatory.  Accurate prediction of dosing of heparin may require new technology, e.g., machine learning which has more power to process mega volume of clinical data, result in best algorithm for estimating heparin dosage.

Key point

Question: Does individualized heparin management reduce post operative blood loss in patients undergoing cardiac surgery requiring cardiopulmonary bypass?

Findings: This systematic review and meta-analysis compared strategies of conventional heparin dosing based on total body weight versus individualized heparin management based on in vitro heparin dose response calculations. The individualized Strategy resulted in significantly reduced post-operative blood loss, as well as reduced protamine/heparin dose ratio, and higher platelet counts.  There is a confounding finding where the total heparin dose was found to be significantly higher in the intervention group. The meta-regression showed that total heparin dose combined with aortic cross clamping time was predictive of post operative blood loss, the likely explanation would be that other than iatrogenic overdosing, the intervention group could have heparin overdose from inaccurate heparin dose response curve.  Bleeding could be further reduced if less heparin were given. In addition, the assumption of one-compartment pharmacokinetic model of heparin metabolism and the linear relationship between heparin dose and ACT/or heparin concentration may be inaccurate.

Meaning: During cardiopulmonary bypass, heparin anticoagulation management should be individualized based on accurate prediction of protamine and heparin dose response. The cumulative clinical data of heparin dose response may require machine learning which offers greater data processing power to generate accurate mathematical equations taking consideration of patient-subjective variables in heparin administration.