Normothermia vs Hypothermia in On-Pump Coronary Artery Bypass Grafting: A Randomised Clinical Trial

Technical Summary: Normothermia vs. Hypothermia in On-Pump Coronary Artery Bypass Grafting

Problem Statement
Coronary artery bypass grafting (CABG) utilizing cardiopulmonary bypass (CPB) traditionally employs hypothermia to protect organs by reducing myocardial and cerebral metabolic demands. However, hypothermia is known to adversely affect platelet function and coagulation, potentially increasing perioperative bleeding. While normothermic or near-normothermic CPB has gained adoption due to shorter bypass times and improved surgical techniques, contemporary randomized evidence comparing these strategies specifically for isolated on-pump CABG remains limited. Previous meta-analyses and randomized trials are often decades old, reflecting outdated surgical and perioperative practices. The study addresses the uncertainty regarding whether normothermic CPB offers a hemostatic advantage over mild hypothermic CPB in reducing clinically relevant postoperative bleeding without compromising safety.

Methodology
The study was a multicentre, prospective, single-blind, randomized clinical trial conducted across eight tertiary hospitals in China between September 2023 and January 2025.

• Participants: 330 adult patients undergoing isolated on-pump CABG were included in the full analysis set (FAS). Exclusion criteria included emergency surgery, combined cardiac procedures, preoperative hemoglobin <9 g/dL, reoperation, and recent antiplatelet therapy.
• Intervention: Patients were randomized 1:1 to two groups:
  • Normothermic CPB: Target oxygenator arterial outlet blood temperature of 35–36°C.
  • Mild Hypothermic CPB: Target oxygenator arterial outlet blood temperature of 32–33°C.
  • Cooling commenced immediately after CPB initiation, and rewarming began during the construction of the left anterior descending artery graft. Core temperature was monitored via nasopharyngeal and bladder/rectal probes, with the oxygenator outlet temperature serving as the primary management reference.
• Outcomes:
  • Primary Outcome: Moderate-to-massive postoperative bleeding, defined as Universal Definition of Perioperative Bleeding (UDPB) grades 2–4. This included criteria such as delayed sternal closure, chest tube output >800 mL in 12 hours, significant blood product transfusion, or surgical re-exploration for bleeding.
  • Secondary Outcomes: Distribution of UDPB grades (0–4), blood product transfusion, chest tube drainage, ICU/hospital stay, and major in-hospital adverse events (e.g., stroke, AKI, mortality).
• Statistical Analysis: The primary analysis followed the FAS using a generalized linear model to calculate the absolute risk difference (ARD) with 95% confidence intervals (CIs), adjusted for study center. The study was initially powered to detect a 15% absolute reduction in bleeding events, assuming a baseline rate of 40%.

Key Results

• Primary Outcome: The incidence of UDPB grades 2–4 was 11.6% (19/164) in the normothermic group and 12.7% (21/166) in the mild hypothermic group. The unadjusted absolute risk difference was −1.1% (95% CI −8.1% to 6.0%), and the adjusted ARD was −0.9% (95% CI −7.6% to 5.8%). There was no statistically significant difference between the groups (P = 0.767 unadjusted; P = 0.791 adjusted).
• Secondary Outcomes: No clinically meaningful differences were observed in secondary outcomes, including the full distribution of UDPB grades, transfusion requirements (RBCs, FFPs, platelets), chest tube output, mechanical ventilation time, or rates of major adverse events (e.g., atrial fibrillation, acute kidney injury, mortality).
• Hospital Stay: The normothermic group had a slightly longer median hospital stay (13 days) compared to the mild hypothermic group (12 days), with a mean difference of 1.551 days (95% CI 0.1 to 3.1; P = 0.043). No deaths occurred in either group.
• Subgroup and Sensitivity Analyses: Results remained consistent across various subgroups (e.g., age, sex, EuroSCORE I, CPB duration) and in per-protocol and single-center sensitivity analyses.

Significance and Claims
The authors conclude that in adults undergoing isolated on-pump CABG, normothermic CPB guided by oxygenator outlet temperature did not demonstrate superiority over mild hypothermic CPB in reducing clinically relevant postoperative bleeding.

The paper highlights several critical points regarding its findings:

1. Contemporary Context: The study provides randomized evidence relevant to current clinical conditions, contrasting with older trials that may not reflect modern surgical and blood management practices.
2. Low Event Rate: The observed incidence of significant bleeding (approx. 12%) was substantially lower than the anticipated 40% used for sample size calculation. The authors attribute this to advances in contemporary perioperative care, including structured blood management, restrictive transfusion strategies, and refined surgical techniques. This lower-than-expected event rate reduced the statistical power to detect modest differences between the groups.
3. Safety Profile: The trial found no safety signal associated with normothermic CPB under contemporary conditions, with similar rates of major complications and mortality between groups.
4. Clinical Implication: The authors suggest that in modern practice, the selection of CPB temperature strategy may be individualized based on the broader clinical context rather than being driven solely by concerns regarding postoperative bleeding.

The study acknowledges limitations, including the single-blind nature (surgeons and perfusionists were unblinded), the lower-than-anticipated event rate limiting precision, and the predominantly male, Asian study population, which may affect generalizability.