Combined Antegrade–Retrograde Cardioplegia for Myocardial Protection During Coronary Artery Bypass Grafting in Patients With Left Main Coronary Artery Disease: The CARDIOPROTECT-LMCAD Randomised Trial

Illustration of a perfusionist managing blood cardioplegia delivery equipment during coronary bypass surgery.

In a randomized trial of 60 adults undergoing CABG for left main coronary artery disease, combined antegrade–retrograde blood cardioplegia did not significantly reduce peak troponin I compared with antegrade-only delivery. Both strategies had similar early clinical outcomes, and no cardioplegia-related complications occurred. The single-center trial supports individualized myocardial protection rather than routine combined delivery, but it cannot establish equivalence or exclude modest benefits.

CARDIOPROTECT-LMCAD examined whether adding retrograde delivery to antegrade blood cardioplegia improves myocardial protection during coronary artery bypass grafting (CABG) in patients with significant left main coronary artery disease. Lesanovic and colleagues randomized 60 adults to the two delivery strategies and measured postoperative cardiac biomarkers and early clinical outcomes. The combined approach was technically successful, but it did not significantly reduce the trial’s primary endpoint: peak high-sensitivity cardiac troponin I during the first 48 hours after surgery. The result supports a selective approach while leaving uncertainty about smaller benefits and higher-risk patients.

The physiological question matters because severe proximal coronary obstruction can interfere with antegrade cardioplegia distribution. Antegrade delivery introduces cardioplegia through the aortic root and coronary arteries; retrograde delivery uses a coronary sinus catheter to reach myocardium through the venous circulation. Combining the routes may improve distribution when arterial delivery is compromised. However, a plausible mechanism does not establish clinical benefit. This trial specifically tested combined delivery in contemporary on-pump CABG practice rather than comparing retrograde cardioplegia alone with antegrade treatment.

The prospective, single-center study took place at the Dedinje Cardiovascular Institute in Belgrade, Serbia. Investigators screened 96 patients between April 2023 and October 2025 and randomized 60, with 30 assigned to each group. Eligible adults had at least 50% left main coronary artery stenosis and were scheduled for elective isolated on-pump CABG. Exclusions included previous cardiac surgery, concomitant procedures, emergency or off-pump CABG, dialysis-dependent kidney failure, and circumstances in which the surgeon considered randomization inappropriate. These criteria restrict applicability to several more complex or urgent surgical populations.

Allocation used a computer-generated sequence implemented through REDCap. Surgeons, anesthesiologists, and perfusionists knew the assigned strategy, but patients, postoperative clinical staff, laboratory personnel, and outcome assessors remained blinded. An independent statistician performed the analyses after completion of follow-up. All participants received their assigned treatment without crossover and completed 30-day follow-up. These features strengthen the comparison, although the study was retrospectively registered and its small sample allowed residual baseline and operative differences despite randomization.

Both groups received standardized surgery and goal-directed cardiopulmonary bypass management. Cardioplegia consisted of oxygenated blood mixed with crystalloid in a 4:1 ratio and delivered at 4–12 °C. The antegrade-only group received aortic-root cardioplegia, while the combined group’s induction dose was divided between antegrade and retrograde routes in a 2:1 ratio. The protocol specified maintenance dosing every 20–30 minutes. Both groups also received terminal warm-blood cardioplegia before cross-clamp removal. Thus, the study examined the incremental effect of an additional delivery route within an otherwise comprehensive protection strategy.

The primary endpoint was the highest high-sensitivity cardiac troponin I concentration measured at 4, 12, 24, or 48 hours after intensive care admission. Median peak values were 529.1 ng/L in the antegrade-only group and 394.1 ng/L in the combined group, with interquartile ranges of 829.9 and 503.4 ng/L, respectively. The difference was not statistically significant (p = 0.21). Although the combined group’s median was numerically lower, the trial did not establish a reliable reduction in postoperative myocardial injury.

Troponin exposure over time gave a similar result. The median area under the concentration–time curve from 4 to 48 hours was 12,910.6 ng·h/L with antegrade-only delivery and 8,112.2 ng·h/L with combined delivery (p = 0.16). Peak CK-MB was 42.0 versus 36.5 IU/L (p = 0.37). Serial CK-MB, total creatine kinase, and serum creatinine measurements also showed no significant differences. Troponin reflects overall perioperative myocardial injury, including surgical manipulation, ischemia, reperfusion, and other factors, so it cannot isolate the effects of cardioplegia alone.

Immediate myocardial electrical recovery was comparable. Spontaneous sinus rhythm returned in 70.0% of antegrade-only patients and 66.7% of combined-treatment patients. Defibrillation requirements and reported arrhythmias did not differ significantly. Cross-clamp duration was numerically shorter with combined delivery: 60.5 versus 72.4 minutes (p = 0.060). Cardiopulmonary bypass duration was 88.9 versus 99.3 minutes (p = 0.20). These differences did not meet the study’s statistical threshold, but unequal ischemic exposure remains relevant when interpreting biomarker comparisons in a small trial.

One exploratory secondary outcome favored combined delivery: median mechanical ventilation duration was 6.0 hours versus 8.5 hours with antegrade-only treatment (p = 0.010). This finding deserves cautious interpretation because secondary analyses were not adjusted for multiple comparisons and the trial was designed around a biochemical endpoint. Intensive care stay, transfusion requirements, and reoperation for bleeding did not differ significantly. Two patients in each group experienced myocardial infarction, and there were no deaths within 30 days. These low event counts cannot establish comparable clinical safety or effectiveness.

Retrograde cardioplegia was delivered successfully in all assigned patients without coronary sinus injury or other reported cardioplegia-related complications. This demonstrates feasibility in an experienced center. It does not rule out uncommon complications or establish the same results during a team’s learning curve. The participating surgeons had substantial experience with retrograde delivery, and the study population was relatively selected. Technical success should therefore be interpreted alongside operator experience, patient anatomy, and the modest number of procedures studied.

The main limitation is statistical uncertainty. The trial’s sample-size calculation assumed a 30% reduction in peak troponin and a distribution that differed from the markedly skewed values actually observed. This may have reduced sensitivity to modest treatment effects. The study was not powered for mortality or other infrequent clinical outcomes, and follow-up ended at 30 days. Baseline ventricular dimension and cross-clamp differences may also have influenced estimates. A nonsignificant superiority result does not establish equivalence, and it does not prove that no patient subgroup could benefit.

For perfusionists and cardiac surgical teams, the practical message is to consider cardioplegia route as one component of myocardial protection, together with solution, temperature, redosing, systemic perfusion, and reperfusion management. These findings do not support routinely adding retrograde delivery solely to lower postoperative troponin in comparable elective CABG patients. They also do not eliminate its potential role when antegrade distribution is inadequate. Larger multicenter randomized trials with adequate clinical endpoint assessment are needed to clarify whether severe obstruction, prolonged ischemic periods, or other high-risk features identify patients who benefit.

Source: Lesanovic J, et al. Medicina. 2026;62(9):1799. DOI: 10.3390/medicina62091799.

PubMed | Full Text Online

Study Ranking

3
Computer-generated randomization, blinded postoperative and laboratory outcome assessment, complete follow-up, and intention-to-treat analysis strengthen this trial. Its single-center sample of 60 patients, retrospective registration, biomarker rather than clinical primary endpoint, residual imbalances, and unadjusted exploratory comparisons limit certainty and generalizability. The nonsignificant result does not establish equivalence or exclude smaller benefits.