Beyond drainage: can pericardial troponin levels predict atrial fibrillation after coronary surgery?

A prospective study exploring pericardial fluid troponin I as a predictor of atrial fibrillation after coronary artery bypass grafting and proposing a new risk stratification model.

Postoperative atrial fibrillation (POAF) is one of the most common complications after cardiac surgery. It is associated with greater morbidity, longer hospital stays and increased healthcare resource use. Despite advances in perioperative management, accurately identifying which patients will develop this arrhythmia remains difficult. POAF prediction has traditionally relied on clinical and structural factors. However, these models do not always account for the myocardial injury inherent to surgery, which may play a major role in the development of electrical disturbances during the early postoperative hours.

Against this background, Shu et al., in a study published in JTCVS Open, proposed that cardiac troponin I (cTnI) concentrations in pericardial drainage fluid (PCF) might reflect local myocardial injury and provide additional prognostic information for predicting POAF.

The authors conducted a prospective single-centre study of 114 patients undergoing elective isolated coronary artery bypass grafting (CABG), with no history of atrial fibrillation (AF). Blood and PCF samples were collected six hours postoperatively to measure cTnI concentrations. The primary outcome was new-onset AF, defined as an episode lasting at least 30 seconds during the first seven days, detected by continuous electrocardiographic monitoring. POAF occurred in 35 patients (30.7%), with a median onset of 46 hours after surgery. Patients who developed the arrhythmia had significantly higher PCF-cTnI concentrations at six hours (25.5 vs. 18.7 ng/mL; p = .006), whereas blood troponin concentrations did not differ significantly between groups. PCF troponin concentrations were also higher than blood concentrations, with no significant correlation between the two.

In multivariable analysis, PCF-cTnI ≥27.2 ng/mL was independently associated with POAF (OR = 5.50; 95% CI = 1.89–16.05; p = .002). On this basis, the authors developed the CALT model, incorporating left circumflex artery lesions, left atrial diameter, left ventricular end-diastolic diameter, postoperative serum calcium and PCF-cTnI concentration.

The model achieved an area under the receiver operating characteristic (ROC) curve of 0.821, outperforming the conventional clinical scores assessed in the same cohort. However, adding pericardial troponin to the preceding model did not produce a statistically significant improvement in the area under the curve (p = .337).

COMMENTARY:

POAF is a complex pathophysiological phenomenon in which a susceptible atrial substrate interacts with multiple surgery-related triggers. From this perspective, the interest of the study by Shu et al. lies in exploring an aspect of arrhythmic risk that traditional clinical models scarcely address: the heart’s local response to surgical injury, beyond simply identifying another biomarker.

Cardiac troponin is an established marker of myocardial injury. Blood concentrations commonly rise after CABG and may reflect injury associated with cardiac manipulation, myocardial ischaemia and reperfusion, even in procedures performed without cardiopulmonary bypass (CPB). This perioperative release complicates its interpretation as a specific marker of an individual postoperative complication. The pericardial space may offer a different perspective. Unlike the systemic circulation, PCF is in direct contact with cardiac surfaces and may more closely reflect certain local processes. The finding of substantially higher troponin concentrations than in plasma, together with the absence of correlation between the two, supports the possibility that the pericardial compartment provides biological information distinct from routine blood testing.

This observation is particularly interesting from a surgical perspective. During CABG, the heart is exposed to a combination of mechanical, ischaemic and inflammatory insults whose magnitude is not always reflected in preoperative variables. PCF analysis might provide insight into this early tissue response and identify patients in whom surgery has created an environment particularly conducive to arrhythmia. However, these findings should not be interpreted causally. Troponin I is primarily a marker of myocardial injury, and its association with POAF does not demonstrate a direct role in arrhythmogenesis. Nor does it establish whether high concentrations reflect atrial injury, ventricular injury or both. Pericardial troponin may serve as an indirect indicator of inflammatory processes or myocardial stress that, in turn, promote electrical instability.

Research into pericardial biomarkers extends beyond troponin. Other studies have examined myeloperoxidase, mitochondrial DNA and their relationship with POAF, reinforcing interest in the pericardial environment as a potential source of prognostic information. Particularly relevant is the study by Luo et al., also published in 2026, which assessed pericardial myeloperoxidase in 469 patients and developed a predictive model validated in an independent cohort. These findings suggest that the potential value of PCF may extend beyond a single biomarker to a broader strategy for characterising the postoperative inflammatory and myocardial response.

A notable feature of the present study is the integration of structural and perioperative variables in the CALT model. The association of atrial and ventricular dilatation with POAF is pathophysiologically plausible: both may represent the cardiac substrate on which acute surgical triggers act. Some findings nevertheless require caution. Left circumflex artery disease was associated with a lower POAF risk, but this does not establish a protective effect of coronary disease in that territory. Uncontrolled confounding, differences in revascularisation patterns or statistical instability caused by the small sample may explain this association.

Sample size is precisely the study’s main limitation. Developing a model with five predictors from only 35 POAF events creates a substantial risk of overfitting, particularly when variable cut-offs are derived from the same cohort. Although the authors used LASSO regression and internal validation, the reduction in the area under the curve from approximately 0.82 to 0.76 after statistical correction shows that the initial performance probably overestimates predictive ability in new patients.

The improvement in the area under the curve after adding pericardial troponin was not statistically significant. This is fundamental: demonstrating an independent association between a biomarker and a complication does not establish a sufficiently meaningful improvement over an existing clinical prediction model. The improvement in net reclassification is promising, but requires confirmation in independent populations and, above all, evidence that it changes decisions with clinically relevant consequences.

The model’s applicability is limited by the selection of patients undergoing isolated CABG. We do not know whether the association is of similar magnitude in valve surgery, aortic surgery or combined procedures, where cardiac manipulation, ischaemic times and inflammatory responses may differ substantially. From a surgical standpoint, it is especially relevant that 93% of patients underwent off-pump CABG. This limits extrapolation to procedures using CPB, in which cardioplegia and the systemic inflammatory response may alter the magnitude and kinetics of troponin release. The 27.2-ng/mL cut-off identified by the authors should therefore not be applied directly to patients undergoing on-pump surgery without specific validation.

Ultimately, the most clinically important question is what we would do differently if we identified a patient with elevated pericardial troponin.

POAF prevention is already part of routine perioperative care, particularly through optimisation of beta-blocker therapy and additional pharmacological strategies in selected patients. For a new biomarker to be truly useful, identifying a higher-risk group is insufficient. It must help select a preventive intervention whose benefits outweigh its adverse effects and whose efficacy has been demonstrated in that setting.

In this respect, the study by Shu et al. is a first step, but it does not yet provide evidence to recommend pharmacological prophylaxis guided by pericardial troponin. Multicentre validation studies are needed, followed by trials testing whether individualised prevention based on this biomarker reduces POAF or improves other clinical outcomes.

The main merit of this work is that it shifts some of our attention from preoperative risk factors towards the biological response developing during the first postoperative hours. Cardiac surgery changes anatomy and haemodynamics and also triggers cellular and inflammatory processes whose local expression may provide valuable information about subsequent complications.

Pericardial fluid, usually regarded as a waste product of surgical drainage, may become a source of prognostic information. However, considerable work remains between identifying a biological signal and turning it into a tool that improves patient outcomes. The study’s real contribution may be less the discovery of another parameter for predicting AF than the reminder that the heart can express the consequences of surgical injury within its own microenvironment before they become clinically apparent. Learning to interpret that information may be another step towards more personalised cardiac surgery.

REFERENCE:

Shu Y, Gao C, Ren X, Deng X, Luo L, Wang T, et al. Pericardial fluid cardiac troponin I predicts postoperative atrial fibrillation after coronary artery bypass grafting. JTCVS Open. 2026 Jul 6;33:101954. doi: 10.1016/j.xjon.2026.101954. PMID: 42948431; PMCID: PMC13711195.

Luo L, Wang T, Peng T, Gao C, Shu Y, Yang H, et al. The pcMPO-AF rule for predicting postoperative atrial fibrillation after coronary artery bypass grafting. Sci Rep. 2026 Jan 21;16(1):2862. doi: 10.1038/s41598-025-32318-x. PMID: 41565722; PMCID: PMC12828024.

Gaudino M, Di Franco A, Rong LQ, Piccini J, Mack M. Postoperative atrial fibrillation: from mechanisms to treatment. Eur Heart J. 2023;44(12):1020-1039. doi:10.1093/eurheartj/ehad019.

Dobrev D, Aguilar M, Heijman J, Guichard JB, Nattel S. Postoperative atrial fibrillation: mechanisms, manifestations and management. Nat Rev Cardiol. 2019;16(7):417-436. doi:10.1038/s41569-019-0166-5.

Liblik K, Zucker J, Baranchuk A, Fernandez AL, Zhang S, Diasty ME. The role of pericardial fluid biomarkers in predicting post-operative atrial fibrillation, a comprehensive review of current literature. Trends Cardiovasc Med. 2024;34(4):244-247. doi:10.1016/j.tcm.2023.02.009.

SUBSCRIBE TO OUR MONTHLY NEWSLETTER..
XXVIII Resident Course
Get to know our magazine

Comparte esta información