Abstract
BACKGROUND: Perioperative myocardial injury (PMI) is a common complication following non-cardiac, particularly thoracic, surgery and is associated with increased cardiovascular risk. Although guidelines recommend cardiac biomarker monitoring to detect PMI, its implementation in routine clinical practice remains limited. OBJECTIVE: To evaluate the combined use of high-sensitivity cardiac troponin I (hs-cTnI) and N-terminal pro-brain natriuretic peptide (NT-proBNP) in predicting major adverse cardiovascular events (MACE) following elective thoracic surgery, and to determine whether NT-proBNP provides incremental prognostic value beyond hs-cTnI alone. DESIGN: Multicentre observational cohort study. SETTING: Conducted between February 2021 and November 2023 in three Spanish tertiary hospitals. PARTICIPANTS: Patients aged ≥45 years scheduled for elective thoracic surgery involving lung resection (pneumonectomy, lobectomy, bilobectomy or segmentectomy) under general anaesthesia. Exclusion criteria included urgent or non-thoracic surgery, active infection or sepsis and a history of severe heart failure (ejection fraction <30%). MAIN OUTCOME MEASURES: Combined measurement of hs-cTnI and NT-proBNP at baseline (preoperatively) and at 24 and 48 hours postoperatively.PMI was defined as hs-cTnI ≥45 ng/L at 24 and/or 48 hours or a ≥20% increase from baseline in patients with elevated preoperative concentrations. RESULTS: Among 475 patients, PMI occurred in 11.8%. PMI had higher rates of prior stroke (12.5% vs 2.9%; p=0.004), smoking history (85.7% vs 64.0%; p=0.001) and severe renal dysfunction (7.1% vs 0.7%; p=0.001), with similar Revised Cardiac Risk Index distribution. Patients with PMI also had greater postoperative elevations of hs-cTnI and NT-proBNP (p<0.001), longer surgeries (3.5 hours vs 2.7 hours; p<0.001) and more frequent lobectomy/bilobectomy (64.3% vs 50.4%; p<0.001). Robotic-assisted thoracic surgery (RATS) was associated with increased PMI risk (OR 2.29; p=0.019). Among 49 patients (10.3%) with dual postoperative elevation of hs-cTnI and NT-proBNP, cardiovascular comorbidities were common (hypertension 81.6%, smoking history 85.7%, stroke 14.3%), and most procedures were minimally invasive (video-assisted thoracic surgery 61.2%, RATS 24.5%), with a median duration of 3 hours 42 min. MACE occurred in 18.4% of this group, indicating a substantially elevated risk than isolated or no biomarker elevations. At 30 days, patients with PMI had higher MACE (14.3% vs 3.3%; p<0.001), mortality (3.6% vs 0.7%; p=0.049) and new-onset arrhythmias (5.3% vs 0.2%; p<0.001), particularly atrial fibrillation (7.1% vs 1.7%; p=0.011). Dual biomarker elevation was associated with the highest MACE risk (15.2%), representing a twofold to threefold increase over single biomarkers. CONCLUSIONS: Combined hs-cTnI and NT-proBNP assessment improves perioperative cardiovascular risk stratification beyond ischaemia. TRIAL REGISTRATION NUMBER: NCT04749212.