This single-center retrospective observational study included adult patients who underwent Mitral Valve Replacement (MVR) at the Affiliated Hospital of Qingdao University between April 2020 and April 2025. The study was approved by the hospital's Ethics Committee (Approval Number: QYFYWZLL30776), and the requirement for informed consent was waived due to the retrospective nature of the study. Patients in the Dapagliflozin group received standardized preoperative treatment according to our institutional protocol for managing concomitant heart failure, which consisted of oral dapagliflozin at a dose of 10 mg once daily. The preoperative treatment duration was at least 2 weeks, with a median duration of 4 weeks.

Inclusion Criteria: (a) Patients who underwent MVR at the Affiliated Hospital of Qingdao University; (b) Patients who underwent preoperative and postoperative echocardiographic and cardiac troponin assessment; (c) Left ventricular ejection fraction (LVEF) < 50%at baseline (to select a cohort with systolic left ventricular dysfunction, the primary driver of secondary mitral regurgitation and post-capillary pulmonary hypertension in this context) (d) Age ≥ 18 years. (e) Surgery was indicated for mitral regurgitation (MR).

Exclusion Criteria: (a) History of pulmonary hypertension (not caused by left heart dysfunction); (b) Emergency cardiac surgery; (c) Preoperative dapagliflozin use not meeting the standard regimen (i.e., a dose of 10 mg once daily for a minimum of 2 weeks prior to surgery); (d) Redo cardiac surgery; (e) Concomitant coronary artery bypass grafting (CABG); (f) Surgery involving artificial vessel replacement; (g) Significant hepatic or renal dysfunction. (h) Isolated predominant mitral stenosis.

Baseline demographic and clinical data were recorded: age, sex, body mass index (BMI), blood pressure, history of hypertension, history of diabetes mellitus, operative time, beta-blocker use, angiotensin receptor-neprilysin inhibitor (ARNI) use, NT-proBNP level, left ventricular ejection fraction (LVEF), main pulmonary artery diameter (MPA), pulmonary artery systolic pressure (PASP), tricuspid annular plane systolic excursion (TAPSE), right ventricular basal diameter (RVBD), and NYHA functional class. Preoperative echocardiographic measurements were obtained during the routine pre-surgical evaluation within one week prior to mitral valve replacement (MVR). Postoperative measurements were conducted at a scheduled follow-up visit one week after hospital discharge, aiming to assess pulmonary vascular status in the early postoperative phase when patients were clinically stable.

The diagnosis of pulmonary hypertension was confirmed by an experienced cardiologist. Transthoracic echocardiography was performed for each patient before and after surgery to assess ejection fraction, main pulmonary artery (MPA) diameter, pulmonary artery systolic pressure (PASP), tricuspid annular plane systolic excursion (TAPSE), and right ventricular basal diameter (RVBD). The MPA diameter was specifically measured in the parasternal pulmonary artery long-axis view at end-systole, just proximal to the bifurcation. All measurements were performed following a standardized institutional protocol aligned with the recommendations of the American Society of Echocardiography (ASE) to ensure consistency and reproducibility. To minimize inter-operator variability and subjective bias, measurements of MPA diameter were performed independently by two experienced cardiac sonographers who were blinded to patient group allocation. The inter-observer agreement was assessed by calculating the intraclass correlation coefficient (ICC) using a two-way random-effects model for absolute agreement. The resulting ICC was (0.93), indicating excellent reliability. For statistical analysis, the average of the two measurements was used.

The primary clinical outcome was the percentage change in echocardiographic parameters from preoperative to postoperative measurements. Secondary clinical outcomes included changes in biomarker levels and other echocardiographic variables.

All statistical analyses were performed using SPSS version 26.0. Continuous variables conforming to a normal distribution are presented as mean ± standard deviation and were analyzed using the t-test; non-normally distributed variables are presented as median and were analyzed using the Wilcoxon signed-rank test. Categorical data are presented as numbers and were compared using the Chi-square test or Fisher's exact test. Pre- and postoperative results were compared using paired t-tests. A p-value < 0.05 was considered statistically significant for all analyses.

Our data consistently indicate that dapagliflozin treatment is associated with a smaller MPA diameter (Figure 2) and significantly inhibits the perioperative percentage change in MPA diameter. After excluding significant baseline imbalances (Table 1) and interference from other cardiac functional parameters (such as LVEF, TAPSE, RVBD), this specific effect on pulmonary artery structure is particularly prominent. It is noteworthy that this effect was observed in a cohort specifically selected for having LVEF < 50%, representing patients with systolic dysfunction as the core mechanism for secondary mitral regurgitation and elevated left-sided filling pressures. MPA diameter is an important morphological indicator reflecting long-term pulmonary artery load and the intrinsic properties of the vascular wall (7). Its dilation is typically recognized as a surrogate imaging marker for pulmonary hypertension and vascular remodeling (7). Therefore, our findings are consistent with the possibility that dapagliflozin may contribute to stabilizing pulmonary artery structure and inhibiting adverse remodeling during the perioperative stress state.

We acknowledge the reviewer's important point that classical pulmonary vascular remodeling is a chronic process. The MPA diameter change observed in our study, following a median preoperative treatment period of 4 weeks, is more likely to reflect rapid modulation of hemodynamics and vascular function by dapagliflozin rather than chronic structural remodeling. Recent evidence from major heart failure trials confirms that dapagliflozin induces rapid hemodynamic and clinical benefits across the spectrum of ejection fraction (18). Potential mechanisms underlying this early effect may include: (1) rapid preload reduction through natriuresis and diuresis, thereby decreasing passive distending forces on the pulmonary arterial wall; and (2) pleiotropic vascular actions, such as anti-inflammatory and endothelial-improving effects, which may optimize vascular compliance in the short term. Thus, the “structural” stabilization we report may represent an early morphological consequence of these functional and hemodynamic improvements.

The most insightful finding of this study is the dissociation between the improvement in MPA and pulmonary artery pressure, i.e., “pressure-structure decoupling”. Although the MPA diameter was smaller and more stable in the Dapagliflozin group, the estimated pulmonary artery systolic pressure (PASP) showed no significant difference between the groups (percentage change p = 0.812). This raises the hypothesis that the benefit of dapagliflozin may not be primarily achieved through direct reduction of pulmonary artery pressure, but rather by pressure-independent mechanisms that could confer vascular protection.

We propose the following mechanistic hypotheses to explain this phenomenon:

Direct Anti-inflammatory and Anti-fibrotic Vascular Protection: Growing evidence indicates that SGLT2 inhibitors possess significant anti-inflammatory and anti-fibrotic properties (14, 19). Heart failure and surgical stress can trigger systemic and pulmonary vascular inflammatory responses, driving vascular matrix degradation and fibrosis, leading to vascular dilation and reduced compliance (20, 21). Dapagliflozin may directly inhibit these pathological processes (8), enhancing the elasticity and structural integrity of the pulmonary arterial wall, thereby resisting dilative remodeling.

This study is the first to reveal an association between SGLT2 inhibitors and improved pulmonary artery structure in a perioperative cardiac surgery population, adding a new dimension to their organ protection profile—pulmonary artery protection. The finding of a smaller postoperative MPA diameter is clinically significant. As the main conduit of the right ventricular (RV) outflow tract, the MPA is a primary determinant of RV afterload. A smaller diameter implies reduced wall tension and, consequently, lower RV afterload. Given the particular sensitivity of the thin-walled RV to afterload changes, attenuating this load during the postoperative period may facilitate RV functional recovery and adaptation.

In cardiac surgery, right ventricular dysfunction is an important prognostic factor. The structure and function of the pulmonary artery are important contributors to right ventricular afterload, and the main pulmonary artery (MPA) diameter is a recognized surrogate imaging marker for these underlying changes (6, 10, 22). This is supported by growing imaging evidence linking MPA diameter to outcomes. For instance, recent studies have identified MPA diameter as an independent prognostic imaging marker for RV dysfunction (23). Although derived from different clinical contexts (e.g., acute pulmonary embolism), the underlying physiological principle is consistent: pulmonary artery dilation reflects chronic pressure or volume overload. Therefore, inhibiting early postoperative MPA dilation, as observed with dapagliflozin, may indicate a more favorable RV-pulmonary artery coupling state. Inhibiting pathological pulmonary artery dilation may provide a more favorable “downstream” environment for the right ventricle, which could have profound implications for patients' long-term outcomes.

It must be noted that this is an observational study and cannot establish causality. Furthermore, pulmonary artery pressure was indirectly estimated by echocardiography; future studies employing direct measurement via right heart catheterization, particularly measuring mean pulmonary artery pressure and pulmonary capillary wedge pressure, could more precisely reveal the hemodynamic mechanisms. Additionally, although the preoperative and postoperative assessment time windows were standardized (within one week before surgery and one week after discharge), the exact timing of echocardiographic measurements was not uniform across all patients, which is a common limitation in retrospective studies. Future prospective studies with strictly fixed time points would be valuable. Finally, we acknowledge that the sample size of this study is limited and the conclusions require validation by larger prospective studies. Furthermore, while we propose plausible mechanisms, they remain hypothesis-generating due to the lack of direct measures of pulmonary vascular properties or left-sided filling pressures. The observed change in MPA diameter is thus best interpreted as a promising surrogate endpoint, with its precise mechanistic underpinnings awaiting future confirmation.