Authors: S. Rafieian, R. Ershadi, Hesam Amini, M. R. Eftekhari, M. R. Ghasri, S. Farzenehfar, Mehrshad Abbasi
Categories: Matters Arising, Lung resection surgery, ERNV scan, Right ventricular systolic pressure (RVSP), Right ventricle, Ejection fraction (EF)
Source: Journal of Cardiothoracic Surgery
Authors: S. Rafieian, R. Ershadi, Hesam Amini, M. R. Eftekhari, M. R. Ghasri, S. Farzenehfar, Mehrshad Abbasi
Lung resection surgeries, including lobectomy and pneumonectomy, are cornerstone treatments for lung cancer and other severe pulmonary conditions. Despite their therapeutic benefits, these procedures can compromise cardiopulmonary function, potentially increasing right ventricular (RV) workload due to reduced pulmonary vascular capacity. Such changes may precipitate RV dysfunction, even in patients with normal preoperative cardiac profiles, contributing to postoperative morbidity like dyspnea and arrhythmias. While pulmonary function tests are standard for preoperative risk assessment, cardiac impacts—particularly on the RV—remain insufficiently characterized. This study employs Equilibrium radionuclide ventriculography (ERNV) scanning, a precise radionuclide ventriculography technique, to evaluate biventricular systolic and diastolic changes post-resection, supplemented by echocardiography to measure Right Ventricular Systolic Pressure (RVSP).
Twenty patients (mean age 43.8 ± 8.3 years, 10 females, 10 male) undergoing lobectomy (n = 15) or pneumonectomy (n = 5) from March 2021 to September 2022 were prospectively studied at a tertiary university hospital. Patients with preexisting coronary disease, hypertension, or abnormal pulmonary function (e.g., predicted postoperative FEV1 or DLCO < 60%) were excluded. Cardiac function was assessed preoperatively and two months postoperatively using ERNV scans to quantify right and left ventricular ejection fraction (EF), peak filling rate (PFR), and time to peak filling rate (TPFR). Transthoracic echocardiography measured RVSP concurrently. Statistical analyses included paired t-tests to compare pre- and postoperative indices, general linear models to assess surgery type effects, and regression analyses to correlate resection extent with RV changes.
Lung resection significantly impaired cardiac performance. Right ventricular EF declined from 48.8 ± 2.6% to 43.7 ± 2.8% (p = 0.035) and left ventricular EF from 54.9 ± 2.7% to 51.3 ± 3.3% (p < 0.001). Diastolic function deteriorated, with the right PFR decreasing from 1.9 ± 0.2 to 1.7 ± 0.2 EDV/s (p < 0.001) and left PFR from 2.4 ± 0.1 to 2.2 ± 0.2 EDV/s (p = 0.001), while TPFR increased bilaterally (right: 151.8 ± 18.6 to 161.4 ± 17.6 ms, p < 0.001; 168.9 ± 11.7 to 176.1 ± 13.5 ms, p < 0.001). RVSP rose from 20 ± 2.6 to 24.9 ± 5.0 mmHg (p = 0.001). Pneumonectomy elicited greater reductions in RV indices than lobectomy, with resection extent strongly predicting RVEF decline (r = 0.7, p < 0.001).
Lung resection induces substantial biventricular systolic and diastolic dysfunction, with severity proportional to resection extent. These findings highlight the utility of ERNV scanning in detecting subtle cardiac changes and emphasize the importance of preoperative cardiac evaluation to anticipate and manage postoperative complications, particularly in extensive resections like pneumonectomy.
The online version contains supplementary material available at 10.1186/s13019-025-03781-4.
Lung cancer remains the leading cause of cancer-related mortality globally. Lung resection, a common therapeutic intervention, is frequently associated with postoperative cardiopulmonary morbidity, manifesting as dyspnea and diminished functional capacity. The reduction in pulmonary vascular bed following surgery may increase right ventricular (RV) workload, potentially leading to RV dysfunction, even in patients with normal preoperative cardiac function [1]. It is well-established that pulmonary resection can precipitate decompensation of pulmonary capacity, impairing the ability to meet postoperative oxygen demands [2]. To mitigate these adverse outcomes, preoperative pulmonary function tests (PFTs) and quantitative assessments of lung perfusion or ventilation are recommended [3, 4]. However, despite satisfactory preoperative pulmonary function, some patients experience postoperative complications. Underlying conditions such as tumors or pulmonary parenchymal diseases can elevate Right Ventricular Systolic Pressure (RVSP), and the loss of pulmonary vascular bed due to resection may further exacerbate RVSP, increasing RV workload and predisposing patients to cardiac-related postoperative complications [5, 6]. Conventional two-dimensional (2D) echocardiography has limitations in accurately assessing RV function. Equilibrium Radionuclide Ventriculography (ERNV), also known as Equilibrium radionuclide ventriculography (ERNV) scanning, is a nuclear medicine imaging modality that provides a robust evaluation of cardiac dynamics, particularly RV ejection fraction (RVEF) and diastolic indices. To our knowledge, ERNV has not previously been utilized to assess RVEF in patients undergoing lung resection. This study aims to investigate RV and left ventricular (LV) systolic and diastolic function indices using ERNV in this patient population.
Twenty patients (mean age 43.8 ± 8.3 years; 10 females, 10 males) scheduled for lung resection were enrolled between March 2021 and September 2022 at the thoracic surgery unit of a tertiary university hospital. Exclusion criteria included a history or clinical suspicion of coronary artery disease, hypertension, abnormal PFT results, prior lung surgery, tricuspid regurgitation, heart failure, or pregnancy. Baseline patient characteristics are detailed in (Table 1).
Table 1Health characteristics, reason for surgery, and type of surgery of participantsHealth indicesFrequency or mean†AGE43.9(8.3)BMI (kg/m^2^)23.9(4.3)SEXMale10(50)Female10(50)Total20(100)SmokingNo4(20)Yes16(80)Surgery typePneumonectomy5(25)Lobectomy15(75)Resection LocationLeft2(10)LLL3(15)LUL4(20)Right3(15)RLL3(15)RML2(10)RUL3(15)DiagnosisAV malformation1(5)Bronchiectasis3(15)Lung Cancer13(65)Hydatid Cyst3(15)†Data are frequency and percentage in parentheses or mean and standard deviation in parentheses
PFTs were performed, and predictive postoperative forced expiratory volume in 1 s (FEV1) and diffusing capacity for carbon monoxide (DLCO) were calculated. Patients with predictive postoperative FEV1 < 60% or DLCO < 60% were excluded. Quantitative pulmonary perfusion scans were conducted using multi-planar projections 5 min following intravenous administration of 3 mCi of macroaggregated albumin (MAA; Pars Isotope, Tehran, Iran) to estimate the percentage of lung resection by laterality (Figures. 1, 2 and 3).Fig. 1Change of ejection fractions of the right and left ventricles (i.e., LVEF and RVEF) after surgery. The type of surgery significantly interacted with RVEF changes; while ejection fraction was reduced by surgery, RVEF reduction was more in patients undergoing pneumonectomyFig. 2Change of peak filling rate (PFR) and time to peak filling rate (TPFR) in the right and left ventricles. Except left PFR, pneumonectomy enhances the negative variations of all other indices after surgeryFig. 3Change of pulmonary artery pressure (RVSP) before and after surgery. RVSP was raised after lung resection surgery more in the patients who underwent pneumonectomy compared to those who had lobectomy surgery
Cardiac function was assessed using ERNV at two time preoperatively and two months postoperatively. For ERNV, 3 mL of blood was withdrawn, labeled with technetium-99 m (Tc99m) using ultrafast kits (Pars Isotope, Tehran, Iran), and reinjected within 1 h. Gated planar imaging was performed using a dual-head gamma camera (ADAC Forte, Philips, CA, USA) at the optimal septal view [7, 8]. Regions of interest (ROIs) for the RV, LV, and background were delineated using a semi-automated method. Systolic and diastolic indices, including ejection fraction (EF), peak filling rate (PFR), and time to peak filling rate (TPFR), were calculated for both ventricles. Transthoracic echocardiography was performed concurrently with ERNV pre- and postoperatively to estimate RVSP. Echocardiographically estimated right ventricular systolic pressure (RVSP) was used as a surrogate for pulmonary artery pressure (PAP), given their equivalence in the absence of right ventricular outflow tract obstruction. For consistency and scientific accuracy, the term RVSP is used throughout this manuscript.
Pre- and postoperative RV and LV function indices were compared using a paired-sample t-test. A general linear model assessed the influence of surgery type (pneumonectomy vs. lobectomy) on postoperative changes in these indices. Bivariate regression analysis evaluated the correlation between the extent of lung resection (based on preoperative perfusion scans) and RVEF reduction. Receiver operating characteristic (ROC) curve analysis determined the accuracy of preoperative LV ejection fraction (LVEF) in predicting postoperative LVEF < 50%, with the optimal preoperative LVEF threshold calculated as the point with the shortest distance to the curve’s ideal sensitivity-specificity coordinate.
All 20 patients completed the study, undergoing reevaluation two months postoperatively without complications necessitating reoperation. Six patients (30%) experienced atrial fibrillation (AF), four of whom had undergone pneumonectomy. AF resolved within one week in all post-lobectomy cases, but persisted in three pneumonectomy patients at two months. Systolic and diastolic function indices for both ventricles are presented in Table 2. RVEF and LVEF were significantly reduced postoperatively (RVEF 5.2 ± 2.8%; p < 0.05). Bilateral PFR decreased significantly, while TPFR increased significantly compared to preoperative values (p < 0.05). RVSP also rose significantly post-resection (p < 0.05). Surgery type significantly influenced postoperative changes in RVEF, right PFR, bilateral TPFR, and RVSP (Table 3), with pneumonectomy exerting a greater impact on RV systolic and diastolic indices, RVSP, and left TPFR than lobectomy. The extent of lung resection, as determined by preoperative perfusion scans, strongly correlated with RVEF reduction (r = 0.7, p = 0.001). ROC analysis revealed that preoperative LVEF predicted postoperative LVEF < 50% with 89.4% accuracy, with an optimal preoperative LVEF cutoff of 55%.
Table 2Cardiac systolic (i.e., ejection fractions) and diastolic (i.e., peak filling rate and time to peak filling rate) indices of right and left ventricles before and after lung resection (n = 20)Before surgeryAfter surgerySignificanceRVEF %48.8(2.6)43.7(2.8)0.035LVEF%54.9(2.7)51.3(3.3)0.000Right PFR (EDV/sec^2^)1.9(0.2)1.7(0.2)0.000Left PFR (EDV/sec)2.4(0.1)2.2(0.2)0.001Right TPFR (mSec)151.8(18.6)161.4(17.6)0.000Left TPFR (mSec)168.9(11.7)176.1(13.5)0.000RVSP (mm-Hg)20(2.6)24.9(5.0)0.001RV, right ventricle; LV, left ventricle; EF, Ejection fraction; PFR, peak filling rate; and TPFR, and time to peak filling rateData are mean and standard deviation in parentheses
Table 3The statistics indicating the effect of type of surgery on the extent of variation of right and left ventricular systolic (ejection fraction, EF) and diastolic (PFR, peak filling rate; and TPFR, and time to peak filling rate) indices as well as pulmonary artery pressure (RVSP). Surgery type effect was significant on RVEF, right PFR, bilateral TPFR, and RVSP changes (P values < 0.05)PneumonectomyN = 5LobectomyN = 15Sig.Partial eta squaredObserved powerBeforesurgeryAfter surgeryBefore surgeryAfter surgeryRVEF49.2(2.6)40.6(1.3)48.7(2.7)44.7(2.4)0.0000.5410.992LVEF57.2(2.2)53.8(3)54.1(2.5)50.5(3.1)0.8650.0020.053Right PFR2.1(0.1)1.8(0.1)1.8(0.2)1.7(0.2)0.0410.2130.551Left PFR2.5(0.1)2.4(0.1)2.3(0.1)2.2(0.2)0.4160.0370.124Right TPFR159.6(23.1)165(20.9)149.1(17)160.1(17)0.0070.3410.823Left TPFR178(2.7)188.4(3.2)165.8(12)172(13.1)0.0220.2590.660RVSP21.6(3.2)30(6.1)19.5(2.3)23.2(3.4)0.0100.3150.777
This study demonstrates that both pneumonectomy and lobectomy impair RV and LV systolic and diastolic function two months postoperatively. To our knowledge, this is the first study to simultaneously assess RVEF, right PFR, and right TPFR using ERNV in the context of lung resection. Pneumonectomy exerted a more pronounced effect on RVEF, right PFR, TPFR, and RVSP compared to lobectomy. Although lobectomy is generally considered to have minimal impact on cardiac function and RVSP [9, 10], our findings align with evidence suggesting RV dimensional changes post-resection [3]. A recent cardiac magnetic resonance study similarly reported RVEF reduction following lung resection, consistent with our ERNV-based results [11]. Notably, all systolic and diastolic indices in our cohort changed significantly post-lobectomy, except for RVSP, which showed a non-significant increase.
Lung resection imposes substantial cardiopulmonary stress, underscoring the importance of preoperative risk stratification. Indications for resection included lung carcinoma, bronchiectasis, hydatid cysts, and arteriovenous malformations, with 75% of patients undergoing lobectomy. Preoperative ventricular function assessment may predict postoperative atrial arrhythmias [12, 13]. In this study, 30% of patients experienced AF, predominantly in the pneumonectomy group, with three cases persisting at two months. AF may result from hypoxemia, vagal stimulation, or atrial inflammation, with prevalence rates aligning with prior reports [12, 13]. The two-month follow-up was chosen to allow resolution of transient postoperative cardiac changes, ensuring observed alterations reflect intrinsic remodeling. Few studies have explored ventricular changes post-lung resection [14]. Our findings indicate that greater resection extent and vascular bed loss correlate with more pronounced RV dysfunction, with RVEF reduction strongly tied to resection magnitude (r = 0.7).
Afterload alterations are a primary driver of RV dysfunction post-resection [15, 16], a finding supported by the correlation between resection extent and RVEF reduction in this study. Pneumonectomy patients exhibited greater RV dysfunction at two months. The observed decline in RV function, despite RVSP remaining within normal limits, suggests a surgically induced increase in pulmonary arterial resistance. Although clinically significant in patients with baseline subnormal function, its relevance in our cohort with normal preoperative indices warrants further investigation. These findings advocate for baseline RVEF measurement to identify patients at risk of falling below normal thresholds post-resection. Limitations This study included relatively young patients without baseline cardiac abnormalities, limiting generalizability to older or comorbid populations. Future research should evaluate patients with preexisting cardiac dysfunction, who may be more susceptible to complications. Additionally, the two-month follow-up may not capture long-term normalization or progression of RV dysfunction, necessitating extended observation.
In addition to right ventricular alterations, a modest but significant reduction in left ventricular ejection fraction (LVEF) was also observed after lung resection. Although this finding may seem unexpected, several interrelated physiological mechanisms may explain it. First, the ventricular interdependence within the shared pericardial sac allows right ventricular (RV) pressure overload and septal displacement following resection to transiently impair left ventricular (LV) filling and geometry [17, 18]. Second, mediastinal shift and pericardial traction after pneumonectomy can alter LV orientation and wall motion, leading to a functional reduction in global LVEF on gated radionuclide imaging [19]. Third, the hemodynamic consequences of reduced preload—owing to loss of pulmonary vascular bed and venous reservoir—decrease LV end-diastolic volume and myocardial fiber stretch. According to the Frank–Starling principle, this diminished preload limits stroke volume and ejection fraction despite preserved intrinsic contractility [20]. Finally, transient myocardial strain induced by postoperative hypoxemia, anesthetic effects, or systemic inflammation may further contribute to mild, reversible LV depression [21]. The small magnitude of LVEF decline observed in our cohort (approximately 3–4%) thus most likely reflects transient geometric and preload-dependent adjustments rather than intrinsic myocardial dysfunction.In this study, ROC curve analysis indicated that a preoperative LVEF of 55% served as the optimal threshold for predicting postoperative LVEF below 50%, with an overall predictive accuracy of 89.4%. This finding indicates that patients with subnormal LVEF before surgery may have limited contractile reserve and are consequently at higher risk of postoperative ventricular decompensation. In such cases, individualized multidisciplinary assessment is warranted, including preoperative optimization of cardiac status, consideration of limited resection volumes guided by quantitative perfusion, and postoperative surveillance with radionuclide or echocardiographic follow-up to detect early functional decline.
The relatively narrow standard deviations of ventricular indices observed in this study likely result from strict exclusion of patients with baseline cardiopulmonary disease and the use of standardized single-center ERNV acquisition by a uniform operator and processing protocol. Such methodological control minimizes inter-observer and equipment-related variability, explaining the lower dispersion compared with population-based MUGA studies involving older or clinically heterogeneous cohorts.
The intent of this study was primarily to assess left and right ventricular functional changes in patients undergoing pulmonary resection, rather than to compare imaging modalities or advocate for one instrument over another. This study does not propose ERNV as a replacement for echocardiography but rather as a complementary imaging approach. While echocardiography remains the standard method for evaluating right ventricle–pulmonary artery coupling, tricuspid annular plane systolic excursion, and PAP, ERNV offers highly reproducible volumetric indices that can sensitively detect biventricular systolic and diastolic alterations, particularly in postoperative thoracic patients with limited acoustic access. Moreover, echocardiography is inherently operator dependent, and quantitative assessment of right ventricular indices is not routinely performed due to technical challenges and variability in measurement reproducibility [22, 23]. In this context, ERNV provides a standardized, operator-independent method for serial evaluation of ventricular function. The integration of both modalities may therefore yield a more comprehensive and reliable assessment of ventricular adaptation following lung resection.
The main limitation of this study is the relatively small sample size, reflecting the limited number of patients undergoing lung resection during the study period who met the eligibility criteria for cardiac evaluation. In addition, quantitative echocardiographic measurements of ventricular ejection fractions were not systematically obtained, as echocardiography in the study protocol was primarily used for estimating RVSP. Future studies with larger cohorts and combined echocardiographic–radionuclide analyses are warranted to enhance cross-modality validation of ventricular function. Furthermore, clinical symptom data were not collected in this study; future investigations should correlate imaging-derived ventricular changes with postoperative symptoms and functional outcomes to better determine the clinical relevance of subclinical dysfunction.
Lung resection increases RVSP and impairs RV systolic and diastolic function proportional to resection extent. These findings highlight the utility of ERNV in assessing postoperative cardiac changes and underscore the need for preoperative cardiac evaluation to optimize surgical outcomes.
Supplementary Material 1.