Tumor-to-thoracic height and width ratios as adjunct imaging markers for preoperative risk stratification in minimally invasive surgery for pediatric mediastinal masses Authors: Satoshi Makita, Hiroo Uchida, Akinari Hinoki, Chiyoe Shirota, Takahisa Tainaka, Hizuru Amano, Katsuhiro Ogawa, Masamune Okamoto, Aitaro Takimoto, Akihiro Yasui, Shunya Takada, Kaito Hayashi, Yoichi Nakagawa, Daiki Kato, Hiroki Ishii, Hajime Asai, Kazuki Ota, Yui Murata, Yaohui Guo, Jiahui Liu Published: 2026 ABSTRACT PurposeSelecting the optimal surgical approach for pediatric mediastinal masses is challenging because absolute tumor size does not adequately reflect the available intrathoracic working space in children. This study evaluated whether the tumor-to-thoracic height ratio (TTH) and tumor-to-thoracic width ratio (TTW) are associated with intraoperative complications during minimally invasive surgery (MIS).MethodsSixty-five children who underwent surgical resection of mediastinal masses between 2007 and 2025 were retrospectively reviewed. Fifty-six underwent MIS, and nine underwent open thoracotomy. TTH and TTW were calculated using preoperative chest radiography and computed tomography or magnetic resonance imaging. Associations with intraoperative complications in the MIS group were analyzed, and receiver operating characteristic analysis was performed as an exploratory assessment.ResultsIntraoperative complications occurred in five of 56 patients undergoing MIS. Both TTH and TTW were significantly higher in patients with complications than in those without (TTH: 0.51 vs. 0.26, p = 0.004; TTW: 0.67 vs. 0.37, p = 0.005). Positive image-defined risk factors were also more frequent in patients with complications. The area under the curve was 0.894 for TTH and 0.882 for TTW.ConclusionTTH and TTW may serve as useful exploratory adjunct imaging markers for preoperative risk stratification in pediatric MIS for mediastinal masses. Introduction Minimally invasive surgery (MIS), including thoracoscopic and robot-assisted approaches, has increasingly been applied to the treatment of pediatric mediastinal masses because of its potential advantages over open thoracotomy, including reduced postoperative pain, shorter hospital stay, and improved cosmetic outcomes [1–6]. However, intraoperative complications, such as bleeding and injury to adjacent vital structures, remain significant concerns, particularly in cases with limited intrathoracic working space. In adult thoracic surgery, tumor size has traditionally been considered an important factor in determining the feasibility of MIS, and larger mediastinal masses are often regarded as less suitable for thoracoscopic resection, with a commonly used cutoff of approximately 5–6 cm [7–9]. However, in pediatric patients, the thoracic cavity varies markedly with age and body size, making absolute tumor diameter alone an insufficient indicator for selecting the surgical approach [10, 11]. Consequently, the choice between MIS and open thoracotomy in children remains largely based on surgeon experience and subjective judgement. To overcome this limitation, several imaging-based indices have been proposed. Shiiya et al. reported that the tumor-to-thoracic height ratio (TTH) may be useful for estimating the feasibility of reduced-port thoracoscopic surgery in children with mediastinal lesions [12]. Young et al. also suggested that imaging characteristics, tumor size relative to body weight, and image-defined risk factors (IDRFs), originally proposed for neuroblastoma, may help guide surgical planning [13]. Nevertheless, the available evidence remains limited, and objective quantitative criteria for assessing technical difficulty and the risk of intraoperative complications in pediatric MIS have not been fully established. Mediastinal masses in children vary not only in overall size but also in their orientation within the thoracic cavity. Some lesions extend predominantly in the craniocaudal direction, whereas others occupy a broader transverse space, potentially reducing the available operative working space during MIS. Therefore, it was hypothesized that relative tumor dimensions referenced to thoracic cavity size, rather than absolute tumor diameter alone, may better reflect technical difficulty in pediatric MIS. In this study, a retrospective analysis was conducted to evaluate whether TTH and the tumor-to-thoracic width ratio (TTW) were associated with intraoperative complications in children undergoing minimally invasive resection of mediastinal masses, with the aim of identifying simple imaging-based parameters that may assist in preoperative surgical planning. Methods Study design and patients A retrospective single-center study was conducted of consecutive pediatric patients who underwent surgical resection of a mediastinal mass at the Nagoya University Graduate School of Medicine between January 2007 and December 2025. Patients who underwent either MIS, including thoracoscopic and robot-assisted procedures, or open thoracotomy were included. At the institution, thoracoscopic surgery has been adopted as the first-line approach for mediastinal mass resection whenever technically feasible. Complete video-assisted thoracoscopic surgery using a four-port technique was introduced in 2007, and robot-assisted surgery was introduced in 2024. Given the long study period, it was confirmed that the occurrence of complications was not clustered in the early phase of MIS adoption, suggesting that the results were not solely driven by the initial learning curve. Open thoracotomy was selected when adequate working space for MIS was not anticipated, when preoperative imaging suggested marked vascular or chest wall involvement, or when the patient’s respiratory condition was considered too unstable to tolerate artificial pneumothorax. Surgical procedures Thoracoscopic procedures were performed without one-lung ventilation, using artificial pneumothorax with CO2 insufflation, in the lateral decubitus position. A complete thoracoscopic approach was employed using four intercostal ports (3–5 mm). Resected specimens were placed in a retrieval bag, fragmented when necessary, and removed through a port site. Robot-assisted procedures were also performed via a lateral approach. Open thoracotomy was performed without one-lung ventilation via a posterolateral approach. In all groups, the operative approach was selected at the discretion of the attending surgeon based on preoperative imaging findings and anticipated technical difficulty. Radiological assessment Preoperative chest radiographs were used to measure thoracic dimensions on the operative side. For lesions located in the anterior mediastinum, measurements were based on the side selected for the surgical approach. Thoracic height was defined as the distance from the lung apex to the base of the costophrenic angle, and thoracic width as the distance from the base of the costophrenic angle to the lateral border of the spine. Tumor height and width were defined as the maximum craniocaudal and transverse diameters of the lesion, respectively, measured on coronal computed tomography images; when such images were unavailable, coronal magnetic resonance imaging was used instead (Fig. 1). TTH was calculated as tumor height divided by thoracic height, and TTW as tumor width divided by thoracic width.Fig. 1Measurement of tumor-to-thoracic height ratio (TTH) and tumor-to-thoracic width ratio (TTW) A Thoracic height (A) was defined as the distance from the lung apex to the base of the costophrenic angle (mm). Thoracic width (B) was defined as the distance from the base of the costophrenic angle to the lateral border of the spine (mm) Tumor height (a) and width (b) were defined as the maximum longitudinal and transverse diameters of the mass (mm), respectively A and B were measured on chest radiographs, whereas a and b were measured on coronal computed tomography (CT) images. If coronal CT images were unavailable, coronal magnetic resonance imaging (MRI) was used instead TTH was calculated as a/A, and TTW as b/B Measurement of tumor-to-thoracic height ratio (TTH) and tumor-to-thoracic width ratio (TTW) A Thoracic height (A) was defined as the distance from the lung apex to the base of the costophrenic angle (mm). Thoracic width (B) was defined as the distance from the base of the costophrenic angle to the lateral border of the spine (mm) Tumor height (a) and width (b) were defined as the maximum longitudinal and transverse diameters of the mass (mm), respectively A and B were measured on chest radiographs, whereas a and b were measured on coronal computed tomography (CT) images. If coronal CT images were unavailable, coronal magnetic resonance imaging (MRI) was used instead TTH was calculated as a/A, and TTW as b/B Based on preoperative imaging findings, tumors were also categorized as predominantly cystic, predominantly solid, or heterogeneous. In addition, IDRFs, originally described for neuroblastoma, were evaluated in all cases as anatomical risk features on preoperative imaging. The following findings were considered positive: compression of the trachea or main bronchus, encasement of the aorta or its major branches, and extension to the costovertebral junction [14]. All radiological measurements were retrospectively performed by a single pediatric surgeon. Outcome measures The primary objective of the study was to evaluate whether TTH and TTW were associated with intraoperative complications in patients undergoing MIS for mediastinal masses. Intraoperative complications were defined as major adverse events, including vascular injury, bronchial injury, and conversion to open thoracotomy due to technical difficulty or uncontrolled bleeding. As secondary descriptive analyses, clinicopathological and radiological characteristics were compared between the open thoracotomy and MIS groups, including patient demographics, tumor location, tumor size, radiological features, operative time, blood loss, histopathological diagnosis, and perioperative complications. Because selection of the surgical approach was not standardized, comparisons between the two groups were considered descriptive rather than comparative assessments of treatment effect. Statistical analysis Continuous variables are presented as medians and interquartile ranges, and categorical variables as numbers and proportions. Comparisons between groups were performed using the Mann-Whitney U test for continuous variables and Fisher’s exact test for categorical variables. Within the MIS cohort, exploratory univariable analyses were performed to examine associations between preoperative factors and intraoperative complications. Because only five intraoperative complication events occurred, multivariable modeling was not considered reliable and was therefore not used for primary inference. Receiver operating characteristic (ROC) curve analysis was conducted as an exploratory assessment of the discriminatory performance of TTH and TTW for intraoperative complications in the MIS group. The area under the curve (AUC), sensitivity, and specificity were calculated, and the derived cutoff values were interpreted as hypothesis-generating rather than definitive clinical thresholds. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R. A two-sided p value < 0.05 was considered statistically significant [15]. Ethics This study was approved by the Institutional Ethics Committee (approval number: #2025-0091). The requirement for informed consent was waived, and an opt-out approach was employed, whereby information about the study was disclosed on the hospital website, allowing participants or their parents or guardians to decline participation. The study was conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan. Results A total of 65 patients underwent surgical resection of mediastinal masses during the study period. Of these, 56 underwent MIS, including 55 thoracoscopic procedures and one robot-assisted procedure, whereas nine underwent open thoracotomy (Table 1). No significant differences were observed between the two groups in terms of age, sex, height, weight, or tumor location. As expected based on approach selection, tumors were significantly larger in the open group, with greater maximum longitudinal diameter (81.9 [55.0–94.3] mm vs. 40.2 [27.0–61.5] mm, p = 0.003) and transverse diameter (80.1 [46.9–90.0] mm vs. 31.0 [20.8–45.2] mm, p = 0.001). The open group also showed significantly higher TTH (0.48 [0.38–0.62] vs. 0.27 [0.21–0.37], p = 0.01) and TTW (0.73 [0.40–1.03] vs. 0.38 [0.23–0.50], p = 0.004), as well as more frequent heterogeneous imaging appearance and positive IDRFs (p = 0.01 and p = 0.007, respectively). Median operative time was longer and blood loss was greater in the open group than in the MIS group (252 [180–271] min vs. 139 [86–215] min, p = 0.046; 5.8 [3.2–31.0] mL/kg vs. 0.1 [0–2.0] mL/kg, p = 0.0006), although these comparisons should be interpreted in the context of case selection. Table 1Comparison between the open thoracotomy and minimally invasive surgery groupsCharacteristicOpen surgeryMinimally invasive surgeryP Value(N = 9)(N = 56)Age, months (median, IQR)27 (3–181)45.5 (13–87)0.79Sex (male : female)5 : 431 : 251Height, cm (median, IQR)82 (64–156)96.2 (73.8–121.8)0.96Weight, kg (median, IQR)11.8 (7.5–45.4)14.2 (8.5–25.4)0.75Tumor locations Upper070.58 Anterior4160.43 Middle0110.34 Posterior5220.47Tumor size Maximum longitudinal diameter, mm (median, IQR)81.9 (55.0–94.3)40.2 (27.0–61.5) 0.003 Maximum transverse diameter, mm (median, IQR)80.1 (46.9–90)31 (20.8–45.2) 0.001 TTH (median, IQR)0.48 (0.38–0.62)0.27 (0.21–0.37) 0.01 TTW (median, IQR)0.73 (0.40–1.03)0.38 (0.23–0.50) 0.004 Tumor appearance on imaging Solid3280.48 Cystic1200.25 Heterogeneous58 0.01 IDRFs positive611 0.007 Preoperative chemotherapy0.135 Yes37 No649Operative procedure Total resection (one MIS case was robot-assisted)847 (1)1 Subtotal resection191 Operation time, min (median, IQR)252 (180–271)139 (86–215) 0.046 Blood loss, mL/kg (median, IQR)5.8 (3.2–31)0.1 (0–2) 0.0006 Pathological type Germ cell tumor68 0.002 Neuroblastoma151 Ganglioneuroblastoma051 Ganglioneuroma0100.33 Bronchogenic cyst0150.1 Lymphatic malformation051 Other (malignant)2 (2)8 (3)0.62Complication Vein injury041 Main bronchial injury011 Open conversion-1- Horner syndrome120.37 Chylothorax011Patient characteristics, tumor location, tumor size, imaging findings, surgical approach, histopathological diagnosis, and perioperative complications are summarizedIQR Interquartile range, IDRFs Image-defined risk factors, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio. One patient with injury to a branch of the brachiocephalic vein required conversion to open thoracotomy and subsequently developed postoperative chylothorax Table 1Comparison between the open thoracotomy and minimally invasive surgery groupsCharacteristicOpen surgeryMinimally invasive surgeryP Value(N = 9)(N = 56)Age, months (median, IQR)27 (3–181)45.5 (13–87)0.79Sex (male : female)5 : 431 : 251Height, cm (median, IQR)82 (64–156)96.2 (73.8–121.8)0.96Weight, kg (median, IQR)11.8 (7.5–45.4)14.2 (8.5–25.4)0.75Tumor locations Upper070.58 Anterior4160.43 Middle0110.34 Posterior5220.47Tumor size Maximum longitudinal diameter, mm (median, IQR)81.9 (55.0–94.3)40.2 (27.0–61.5) 0.003 Maximum transverse diameter, mm (median, IQR)80.1 (46.9–90)31 (20.8–45.2) 0.001 TTH (median, IQR)0.48 (0.38–0.62)0.27 (0.21–0.37) 0.01 TTW (median, IQR)0.73 (0.40–1.03)0.38 (0.23–0.50) 0.004 Tumor appearance on imaging Solid3280.48 Cystic1200.25 Heterogeneous58 0.01 IDRFs positive611 0.007 Preoperative chemotherapy0.135 Yes37 No649Operative procedure Total resection (one MIS case was robot-assisted)847 (1)1 Subtotal resection191 Operation time, min (median, IQR)252 (180–271)139 (86–215) 0.046 Blood loss, mL/kg (median, IQR)5.8 (3.2–31)0.1 (0–2) 0.0006 Pathological type Germ cell tumor68 0.002 Neuroblastoma151 Ganglioneuroblastoma051 Ganglioneuroma0100.33 Bronchogenic cyst0150.1 Lymphatic malformation051 Other (malignant)2 (2)8 (3)0.62Complication Vein injury041 Main bronchial injury011 Open conversion-1- Horner syndrome120.37 Chylothorax011Patient characteristics, tumor location, tumor size, imaging findings, surgical approach, histopathological diagnosis, and perioperative complications are summarizedIQR Interquartile range, IDRFs Image-defined risk factors, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio. One patient with injury to a branch of the brachiocephalic vein required conversion to open thoracotomy and subsequently developed postoperative chylothorax Comparison between the open thoracotomy and minimally invasive surgery groups Patient characteristics, tumor location, tumor size, imaging findings, surgical approach, histopathological diagnosis, and perioperative complications are summarized IQR Interquartile range, IDRFs Image-defined risk factors, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio. One patient with injury to a branch of the brachiocephalic vein required conversion to open thoracotomy and subsequently developed postoperative chylothorax Among the 17 patients with positive IDRFs, six underwent open thoracotomy (tracheal compression in five and vascular involvement in one) and 11 underwent MIS (tracheal compression in four, vascular involvement in four, both tracheal compression and vascular involvement in one, and costovertebral involvement in one). Of the 10 patients with tracheal compression, five underwent open thoracotomy because of unstable respiratory status. Intraoperative complications occurred in five of the 56 patients undergoing MIS (Table 2). Four patients developed major bleeding due to injury to branches of the brachiocephalic vein. These lesions were located in the anterior mediastinum in three patients and in the superior mediastinum in one patient; histopathological diagnoses were lymphatic malformation in two cases and mature teratoma and ganglioneuroblastoma in one case each. One patient with a lymphatic malformation required conversion to open thoracotomy due to uncontrollable bleeding and subsequently developed postoperative chylothorax. Three of the four bleeding cases were positive for IDRFs, with preoperative evidence of vascular encasement. In addition, one patient with a bronchogenic cyst in the middle mediastinum sustained a main bronchial injury requiring intraoperative repair. Horner syndrome occurred in three patients overall, including one in the open group and two in the MIS group; all three patients had neuroblastoma. Table 2Exploratory comparison of preselected imaging variables associated with intraoperative complications in patients undergoing minimally invasive surgeryNo intraoperative complicationIntraoperative complicationsP Value(N = 51)(N = 5)TTH (median, IQR)0.26 (0.21–0.33)0.51 (0.50–0.55) 0.004 TTW (median, IQR)0.37 (0.21–0.47)0.67 (0.64–0.71) 0.005 IDRFs positive7/514/5 0.004 Tracheal compression31 Vascular involvement32 Both tracheal compression and vascular involvement01 Costovertebral involvement10Tumor locations Upper610.5 Anterior1330.135 Middle1011 Posterior2200.14Preoperative chemotherapy701Pathological type Germ cell tumor710.55 Neuroblastoma501 Ganglioneuroblastoma410.385 Ganglioneuroma1000.57 Bronchogenic cyst1411 Lymphatic malformation320.058 Other (malignant)8 (3)01Clinicopathological and imaging characteristics are compared between patients with and without intraoperative complications in the MIS cohort. Continuous variables are presented as median (interquartile range) and compared using the Mann–Whitney U test, and categorical variables are presented as numbers and were compared using Fisher’s exact testIQR Interquartile range, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio, IDRFs Image-defined risk factors Table 2Exploratory comparison of preselected imaging variables associated with intraoperative complications in patients undergoing minimally invasive surgeryNo intraoperative complicationIntraoperative complicationsP Value(N = 51)(N = 5)TTH (median, IQR)0.26 (0.21–0.33)0.51 (0.50–0.55) 0.004 TTW (median, IQR)0.37 (0.21–0.47)0.67 (0.64–0.71) 0.005 IDRFs positive7/514/5 0.004 Tracheal compression31 Vascular involvement32 Both tracheal compression and vascular involvement01 Costovertebral involvement10Tumor locations Upper610.5 Anterior1330.135 Middle1011 Posterior2200.14Preoperative chemotherapy701Pathological type Germ cell tumor710.55 Neuroblastoma501 Ganglioneuroblastoma410.385 Ganglioneuroma1000.57 Bronchogenic cyst1411 Lymphatic malformation320.058 Other (malignant)8 (3)01Clinicopathological and imaging characteristics are compared between patients with and without intraoperative complications in the MIS cohort. Continuous variables are presented as median (interquartile range) and compared using the Mann–Whitney U test, and categorical variables are presented as numbers and were compared using Fisher’s exact testIQR Interquartile range, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio, IDRFs Image-defined risk factors Exploratory comparison of preselected imaging variables associated with intraoperative complications in patients undergoing minimally invasive surgery Clinicopathological and imaging characteristics are compared between patients with and without intraoperative complications in the MIS cohort. Continuous variables are presented as median (interquartile range) and compared using the Mann–Whitney U test, and categorical variables are presented as numbers and were compared using Fisher’s exact test IQR Interquartile range, TTH Tumor-to-thoracic height ratio, TTW Tumor-to-thoracic width ratio, IDRFs Image-defined risk factors Within the MIS cohort, both TTH and TTW were significantly higher in patients with intraoperative complications than in those without. Median TTH was 0.51 [0.50–0.55] in patients with complications and 0.26 [0.21–0.33] in those without (p = 0.004). Median TTW was 0.67 [0.64–0.71] and 0.37 [0.21–0.47], respectively (p = 0.005). Positive IDRFs were also more frequent in patients with complications than in those without (4/5 vs. 7/51, p = 0.004; Table 2). No patient with intraoperative complications had received preoperative chemotherapy. Bold values indicate statistically significant differences (P < 0.05). Exploratory ROC analysis demonstrated good discriminatory performance for both indices (Figs. 2 and 3). For TTW, the AUC was 0.882 (95% confidence interval [CI], 0.685–1.000), and the exploratory cutoff value was 0.64, with a sensitivity of 0.80 and a specificity of 0.98. For TTH, the AUC was 0.894 (95% CI 0.803–0.985), and the exploratory cutoff value was 0.356, with a sensitivity of 1.00 and a specificity of 0.78. These cutoff values should be interpreted cautiously because of the limited number of events. Fig. 2Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery The area under the curve (AUC) was 0.894 (95% confidence interval, 0.803–0.985). The exploratory cutoff value was 0.356, yielding a sensitivity of 1.00 and a specificity of 0.78 Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery The area under the curve (AUC) was 0.894 (95% confidence interval, 0.803–0.985). The exploratory cutoff value was 0.356, yielding a sensitivity of 1.00 and a specificity of 0.78 The distribution of TTH and TTW according to surgical approach and the presence or absence of intraoperative complications is shown in Fig. 4. Cases requiring open thoracotomy and MIS cases with intraoperative complications tended to cluster in the higher TTH and TTW ranges. Fig. 3Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery. The area under the curve (AUC) was 0.894 (95% confidence interval, 0.803–0.985). The exploratory cutoff value was 0.356, yielding a sensitivity of 1.00 and a specificity of 0.78 Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery Receiver operating characteristic (ROC) curve for the tumor-to-thoracic height ratio (TTH) for intraoperative complications in patients undergoing minimally invasive surgery. The area under the curve (AUC) was 0.894 (95% confidence interval, 0.803–0.985). The exploratory cutoff value was 0.356, yielding a sensitivity of 1.00 and a specificity of 0.78 Discussion In this study, both TTH and TTW were significantly associated with intraoperative complications in patients undergoing MIS for mediastinal masses. Exploratory ROC analysis suggested that both indices demonstrated reasonable discriminatory performance, with TTH showing higher sensitivity and TTW higher specificity. These findings suggest that simple tumor-to-thoracic cavity ratios may provide useful adjunctive information for estimating technical difficulty before pediatric MIS. Selection of the surgical approach for pediatric mediastinal masses remains challenging. In adults, absolute tumor diameter is often used as a practical indicator of thoracoscopic feasibility [7–9]. However, in children, thoracic dimensions vary substantially according to age and body size [10, 11], and the same tumor diameter may imply markedly different operative working conditions depending on the size of the hemithorax. For this reason, relative indices that reflect the relationship between tumor size and thoracic cavity size may be more informative than absolute measurements alone in pediatric patients. Fig. 4Distribution of tumor-to-thoracic height ratio (TTH) and tumor-to-thoracic width ratio (TTW) according to surgical approach and intraoperative complications. Scatter plot showing the distribution of TTH and TTW according to surgical approach and intraoperative complications. Closed circles indicate patients who underwent open thoracotomy, open triangles indicate minimally invasive surgery (MIS) cases without complications, and crosses indicate MIS cases with intraoperative complications Distribution of tumor-to-thoracic height ratio (TTH) and tumor-to-thoracic width ratio (TTW) according to surgical approach and intraoperative complications. Scatter plot showing the distribution of TTH and TTW according to surgical approach and intraoperative complications. Closed circles indicate patients who underwent open thoracotomy, open triangles indicate minimally invasive surgery (MIS) cases without complications, and crosses indicate MIS cases with intraoperative complications An important feature of the present study is the assessment of TTW in addition to TTH. Whereas TTH reflects craniocaudal tumor occupancy, TTW may better capture the extent to which the lesion narrows the transverse operative workspace and interferes with instrument handling. This may explain why TTW showed high specificity for intraoperative complications, whereas TTH showed higher sensitivity. Rather than representing competing parameters, these indices may reflect complementary aspects of technical difficulty during MIS. Simultaneously, the findings indicate that geometric indices alone are insufficient for surgical decision-making. Most intraoperative complications in the MIS cohort were related to injury of branches of the brachiocephalic vein, and three of the four bleeding cases had positive anatomical risk features on preoperative imaging, including vascular encasement. In addition, one patient with a bronchogenic cyst sustained a bronchial injury. These findings suggest that mass size relative to thoracic dimensions should be interpreted in conjunction with anatomical risk features, particularly vascular and airway involvement. Accordingly, TTH and TTW should be regarded as adjunctive markers that complement, rather than replace, preoperative anatomical assessment. While IDRFs provide qualitative information regarding the involvement of vital structures (e.g., vessel encasement), TTH and TTW offer a quantitative and spatial assessment of the available intrathoracic working space. The high specificity of TTW observed in this study suggests that transverse mass occupancy may be a critical limiting factor for instrument maneuverability in the narrow pediatric thorax. Notably, one patient who experienced an intraoperative complication was negative for IDRFs but had a TTH value above the exploratory cutoff, suggesting that technical difficulty related to limited intrathoracic working space may not be fully captured by anatomical risk assessment alone. Therefore, integrating both anatomical (IDRFs) and geometric (TTH/TTW) assessments may allow for more comprehensive risk stratification. The nature of surgical complications may also vary according to lesion characteristics. In the present cohort, vascular injuries occurred in lesions located near major mediastinal vessels, whereas airway injury occurred in a bronchogenic cyst closely associated with the tracheomembranous wall, requiring dissection adjacent to the airway. These findings suggest that lesion-specific anatomical relationships, in addition to lesion size and available working space, may influence the type of complication encountered during MIS. However, because complication events were infrequent and distributed across several histopathological subtypes, disease-specific analyses were not feasible in the present study. The influence of lesion location should be considered when interpreting TTH and TTW. In the present study, three of the five intraoperative complications occurred in anterior mediastinal lesions, while the remaining complications occurred in superior and middle mediastinal lesions. These regions contain critical structures such as branches of the brachiocephalic vein, the trachea, and the sympathetic chain, which may increase operative complexity independently of lesion size. Accordingly, anatomical location may contribute to operative risk, and the predictive performance of TTH and TTW may vary by mediastinal compartment. In this study, TTH and TTW were calculated relative to the operative hemithorax, reflecting the surgical workspace encountered during MIS. However, some anterior mediastinal lesions are located near the midline and may extend across both thoracic cavities. Whether hemithorax-based measurements are fully applicable to such lesions warrants further investigation. Therefore, future studies are needed to evaluate the applicability of these indices according to mediastinal compartment and lesion configuration. The distribution of cases on the TTH-TTW plot is also clinically informative (Fig. 4). Patients who underwent open thoracotomy and those who experienced intraoperative complications during MIS tended to cluster in the higher TTH and TTW ranges. This pattern supports the concept that these indices reflect overall technical difficulty. However, they should not be interpreted as defining absolute indications or contraindications for MIS. Instead, they may serve as practical reference points within a broader decision-making process that also considers tumor location, suspected pathology, vascular or airway involvement, and patient respiratory status. This study has several limitations. First, it was a retrospective single-center study, and selection of the surgical approach was not standardized, resulting in inherent selection bias. At the institution, open thoracotomy was preferentially selected for cases in which adequate working space for MIS was not anticipated or when significant anatomical risk or respiratory instability was present. Therefore, comparisons between the open and MIS groups should be interpreted as descriptive and are not suitable for causal inference regarding the superiority of one approach over the other. Second, the number of intraoperative complication events in the MIS cohort was small, which precluded formal multivariable analysis to adjust for potential confounders such as histopathology or surgical era. Consequently, the AUC and exploratory cutoff values derived from the ROC analysis should be regarded as hypothesis-generating rather than definitive clinical thresholds. These parameters require further validation in larger, multi-institutional prospective cohorts. Third, the study period was long, and changes in surgical experience, perioperative management, and device availability may have influenced outcomes. In addition, evolving treatment strategies, such as the use of medical therapy (e.g., sirolimus) for lymphatic malformations, may have influenced surgical indications and case selection, as some lesions included in this study would currently be managed non-surgically. Fourth, mediastinal masses with different histopathological backgrounds were analyzed together, and the relevance of anatomical risk features may differ across disease entities. Finally, measurement reproducibility was not formally assessed, and further validation will require evaluation of interobserver agreement and external generalizability. Conclusion TTH and TTW were associated with intraoperative complications in children undergoing minimally invasive resection of mediastinal masses. These simple imaging-based indices may provide preliminary insights for preoperative risk stratification, complementing anatomical risk features such as vascular or airway involvement. Larger multicenter studies are needed to validate these findings before broader clinical application can be recommended.