Development and validation of machine learning models and nomograms for predicting the surgical difficulty of laparoscopic resection in rectal cancer.

BACKGROUND: The objective of this study is to develop and validate a machine learning (ML) prediction model for the assessment of laparoscopic total mesorectal excision (LaTME) surgery difficulty, as well as to identify independent risk factors that influence surgical difficulty. Establishing a nomogram aims to assist clinical practitioners in formulating more effective surgical plans before the procedure. METHODS: This study included 186 patients with rectal cancer who underwent LaTME from January 2018 to December 2020. They were divided into a training cohort (n = 131) versus a validation cohort (n = 55). The difficulty of LaTME was defined based on Escal's et al. scoring criteria with modifications. We utilized Lasso regression to screen the preoperative clinical characteristic variables and intraoperative information most relevant to surgical difficulty for the development and validation of four ML models: logistic regression (LR), support vector machine (SVM), random forest (RF), and decision tree (DT). The performance of the model was assessed based on the area under the receiver operating characteristic curve(AUC), sensitivity, specificity, and accuracy. Logistic regression-based column-line plots were created to visualize the predictive model. Consistency statistics (C-statistic) and calibration curves were used to discriminate and calibrate the nomogram, respectively. RESULTS: In the validation cohort, all four ML models demonstrate good performance: SVM AUC = 0.987, RF AUC = 0.953, LR AUC = 0.950, and DT AUC = 0.904. To enhance visual evaluation, a logistic regression-based nomogram has been established. Predictive factors included in the nomogram are body mass index (BMI), distance between the tumor to the dentate line ≤ 10 cm, radiodensity of visceral adipose tissue (VAT), area of subcutaneous adipose tissue (SAT), tumor diameter >3 cm, and comorbid hypertension. CONCLUSION: In this study, four ML models based on intraoperative and preoperative risk factors and a nomogram based on logistic regression may be of help to surgeons in evaluating the surgical difficulty before operation and adopting appropriate responses and surgical protocols.

Simple pelvimetry predicts the pelvic manipulation time in robot-assisted low and ultra-low anterior resection for rectal cancer.

PURPOSE: This study explored the difficulty factors in robot-assisted low and ultra-low anterior resection, focusing on simple measurements of the pelvic anatomy. METHODS: This was a retrospective analysis of the clinical data of 61 patients who underwent robot-assisted low and ultra-low anterior resection for rectal cancer between October 2018 and April 2023. The relationship between the operative time in the pelvic phase and clinicopathological data, especially pelvic anatomical parameters measured on X-ray and computed tomography (CT), was evaluated. The operative time in the pelvic phase was defined as the time between mobilization from the sacral promontory and rectal resection. RESULTS: Robot-assisted low and ultra-low anterior resections were performed in 32 and 29 patients, respectively. The median operative time in the pelvic phase was 126 (range, 31-332) min. A multiple linear regression analysis showed that a short distance from the anal verge to the lower edge of the cancer, a narrow area comprising the iliopectineal line, short anteroposterior and transverse pelvic diameters, and a small angle of the pelvic mesorectum were associated with a prolonged operative time in the pelvic phase. CONCLUSION: Simple pelvic anatomical measurements using abdominal radiography and CT may predict the pelvic manipulation time in robot-assisted surgery for rectal cancer.

Effect of thoracic cage width on surgery time and postoperative outcome in minimally invasive esophagectomy.

INTRODUCTION: Minimally invasive esophagectomy (MIE) for esophageal cancer is a complex procedure that reduces postoperative morbidity in comparison to open approach. In this study, thoracic cage width as a factor to predict surgical difficulty in MIE was evaluated. METHODS: All patients of our institution receiving either total MIE or robotic-assisted MIE (RAMIE) with intrathoracic anastomosis between February 2016 and April 2021 for esophageal cancer were included in this study. Right unilateral thoracic cage width on the level of vena azygos crossing the esophagus was measured by the horizontal distance between the esophagus and parietal pleura on preoperative computer tomography. Patients' data as well as operative and postoperative details were collected in a prospective database. Correlation between thoracic cage width with duration of the thoracic procedure and postoperative complication rates was analyzed. RESULTS: Overall, 313 patients were eligible for this study. Thoracic width on vena azygos level ranged from 85 to 149 mm with a mean of 116.5 mm. In univariate analysis, a small thoracic width significantly correlated with longer duration of the thoracic procedure (p = 0.014). In multivariate analysis, small thoracic width and neoadjuvant therapy were identified as independent factors for long duration of the thoracic procedure (p = 0.006). Regarding postoperative complications, thoracic cage width was a significant risk factor for occurrence of postoperative pneumonia in the multivariate analysis (p = 0.045). Dividing the cohort into two groups of patients with narrow (≤ 107 mm, 19.5%) and wide thoraces (≥ 108 mm, 80.5%), the thoracic procedure was significantly prolonged by 17 min (204 min vs. 221 min, p = 0.014). CONCLUSION: A small thoracic cage width is significantly correlated with longer operation time during thoracic phase of a MIE in Europe, which suggests increased surgical difficulty. Patients with small thoracic cage width may preferably be operated by MIE-experienced surgeons.

Identifying the preoperative factors predicting the surgical difficulty of robotic distal pancreatectomy.

BACKGROUND: Few studies have evaluated the preoperative factors predicting the surgical difficulty of robotic distal pancreatectomy (RDP). This study aims to explore such factors and provide guidance on the selection of suitable patients to aid surgeons lacking extensive experience in RDP. METHODS: A retrospective study was conducted on consecutive patients who underwent RDP to identify preoperative factors predicting surgical difficulty. High surgical difficulty was defined by both operation time and intraoperative estimated blood loss exceeding their median, or by conversion to laparotomy. RESULTS: A total of 161 patients were ultimately enrolled, including 51 patients with high levels of surgical difficulty. Multivariate analysis revealed that male sex [OR (95% CI): 4.07 (1.77,9.40), p = 0.001], body mass index (BMI) ≥ 25 kg/m2 OR (95% CI): 2.27 (1.03,5.00), p = 0.042], tumors located at the neck of the pancreas [OR (95% CI): 4.15 (1.49,11.56), p = 0.006] and splenic artery type B [OR (95% CI): 3.28 (1.09,9.91), p = 0.035] were independent risk factors for surgical difficulty. Regarding postoperative complications, high surgical difficulty was associated with the risk of overall complications and pancreatic fistula (grade B/C) (49.0% vs. 22.7%, p < 0.001; 39.2% vs. 19.1%, p = 0.006). CONCLUSION: Male sex, body mass index ≥ 25 kg/m2, tumor located at the neck of the pancreas and splenic artery type B are associated with a high RDP difficulty level. These factors can be used preoperatively to assess the difficulty level of surgery, to help surgeons choose patients suitable for them and ensure surgical safety.

The preoperative M2BPGi score predicts operative difficulty and the incidence of postoperative complications in laparoscopic liver resection.

BACKGROUND: Liver fibrosis or cirrhosis frequently makes parenchymal transection more difficult, but the difficulty score of laparoscopic liver resection (LLR), including the IWATE criteria, does not include a factor related to liver fibrosis. Therefore, this study aimed to evaluate M2BPGi as a predictor of the difficulty of parenchymal transection and the incidence of postoperative complications in LLR. METHODS: Data from 54 patients who underwent laparoscopic partial liver resection (LLR-P) and 24 patients who underwent laparoscopic anatomical liver resection between 2017 and 2019 in our institution were retrospectively analyzed. All cases were classified according to M2BPGi scores, and reserve liver function, intraoperative blood loss, and postoperative complications were compared among these groups. RESULTS: Sixteen cases (29.6%) were M2BPGi negative (cut-off index < 1.0), 25 cases (46.3%) were 1+ (1.0 ≤ cut-off index < 3.0), and 13 cases (24.1%) were 2+ (cut-off index ≥ 3.0). M2BPGi-positive cases had significantly worse hepatic reserve function (K-ICG: 0.16 vs 0.14 vs 0.08, p < 0.0001). Intraoperative bleeding was significantly greater in M2BPGi-positive cases [50 ml vs 150 ml vs 200 ml, M2BPGi (-) or (1+) vs M2BPGi (2+), p = 0.045]. Postoperative complications (Clavien-Dindo ≥ II) were significantly more frequent in M2BPGi-positive cases [0% vs 4% vs 33%, M2BPGi (-) or (1+) vs M2BPGi (2+), p = 0.001]. CONCLUSION: M2BPGi could predict surgical difficulty and complications in LLR-P. In particular, it might be better not to select M2BPGi (2+) cases as teaching cases because of the massive bleeding during parenchymal transection.

Perioperative factors influencing the difficulty of retroperitoneal laparoscopic adrenalectomy: a single-center retrospective study.

PURPOSE: Identifying patients in whom adrenalectomy may be more difficult can help with surgical decision-making. This study investigated the perioperative factors affecting the difficulty of retroperitoneal laparoscopic adrenalectomy (RLA). METHODS: Sixty-eight patients who underwent RLA at our hospital between December 1, 2020 and May 1, 2021 were included. The difficulty of RLA was assessed by operating time and intraoperative blood loss. We analyzed the relationship between surgical difficulty and patient sex, age, and body mass index, pathological type, tumor side, tumor size, distance from the lower pole of the adrenal tumor to the upper pole of the kidney (DAK), and distance from the lower pole of the adrenal tumor to the renal pedicle (DARP). RESULTS: Mean operating time was 105.38 ± 33.31 min and mean intraoperative blood loss was 32.28 ± 22.88 ml. Univariate linear regression analysis showed that age (P = 0.047), tumor size (P = 0.002), DAK (P = 0.002), and DARP (P < 0.001) were significantly correlated with a longer operating time. Univariate logistic regression analysis showed that DARP (P = 0.001), DAK (P = 0.001), tumor size (P = 0.002), and age (P = 0.033) were significantly correlated with a longer operating time. Multivariate logistic regression indicated that DARP (OR 5.341; 95% CI 1.704-16.739; P = 0.004), and tumor size (OR 4.433; 95% CI 1.434-13.709; P = 0.010) were independent predictors of operating time. CONCLUSION: Age, tumor size, DAK, and DARP were predictors of the difficulty of RLA. Older age, lower DARP and DAK, and a larger tumor size were associated with a longer operating time. DARP and tumor size were independent predictors of surgical difficulty.

A retrospective and prospective study to establish a preoperative difficulty predicting model for video-assisted thoracoscopic lobectomy and mediastinal lymph node dissection.

BACKGROUND: In previous studies, the difficulty of surgery has rarely been used as a research object. Our study aimed to develop a predictive model to enable preoperative prediction of the technical difficulty of video-assisted thoracoscopic lobectomy and mediastinal lymph node dissection using retrospective data and to validate our findings prospectively. METHODS: Collected data according to the designed data table and took the operation time as the outcome variable. A nomogram to predict the difficulty of surgery was established through Lasso logistic regression. The prospective datasets were analyzed and the outcome was the operation time. RESULTS: This retrospective study enrolled 351 patients and 85 patients were included in the prospective datasets. The variables in the retrospective research were selected by Lasso logistic regression (only used for modeling and not screening), and four significantly related influencing factors were obtained: FEV1/FVC (forced expiratory volume in the first second/forced vital capacity) (p < 0.001, OR, odds ratio = 0.89, 95% CI, confidence interval = 0.84-0.94), FEV1/pred FEV1 (forced expiratory volume in the first second/forced expiratory volume in the first second in predicted) (p = 0.076, OR = 0.98, 95% CI = 0.95-1.00), history of lung disease (p = 0.027, OR = 4.00, 95% CI = 1.27-15.64), and mediastinal lymph node enlargement or calcification (p < 0.001, OR = 9.78, 95% CI = 5.10-19.69). We used ROC (receiver operating characteristic) curves to evaluate the model. The training set AUC (area under curve) value was 0.877, the test set's AUC was 0.789, and the model had a good calibration curve. In a prospective study, the data obtained in the research cohort were brought into the model again for verification, and the AUC value was 0.772. CONCLUSION: Our retrospective study identified four preoperative variables that are correlated with a longer surgical time and can be presumed to reflect more difficult surgical procedures. Our prospective study verified that the variables in the prediction model (including prior lung disease, FEV1/pred FEV1, FEV1/FVC, mediastinal lymph node enlargement or calcification) were related to the difficulty.

Establishment of Thoracic Surgical Difficulty Assessment Scale based on Delphi method. / 基于德尔菲法构建胸外科手术难度评价量表.

OBJECTIVES: The difficulty of surgery, which is related to surgical safety, has only been mentioned as a subjective perception for a long time. There are few studies to quantitatively and systematically evaluate the difficulty of thoracic surgery. This study aims to establish a quantitative evaluation index system for thoracic surgical difficulty, and to evaluate its reliability and validity. METHODS: During the 2 national thoracic surgery academic conferences, the factors that may affect the difficulty of thoracic surgery were evaluated by the thoracic surgeons via semi open questionnaires, and then the evaluation item pool of thoracic surgery difficulty was established. The importance of each indicator in the evaluation item pool was graded by 2 rounds of Delphi method. The average score, full score rate and coefficient of variation of each index were calculated, and the composite index method was used to decide whether to delete the indicator.Finally, the difficulty evaluation scale of thoracic surgery was constructed. The surgical data of patients with thoracic tumors were collected. The scale was used to evaluate the difficulty of thoracic surgery for lung, esophageal, and mediastinal tumors. The reliability and validity of the scale were evaluated by the commonly used difficulty evaluation indexes: Operation time, intraoperative estimated blood loss, Visual Analog Scale (VAS), side injury rate, and blood transfusion rate as standards. RESULTS: A total of 230 questionnaires were distributed in the 2 rounds of survey, and 149 valid questionnaires were collected after eliminating duplicate questionnaires. Through 2 rounds of Delphi consultation with 20 experts, the difficulty evaluation indexes were scored and screened, and the difficulty evaluation scale of thoracic surgery was established. It included 5 main indexes (surgical decision-making, operation space, separation interface, reconstruction method, and surgical materials) and 16 secondary indexes [American Society of Anesthesiologists (ASA) classification, surgical trauma, operator experience, space size, space depth, space source, space adjacent, interface content, anatomical gap, visual field, interface size, reconstruction complexity, reconstruction scope, autologous materials, artificial biomaterials and instruments]. After weighting, the total score of Thoracic Surgery Difficulty Evaluation Scale was from 1 to 3. A Score at 1 standed for simplicity, and score at 3 standed for difficulty. Further data were collected for 127 cases of thoracic tumor surgery. The difficulty scores of surgery for lung, esophageal, and mediastinal tumor were 1.69±0.26, 1.86±0.18, and 1.56±0.31, respectively, and the Cronbach's α coefficients of the scale in 3 tumor surgeries were 0.993, 0.974, and 0.989, repectively, and the Spearman Brown coefficients were 0.996, 0.984, and 0.996, respectively. The Spearman correlation coefficients of operation difficulty score with operation time, estimated blood loss, and VAS were 0.360 and 0.634, 0.632 and 0.578, 0.696 and 0.875, respectively (all P<0.05). The incidence of postoperative complications in the difficult operation group (difficulty score >1.85) was higher than that in the non-difficult operation group (P=0.02). CONCLUSIONS: The quantitative Thoracic Surgical Difficulty Assessment Scale has been successfully established, which shows good reliability and validity in thoracic tumor surgery. The Thoracic Surgical Difficulty Assessment Scale has broad application prospects in reducing the difficulty of the surgery, controlling surgical complications, and training surgeons.

Analysis of economic impact of laparoscopic liver resection according to surgical difficulty.

INTRODUCTION: The advantages of laparoscopic liver resection (LLR) are well known, but their financial costs are poorly evaluated. The aim of this study was to analyze the economic impact of surgical difficulty on LLR costs, and to identify clinical factors that most affect global charges. METHODS: All patients who underwent LLR from 2014 to 2018 in a single French center were included. The IMM classification was used to stratify surgical difficulty, from group I through group III. The costing method was done combining top-down and bottom-up approaches. A multivariate analysis was performed in order to identify clinical factors that most affect global charges. RESULTS: Two hundred seventy patients were included (Group I: n = 136 (50%), Group II: n = 60 (22%), Group III: n = 74 (28%)). Total expenses significantly increased (p < 0.001) from Group I to Group III, but there was no difference regarding financial income (p = 0.133). Technical platform expenses significantly increased (p < 0.001) from Group I to Group III and represented the main expense among all costs with a total of 4 930 ± 2 601€. Among technical platform expenses, the anesthesia platform represented the main expense. In multivariate analysis, the four clinical factors that affected global charges in the whole study population were operating time (p < 0.001), length of stay (p < 0.001), admission in ICU (p < 0.001) and the occurrence of major complication (p < 0.05). An admission in ICU was the clinical factor that affected most global charges, as an ICU stay had a 39.1% increase effect on global charges in the whole study population. CONCLUSION: LLR is a cost-effective procedure. The more complex is the LLR, the higher is the hospital cost. An admission in ICU was the clinical factor that most affected global charges.

Establishment of a nomogram for predicting the surgical difficulty of anterior cervical spine surgery.

BACKGROUND: For a long time, surgical difficulty is mainly evaluated based on subjective perception rather than objective indexes. Moreover, the lack of systematic research regarding the evaluation of surgical difficulty potentially has a negative effect in this field. This study was aimed to evaluate the risk factors for the surgical difficulty of anterior cervical spine surgery (ACSS). METHODS: This was a retrospective cohort study totaling 291 consecutive patients underwent ACSS from 2012.3 to 2017.8. The surgical difficulty of ACSS was defined by operation time longer than 120 min or intraoperative blood loss equal to or greater than 200 ml. Evaluation of risk factors was performed by analyzing the patient's medical records and radiological parameters such as age, sex, BMI, number of operation levels, high signal intensity of spinal cord on T2-weighted images, ossified posterior longitudinal ligament (OPLL), sagittal and coronal cervical circumference, cervical length, spinal canal occupational ratio, coagulation function index and platelet count. RESULTS: Significant differences were reported between low-difficulty and high-difficulty ACSS groups in terms of age (p = 0.017), sex (p = 0.006), number of operation levels (p < 0.001), high signal intensity (p < 0.001), OPLL (p < 0.001) and spinal canal occupational ratio (p < 0.001). Multivariate logistic regression analysis revealed that number of operation levels (OR = 5.224, 95%CI = 2.125-12.843, p < 0.001), high signal intensity of spinal cord (OR = 4.994, 95%CI = 1.636-15.245, p = 0.005), OPLL (OR = 6.358, 95%CI = 1.932-20.931, p = 0.002) and the spinal canal occupational ratio > 0.45 (OR = 3.988, 95%CI = 1.343-11.840, p = 0.013) were independently associated with surgical difficulty in ACSS. A nomogram was established and ROC curve gave a 0.906 C-index. There was a good calibration curve for difficulty estimation. CONCLUSION: This study indicated that the operational level, OPLL, high signal intensity of spinal cord, and spinal canal occupational ratio were independently associated with surgical difficulty and a predictive nomogram can be established using the identified risk factors. Optimal performance was achieved for predicting surgical difficulty of ACSS based on preoperative factors.

Utility of thoracic cage width in assessing surgical difficulty of minimally invasive esophagectomy in left lateral decubitus position.

BACKGROUND: This study aimed to assess the surgical difficulty of minimally invasive esophagectomy in the left lateral decubitus position for patients with esophageal cancer from the perspective of short-term outcomes, including operation time, blood loss, and morbidity. MATERIALS AND METHODS: The initial 44 consecutive patients with esophageal cancer who underwent minimally invasive esophagectomy were statistically analyzed retrospectively. Thoracic cage area was measured from preoperative computed tomography as a factor affecting the surgical difficulty of minimally invasive esophagectomy, as well as other patient characteristics. Correlations with short-term outcomes including chest operation time, blood loss, and morbidity rate were then examined. RESULTS: In univariate analyses, smaller area of the upper thoracic cage width correlated with prolonged thoracic procedure time (p = 0.0119) and greater blood loss during thoracic procedures (p = 0.0283), but area of the lower thoracic cage showed no correlations. History of respiratory disease was associated with thoracic procedure time (p < 0.0001), but not blood loss. In multivariate analysis, small area of the upper thoracic cage was independently associated with prolonged thoracic procedure time (p = 0.0253). Small upper thoracic cage area was not directly correlated with morbidity rate, but prolonged thoracic procedure time was associated with increased blood loss (p < 0.0001) and morbidity rate (p = 0.0204). Empirical time reduction (p = 0.0065), but not blood loss, was associated with thoracic procedure time. However, area of the upper thoracic cage did not correlate with empirical case number. In multivariate analysis, area of the upper thoracic cage (p = 0.0317) and empirical case number (p = 0.0193) correlated independently with thoracic procedure time. CONCLUSION: A small area of the upper thoracic cage correlated significantly with prolonged thoracic procedure time and increased thoracic blood loss for minimally invasive esophagectomy in the left lateral decubitus position, suggesting the surgical difficulty of minimally invasive esophagectomy in the left lateral decubitus position.

Assessing surgical difficulty in locally advanced mid-low rectal cancer: the accuracy of two MRI-based predictive scores.

AIM: Predicting surgical difficulty is a critical factor in the management of locally advanced rectal cancer (LARC). This study evaluates the accuracy and external validity of a recently published morphometric score to predict surgical difficulty and additionally proposes a new score to identify preoperatively LARC patients with a high risk of having a difficult surgery. METHODS: This is a retrospective study based on the European MRI and Rectal Cancer Surgery (EuMaRCS) database, including patients with mid/low LARC who were treated with neoadjuvant chemoradiation therapy and laparoscopic total mesorectal excision (L-TME) with primary anastomosis. For all patients, pretreatment and restaging MRI were available. Surgical difficulty was graded as high and low based upon a composite outcome, including operative (e.g. duration of surgery) and postoperative variables (e.g. hospital stay). Score accuracy was assessed by estimating sensitivity, specificity and area under the receiver operating characteristic curve (AROC). RESULTS: In a total of 136 LARC patients, 17 (12.5%) were graded as high surgical difficulty. The previously published score (calculated on body mass index, intertuberous distance, mesorectal fat area, type of anastomosis) showed low predictive value (sensitivity 11.8%; specificity 92.4%; AROC 0.612). The new EuMaRCS score was developed using the following significant predictors of surgical difficulty: body mass index > 30, interspinous distance < 96.4 mm, ymrT stage ≥ T3b and male sex. It demonstrated high accuracy (AROC 0.802). CONCLUSION: The EuMaRCS score was found to be more sensitive and specific than the previous score in predicting surgical difficulty in LARC patients who are candidates for L-TME. However, this score has yet to be externally validated.

Mesorectal fat area as a useful predictor of the difficulty of robotic-assisted laparoscopic total mesorectal excision for rectal cancer.

BACKGROUND: Various predictors of the difficulty of total mesorectal excision for rectal cancer have been described. Although a bulky mesorectum was considered to pose technical difficulties in total mesorectal excision, no studies have evaluated the influence of mesorectum morphology on the difficulty of total mesorectal excision. Mesorectal fat area at the level of the tip of the ischial spines on magnetic resonance imaging was described as a parameter characterizing mesorectum morphology. This study aimed to evaluate the influence of clinical and anatomical factors, including mesorectal fat area, on the difficulty of total mesorectal excision for rectal cancer. METHODS: This study enrolled 98 patients who underwent robotic-assisted laparoscopic low anterior resection with total mesorectal excision for primary rectal cancer, performed by a single expert surgeon, between 2010 and 2015. Magnetic resonance imaging-based pelvimetry data were collected. Linear regression was performed to determine clinical and anatomical factors significantly associated with operative time of the pelvic phase, which was defined as the time interval from the start of rectal mobilization to the division of the rectum. RESULTS: The median operative time of the pelvic phase was 68 min (range 33-178 min). On univariate analysis, the following variables were significantly associated with longer operative time of the pelvic phase: male sex, larger tumor size, larger visceral fat area, larger mesorectal fat area, shorter pelvic outlet length, longer sacral length, shorter interspinous distance, larger pelvic inlet angle, and smaller angle between the lines connecting the coccyx to S3 and to the inferior middle aspect of the pubic symphysis. On multiple linear regression analysis, only larger mesorectal fat area remained significantly associated with longer operative time of the pelvic phase (p = 0.009). CONCLUSIONS: Mesorectal fat area may serve as a useful predictor of the difficulty of total mesorectal excision for rectal cancer.

Factors influencing difficulty of the thoracic procedure in minimally invasive esophagectomy.

BACKGROUND: Minimally invasive esophagectomy (MIE) is being increasingly performed worldwide. When performing MIE, we sometimes experienced difficulties due to a narrow upper mediastinum or a middle to lower thoracic esophagus hidden by the projection of the vertebral body. However, there were no reports regarding the influence of anatomical factors on the difficulty of performing MIE. The aim of this study was to evaluate whether anatomical factors could be related to the difficulty of the thoracic procedure in MIE. METHODS: We investigated 87 consecutive patients undergoing MIE for primary esophageal cancer between 2013 and 2015 and created novel indices to assess the upper mediastinal narrowness and vertebral body projection at middle thoracic part on preoperative computed tomography images. We assessed clinicopathological and anatomical factors and determined the factors influencing the thoracic procedural difficulty in MIE. The thoracic procedure duration was selected as the variable representing technical difficulty. RESULTS: The mean thoracic procedure duration was 280.2 ± 52.5 min. There were no significant correlations between the indices and patient factors such as age, sex, and body mass index. Meanwhile, there was a significant correlation between the upper mediastinal narrowness and the vertebral body projection (p < 0.01). Of the clinicopathological and anatomical factors, blood loss during the thoracic procedure, thoracic duct resection, and vertebral body projection independently were related to the prolonged thoracic procedure duration in multiple linear regression analysis (p = 0.01, 0.03, and <0.01, respectively). The other factors including upper mediastinal narrowness were not statistically significant. CONCLUSIONS: This is the first study to reveal the influence of anatomical factors on the difficulty of the thoracic procedure in MIE. The vertebral body projection at middle thoracic part appears to be a useful tool for predicting the thoracic procedural difficulty in MIE preoperatively.

A preoperatively predictive difficulty scoring system for laparoscopic spleen-preserving splenic hilar lymph node dissection for gastric cancer: experience from a large-scale single center.

OBJECTIVES: The purpose of this study was to evaluate the risk factors that influence the difficulty of performing laparoscopic spleen-preserving splenic hilar lymph node dissection (SHLND) for gastric cancer and to establish a simple and effective scoring system to predict the surgical difficulty preoperatively. METHODS: Between January 2011 and December 2013, we prospectively collected and retrospectively analyzed the medical records of 317 patients with upper- or middle-third gastric cancer who underwent laparoscopic spleen-preserving SHLND. Univariate and multivariate analyses were performed to determine the independent predictors for surgical difficulty based on the operation time during spleen-preserving SHLND. A logistic regression model was used to identify determinant variables and construct a predictive difficulty scoring system. RESULTS: Multivariate analysis showed that gender, body mass index (BMI), number of splenic lobar arteries (SLAs), and type of SLA were independently predictive factors of operation time. According to these factors, we developed a predictive surgical difficulty scoring system and the difficulty levels are divided into 0, 1, 2, 3, and 4 points or more. Based on the relative risk, we stratified the surgical difficulty into the following three divisions: 0 low difficulty, 1-2 intermediate difficulty, and 3 points or more high difficulty. The patients with long operation time accounted for 19.6, 43.6, and 90.9 % for the three groups, respectively (p < 0.001). The area under the receiver operating characteristic curve for the logistic regression model and the simplified difficulty scoring prediction model was 0.717 and 0.715, respectively. CONCLUSIONS: Based on four independent risk factors, including gender, BMI, number of SLAs, and type of SLA, we developed a simple and effective scoring system to predict the difficulty of laparoscopic spleen-preserving SHLND preoperatively. This novel scoring system might aid surgeons with different experience in performing operations at different levels of difficulty.

Validation of a surgical difficulty grading system in laparoscopic cholecystectomy for acute cholecystitis.

BACKGROUND: The surgical difficulty of laparoscopic cholecystectomy (LC) for acute cholecystitis varies from case to case, and appropriate intraoperative evaluation would help prevent bile duct injury (BDI). METHODS: We analyzed 178 patients who underwent LC for acute cholecystitis. Expert surgeons and trainees individually evaluated the surgical difficulty. The inter-rater agreement was analyzed using Conger's κ and Gwet's agreement coefficient (AC). Furthermore, we analyzed the predictive surgical difficulty item for performing subtotal cholecystectomy (STC). RESULTS: Regarding the inter-rater agreement between expert surgeons and trainees, 15 of the 17 surgical difficulty items had a Gwet's AC of 0.5 or higher, indicating "moderate" agreement or higher. Furthermore, the highest and total surgical difficulty scores were deemed "substantial" agreement. Scarring and dense fibrotic changes around the Calot's triangle area with easy bleeding with/without necrotic changes were predictive of whether STC should be performed. CONCLUSIONS: This surgical difficulty grading system is expected to be a tool that can be used by any surgeon with LC experience. STC should be performed to prevent BDI according to the changes around the Calot's triangle area.

Prediction of extraction difficulty for impacted maxillary third molars with deep learning approach.

OBJECTIVE: The aim of this study is to determine if a deep learning (DL) model can predict the surgical difficulty for impacted maxillary third molar tooth using panoramic images before surgery. MATERIALS AND METHODS: The dataset consists of 708 panoramic radiographs of the patients who applied to the Oral and Maxillofacial Surgery Clinic for various reasons. Each maxillary third molar difficulty was scored based on dept (V), angulation (H), relation with maxillary sinus (S), and relation with ramus (R) on panoramic images. The YoloV5x architecture was used to perform automatic segmentation and classification. To prevent re-testing of images, participate in the training, the data set was subdivided as: 80 % training, 10 % validation, and 10 % test group. RESULTS: Impacted Upper Third Molar Segmentation model showed best success on sensitivity, precision and F1 score with 0,9705, 0,9428 and 0,9565, respectively. S-model had a lesser sensitivity, precision and F1 score than the other models with 0,8974, 0,6194, 0,7329, respectively. CONCLUSION: The results showed that the proposed DL model could be effective for predicting the surgical difficulty of an impacted maxillary third molar tooth using panoramic radiographs and this approach might help as a decision support mechanism for the clinicians in peri­surgical period.

Interpretable machine learning model to predict surgical difficulty in laparoscopic resection for rectal cancer.

Background: Laparoscopic total mesorectal excision (LaTME) is standard surgical methods for rectal cancer, and LaTME operation is a challenging procedure. This study is intended to use machine learning to develop and validate prediction models for surgical difficulty of LaTME in patients with rectal cancer and compare these models' performance. Methods: We retrospectively collected the preoperative clinical and MRI pelvimetry parameter of rectal cancer patients who underwent laparoscopic total mesorectal resection from 2017 to 2022. The difficulty of LaTME was defined according to the scoring criteria reported by Escal. Patients were randomly divided into training group (80%) and test group (20%). We selected independent influencing features using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression method. Adopt synthetic minority oversampling technique (SMOTE) to alleviate the class imbalance problem. Six machine learning model were developed: light gradient boosting machine (LGBM); categorical boosting (CatBoost); extreme gradient boost (XGBoost), logistic regression (LR); random forests (RF); multilayer perceptron (MLP). The area under receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity and F1 score were used to evaluate the performance of the model. The Shapley Additive Explanations (SHAP) analysis provided interpretation for the best machine learning model. Further decision curve analysis (DCA) was used to evaluate the clinical manifestations of the model. Results: A total of 626 patients were included. LASSO regression analysis shows that tumor height, prognostic nutrition index (PNI), pelvic inlet, pelvic outlet, sacrococcygeal distance, mesorectal fat area and angle 5 (the angle between the apex of the sacral angle and the lower edge of the pubic bone) are the predictor variables of the machine learning model. In addition, the correlation heatmap shows that there is no significant correlation between these seven variables. When predicting the difficulty of LaTME surgery, the XGBoost model performed best among the six machine learning models (AUROC=0.855). Based on the decision curve analysis (DCA) results, the XGBoost model is also superior, and feature importance analysis shows that tumor height is the most important variable among the seven factors. Conclusions: This study developed an XGBoost model to predict the difficulty of LaTME surgery. This model can help clinicians quickly and accurately predict the difficulty of surgery and adopt individualized surgical methods.

Prediction of intraoperative surgical difficulty during laparoscopic cholecystectomy using drip infusion cholangiography with computed tomography.

BACKGROUND: Although findings from drip infusion cholangiography with computed tomography (DIC-CT) are useful in preoperative anatomic evaluation for laparoscopic cholecystectomy (LC), their relationship with intraoperative surgical difficulty based on the difficulty score (DS) proposed by Tokyo Guidelines 2018 is unclear. We examined this relationship. METHODS: Data were collected from 202 patients who underwent LC for benign gallbladder (GB) disease with preoperative DIC-CT in our department. DIC-CT findings were classified into GB-positive and GB-negative groups based on GB opacification, and clinical characteristics were compared. DS assessed only on findings from around Calot's triangle was considered "cDS", and patients were divided into cDS ≤2 and ≥3 groups. Preoperative data including DIC-CT findings were evaluated using multivariate analysis. RESULTS: DIC-CT findings showed 151 (74.8%) GB-positive and 51 (25.2%) GB-negative patients. Surgical outcomes were significantly better in the GB-positive versus GB-negative group for operation time (107 vs. 154 min, p < .001), blood loss (8 vs. 25 mL, p < .001), cDS (0.8 vs. 2.2, p < .001), and critical view of safety score (4.0 vs. 3.1, p < .001). cDS was ≤2 in 174 (86.1%) and ≥3 in 28 (13.9%) patients. By multivariate analysis, DIC-CT findings and alkaline phosphatase values were independent factors predicting intraoperative difficulty. CONCLUSION: DIC-CT findings are useful for predicting cDS in LC.

Preoperative systemic and local inflammation are independent risk factors for difficult laparoscopic cholecystectomy after percutaneous transhepatic gallbladder drainage.

Background: Laparoscopic cholecystectomy (LC) is required for acute cholecystitis patient with percutaneous transhepatic gallbladder drainage (PTGBD). However, it's unknown how to distinguishing the surgical difficulty for these patients. Methods: Data of patients who underwent LC after PTGBD between 2016 and 2022 were collected. Patients were categorized into difficult and non-difficult operations based on operative time, blood loss, and surgical conversion. Performance of prediction model was evaluated by ROC, calibration, and decision curves. Results: A total of 127 patients were analyzed, including 91 in non-difficult operation group and 36 in difficult operation group. Elevated CRP (P = 0.011), pericholecystic effusion (P < 0.001), and contact with stomach or duodenal (P = 0.015) were independent risk factors for difficult LC after PTGBD. A nomogram was developed according to these risk factors, and was well-calibrated and good at distinguishing difficult LC after PTGBD. Conclusion: Preoperative elevated systemic and local inflammation indictors are predictors for difficult LC after PTGBD.