Clinical, Anatomical, and Pathological Grading Score to Predict Technical Difficulty in Laparoscopic Splenectomy for Non-traumatic Diseases

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.

Development of a difficulty scoring system for laparoscopic pancreatoduodenectomy in the initial stage of the learning curve: a retrospective cohort study

BACKGROUND

It remains uncertain how surgeons can safely pass the learning curve of laparoscopic pancreatoduodenectomy (LPD) without potentially harming patients. We aimed to develop a difficulty scoring system (DSS) to select an appropriate patient for surgeons.

MATERIALS AND METHODS

A total of 773 elective pancreatoduodenectomy surgeries between July 2014 and December 2019, including 346 LPD and 427 open pancreatoduodenectomy cases, were included. A 10-level DSS for LPD was developed, and an additional 77 consecutive LPD surgeries which could provide information of the learning stage I of LPD externally validated its performance between December 2019 and December 2021.

RESULTS

The incidences of postoperative complications (Clavien-Dindo≥III) gradually decreased from the learning curve stage I-III (20.00, 10.94, 5.79%, P =0.008, respectively). The DSS consisted of the following independent risk factors: (1) tumor location, (2) vascular resection and reconstruction, (3) learning curve stage, (4) prognostic nutritional index, (5) tumor size, and (6) benign or malignant tumor. The weighted Cohen's κ statistic of concordance between the reviewer's and calculated difficulty score index was 0.873. The C -statistics of DSS for postoperative complication (Clavien-Dindo≥III) were 0.818 in the learning curve stage I. The patients with DSS<5 had lower postoperative complications (Clavien-Dindo≥III) than those with DSS≥5 (4.35-41.18%, P =0.004) in the training cohort and had a lower postoperative pancreatic fistula (19.23-57.14%, P =0.0352), delayed gastric emptying (19.23-71.43%, P =0.001), and bile leakage rate (0.00-21.43%, P =0.0368) in validation cohort in the learning curve stage I.

CONCLUSION

We developed and validated a difficulty score model for patient selection, which could facilitate the stepwise adoption of LPD for surgeons at different stages of the learning curve.

Impact of Underlying Disease and Preoperative Factors on Postoperative Outcomes After Laparoscopic Splenectomy: A Bicentric Retrospective Analysis

BACKGROUND

Although the laparoscopic approach is considered the gold standard for elective splenectomy, it is still debated whether the underlying disease affects postoperative outcomes. Given the importance of good patient selection in the early stages of the learning curve for laparoscopic splenectomy (LS), this study aimed to compare the postoperative outcomes following LS for malignant diseases and benign diseases (MDs and BDs).

MATERIALS AND METHODS

A retrospective review of patients who underwent LS was performed at 2 different institutions between January 2013 and September 2020. Patients were classified into 2 groups based on the underlying BDs or MDs, and the 30 days postoperative outcomes were compared. Risk factors for overall complications were determined using logistic regression analysis.

RESULTS

LS was performed for BDs and MDs in 51 (67%) and 25 (33%) patients, respectively. The overall morbidity rate and the intraoperative and postoperative complication rates were significantly higher in the MD group (P<0.05). In the univariate analysis, the underlying MD, age above 49.5 years, body mass index >24.9, the long axis of the spleen >16 cm, and spleen weight >600 g were significantly associated with increased postoperative morbidity.

CONCLUSION

In addition to the underlying disease, preoperative conditions may also affect the complication rates after LS. These findings may be helpful in patient selection, especially in the early stages of the learning curve for minimally invasive splenectomies.

Prognostic scoring system of laparoscopic splenectomy in children with benign hematological diseases, a retrospective cohort study

Purpose

laparoscopic splenectomy (LS) gained popularity during the last years among pediatric surgeons. Benign hematological diseases represent the main indications of that procedure. There are some score systems of difficulty of the technique. Most of them are for adult patients. The aim is to develop a prognostic score system of LS in children with benign hematological diseases.

Methods

LS was performed in all children with benign hematological diseases. The children were operated in lateral decubitus position. The control of pedicle was achieved using bipolar sealing devices, polymer clips or endoscopic linear stablers. Preoperative, operative, and postoperative data were recorded and analyzed. Children were divided into two groups A and B according to the onset of complications.

Results

137 successive children were operated. The mean age of patients was 8.7 (4–15) years. The main indications of LS were thalassemia, spherocytosis, idiopathic thrombocytopenic purpura (ITP), and sickle cell anemia (SCA). The mean length of splenic axis in thalassemia and spherocytosis cases was15.6 ± 2.3 cm. The mean amount of estimated blood loss (EABL) was 149 ± 43.2 ml in group A while it was 185 ± 44.4 ml in group B. The mean operative time in group A was 75.5 ± 22.5 min while it was 89.2 ± 20.6 min in group B.

Conclusion

the development of prognostic score of LS in children with benign hematological diseases can predict children who are more susceptible for complications. It helps to minimize these complications and improve the outcome of LS.

•

Laparoscopic splenectomy in children gained popularity during last years.

•

The outcome of laparoscopic surgery is superior to conventional surgery.

•

Benign hematological diseases represent the main indication of splenectomy in children.

•

There is a need to develop a prognostic score of laparoscopic splenectomy in pediatric age group.

•

Method of pedicle control represents one of main prognostic factors.

The usefulness of modified splenic hilum hanging maneuver in laparoscopic splenectomy, especially for patients with huge spleen: a case-control study with propensity score matching

BACKGROUND

Although Laparoscopic splenectomy (LS) have been proven to the standard operation for removal of spleen, the rate of conversion to open surgery is still higher than those of other laparoscopic surgeries, especially for huge spleen. In order to reduce the rate of conversion to open surgery, we had developed LS using modified splenic hilum hanging (MSHH) maneuver: the splenic pedicle was transected en bloc using a surgical stapler after hanging splenic hilum with an atraumatic penrose drain tube.

METHODS

Between January 2005 and December 2019, we retrospectively assessed 94 patients who underwent LS. MSHH maneuver was performed in 37 patients (39.4%). We compared the intra- and postoperative outcomes between patients with or without MSHH maneuver. To adjust for differences in preoperative characteristics and blood examination, propensity score matching was used at a 1:1 ratio, resulting in a comparison of 29 patients per group. Predictive factors of conversion from LS to open surgery were elucidated using the uni- and multi-variate analyses.

RESULTS

After the propensity score matching, blood loss (268 ml vs. 50 ml), the rate of conversion to open surgery (27.6% vs. 0%), and postoperative hospital stays (15 days vs. 10 days) were significantly decreased in patients with MSHH maneuver, respectively. Among 94 patients, 19 patients (20.2%) underwent conversion to open surgery. In multivariate analysis, spleen volume (SV) and LS without MSHH maneuver were independent predictive factors of conversion to open surgery, respectively. Additionally, cut-off value of SV for conversion to open surgery was 802 ml (sensitivity: 0.684, specificity: 0.827, p < 0.001).

CONCLUSIONS

LS using MSHH maneuver seems to be useful surgical technique to improve intraoperative outcomes and reduce the rate of conversion from LS to open surgery resulting in shorten postoperative hospital stay.

Laparoscopic Resection of a Left Upper Quadrant Mass Leading to a Surprise Diagnosis

Background. Resplenectomy is most commonly done for the treatment of recurrent idiopathic thrombocytopenic purpura (ITP) refractory to medical management due to the regrowth of a missed accessory spleen. Case Report. A 66-year-old male had undergone open splenectomy for traumatic rupture 40 years ago. He presented with a leiomyosarcoma of his leg, which was surgically removed. When he developed metastatic disease, chemotherapy was started. He developed left upper quadrant pain, and on CT scan, a 5 cm mass compatible with a sarcoma was found between the tail of the pancreas and the left adrenal gland. During laparoscopy, dense adhesion of the omentum to the abdominal wall and the stomach from his previous splenectomy was divided. The lesser sac was opened through the gastrocolic ligament, and the splenic flexure was taken down. Superior and dorsal to the tail of the pancreas next to the left adrenal gland, the mass was identified and carefully dissected out. The vascular pedicle, which originated from a side branch of the splenic vessels at the tail of the pancreas, was stapled. The gastric fundus showed multiple nodules, and therefore, a modified sleeve gastrectomy was done; also, a 2 cm nodule in segment 5 of the liver and an omental nodule were removed. The tumors and gastrectomy specimen were placed in an endobag and removed through a periumbilical mini-incision. The patient recovered without any complications from the procedure and his LUQ pain resolved. Pathology revealed no sarcoma metastases but accessory spleens in all specimens. Discussion. Splenosis with multiple implants within the abdomen after splenectomy for trauma is a rare condition. In our patient, this seems to have been triggered by chemotherapy for his sarcoma resulting in extramedullary hemopoiesis. Laparoscopic removal of accessory spleens can be safely done.

A novel preoperative scoring system to predict technical difficulty in laparoscopic splenectomy for non-traumatic diseases

BACKGROUND

Laparoscopic splenectomy (LS) has been proven to be a safe and advantageous procedure. To ensure that resections of appropriate difficulty are selected, an objective preoperative grading of difficulty is required. We aimed to develop a predictive difficulty grading of LS based on intraoperative complications.

METHODS

A total of 272 non-traumatic patients who underwent LS were identified from a regional medical center. Patients were randomized into a training cohort (n = 222) and a validation cohort (n = 50). Data on demographics, medical and surgical history, operative and pathological characteristics, and postoperative outcome details were collected. Univariate and multivariate analyses of risk factors for intraoperative complications were performed to develop a difficulty scoring system. The Spearman correlation coefficient was used to evaluate the relationship between the difficulty grading score and intraoperative outcomes. Receiver operating characteristic (ROC) curve was used to evaluate the discriminatory power of this scoring system.

RESULTS

Three preoperative factors (spleen weight, esophagogastric varices, and INR) had a significant effect on operative time, bleeding, and conversion to open surgery. We created a difficulty grading score with three levels of difficulty: low (≤ 4 points), medium (5-6 points), and high (≥ 7 points), based on the three preoperative parameters. The correlation was highly significant (P < 0.01) according to Spearman's correlation. The area under the ROC curve was 0.695 (95% CI 0.630-0.755). The external validation showed significant correlations with the present model, with an AUC of 0.725 (95% CI 0.580-0.842). The comparison between our difficulty score and the previous grading system in the 272-patient cohort presented a significant difference in the AUC (0.701, 95% CI 0.643-0.755 vs. 0.644, 95% CI 0.584-0.701, P = 0.0452).

CONCLUSION

The present difficulty scoring system, based on preoperative factors, has good performance in predicting the risk of intraoperative complications of LS and could be helpful for enabling appropriate case selection with respect to the current experience of a surgeon.

Laparoscopic Splenectomy Versus Open Splenectomy In Massive and Giant Spleens: Should we Update the 2008 EAES Guidelines?

The objective of this study was to derive some useful parameters to define the feasibility of laparoscopic splenectomy (LS) in massive [spleen longitudinal diameter (SLD)>20 cm] and giant spleens (SLD>25 cm). Between December 1996 and May 2017, 175 patients underwent an elective splenectomy. A laparoscopic approach was used in 133 (76%) patients. Massive spleens were treated in 65 (37.1%) patients, of which 24 were treated laparoscopically. In this subset of massive spleens, the results of laparoscopic splenectomy in massive spleens (LSM) and open splenectomy in massive spleens (OSM) were compared. The clinical outcome of a subgroup of patients with giant spleens was also analyzed. The LSM group resulted in significant longer operative times (143±31 vs. 112±40 min; P=0.001), less blood loss (278±302 vs. 575±583 mL; P=0.007), and shorter hospital stay (6±3 vs. 9±4 d; P=0.004). No conversions were experienced in the LSM group, and the morbidity rate was similar in both the LSM and OSM groups (16.6% vs. 20%; P=0.75). When considering the subset of 9 LSM patients and 26 OSM patients with giant spleens, the same favorable tendency of the laparoscopic group as regards surgical conversion, blood loss, and hospital stay was maintained. The laparoscopic approach can be successfully proposed in the presence of massive splenomegaly also after a careful preoperative evaluation of the expected abdominal "working space." In experienced hands, LS is safe, feasible, and associated with better outcomes than open splenectomy for the treatment of massive and giant spleen, with a maximum SLD limit of 31 cm.

Prediction of Technical Difficulties in Laparoscopic Splenectomy and Analysis of Risk Factors for Postoperative Complications in 468 Cases

Prediction of intraoperative difficulties may be helpful in planning surgery; however, few studies explored this issue in laparoscopic splenectomy (LS). We performed retrospective analysis of consecutive 468 patients undergoing LS from 1998 to 2017 (295 women; median age 47 years). The patients were divided into difficult LS and control groups. The inclusion criteria for difficult LS were operative time ≥mean + 2SD; intraoperative blood loss ≥500 mL, intraoperative adverse events (IAE), conversion. Primary outcomes were risk factors for difficult splenectomy and secondary outcomes for perioperative morbidity. Fifty-six patients were included in the difficult LS group (12%). Spleens ≥19 cm and higher participation of younger surgeons in consecutive years were predictive for difficult splenectomy. Age ≥53 years and diagnosis other than idiopathic thrombocytopenic purpura (ITP) were independent risk factors of spleen ≥19 cm. The perioperative morbidity was 8.33%; its OR was increased only by blood loss and IAEs. Only blood loss significantly increased serious morbidity. Male sex, spleens ≥19 cm, and IAEs were independent risk factors for intraoperative hemorrhage. Spleen length ≥19 cm was a risk factor for difficult LS and intraoperative hemorrhage. Diagnoses other than ITP in patients aged ≥53 years with ≥19 cm spleens are predictive for intraoperative difficulties and perioperative complications.

Difficulty scoring system in laparoscopic distal pancreatectomy

BACKGROUND

Several factors affect the level of difficulty of laparoscopic distal pancreatectomy (LDP). The purpose of this study was to develop a difficulty scoring (DS) system to quantify the degree of difficulty in LDP.

METHODS

We collected clinical data for 80 patients who underwent LDP. A 10-level difficulty index was developed and subcategorized into a three-level difficulty index; 1-3 as low, 4-6 as intermediate, and 7-10 as high index. The automatic linear modeling (LINEAR) statistical tool was used to identify factors that significantly increase level of difficulty in LDP.

RESULTS

The operator's 10-level DS concordance between the 10-level DS by the reviewers, LINEAR index DS, and clinical index DS systems were analyzed, and the weighted Cohen's kappa statistic were at 0.869, 0.729, and 0.648, respectively, showing good to excellent inter-rater agreement. We identified five factors significantly affecting level of difficulty in LDP; type of operation, resection line, proximity of tumor to major vessel, tumor extension to peripancreatic tissue, and left-sided portal hypertension/splenomegaly.

CONCLUSIONS

This novel DS for LDP adequately quantified the degree of difficulty, and can be useful for selecting patients for LDP, in conjunction with fitness for surgery and prognosis.

Validation of a Difficulty Grading Score in Laparoscopic Splenectomy

INTRODUCTION

Laparoscopic splenectomy (LS) is the gold standard for the treatment of many spleen-related disorders. However, in the presence of splenomegaly, the rate of conversion to open surgery can be as high as 33.9% and is associated with longer operative times and higher costs. In an attempt to improve risk stratification and patient selection for LS, a score that includes four preoperative parameters (age, gender, diagnosis, and spleen size) has been developed. The aim was the validation of a difficulty grading score, in predicting conversion and poorer outcomes.

METHODS

Retrospective analysis of 153 consecutive patients subjected to LS from January 2006 through December 2016 was performed. Several parameters were reviewed and correlation with evaluated outcomes was analyzed.

RESULTS

Conversion to open surgery occurred in 13 (8.50%) patients and was highly associated with serious intraoperative complications. Spearman correlation showed a significant association between the score and risk of conversion, operative time, and postoperative complications, but not with intraoperative bleeding.

DISCUSSION

Patient and disease features, incorporated in a difficulty grading score, can reliably determine the difficulty of LS and predict risk of conversion, intraoperative, and postoperative complications. This simple and reproducible score improves risk stratification for LS and could be practical in daily clinical activities.