A Nomogram to Predict Anastomotic Leakage in Open Rectal Surgery-Hope or Hype?

Systematic review of preoperative and intraoperative colorectal Anastomotic Leak Prediction Scores (ALPS)

OBJECTIVE

To systematically review preoperative and intraoperative Anastomotic Leak Prediction Scores (ALPS) and validation studies to evaluate performance and utility in surgical decision-making. Anastomotic leak (AL) is the most feared complication of colorectal surgery. Individualised leak risk could guide anastomosis and/or diverting stoma.

METHODS

Systematic search of Ovid MEDLINE and Embase databases, 30 October 2020, identified existing ALPS and validation studies. All records including >1 risk factor, used to develop new, or to validate existing models for preoperative or intraoperative use to predict colorectal AL, were selected. Data extraction followed CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies guidelines. Models were assessed for applicability for surgical decision-making and risk of bias using Prediction model Risk Of Bias ASsessment Tool.

RESULTS

34 studies were identified containing 31 individual ALPS (12 colonic/colorectal, 19 rectal) and 6 papers with validation studies only. Development dataset patient populations were heterogeneous in terms of numbers, indication for surgery, urgency and stoma inclusion. Heterogeneity precluded meta-analysis. Definitions and timeframe for AL were available in only 22 and 11 ALPS, respectively. 26/31 studies used some form of multivariable logistic regression in their modelling. Models included 3-33 individual predictors. 27/31 studies reported model discrimination performance but just 18/31 reported calibration. 15/31 ALPS were reported with external validation, 9/31 with internal validation alone and 4 published without any validation. 27/31 ALPS and every validation study were scored high risk of bias in model analysis.

CONCLUSIONS

Poor reporting practices and methodological shortcomings limit wider adoption of published ALPS. Several models appear to perform well in discriminating patients at highest AL risk but all raise concerns over risk of bias, and nearly all over wider applicability. Large-scale, precisely reported external validation studies are required.

PROSPERO REGISTRATION NUMBER

CRD42020164804.

Artificial intelligence assists surgeons' decision-making of temporary ileostomy in patients with rectal cancer who have received anterior resection

BACKGROUND

Due to the difficult evaluation of the risk of anastomotic leakage (AL) after rectal cancer resection, the decision to perform a temporary ileostomy is not easily distinguishable. The aim of the present study was to develop an artificial intelligence (AI) model for identifying the risk of AL to assist surgeons in the selective implementation of a temporary ileostomy.

MATERIALS AND METHODS

The data from 2240 patients with rectal cancer who received anterior resection were collected, and these patients were divided into one training and two test cohorts. Five AI algorithms, such as support vector machine (SVM), logistic regression (LR), Naive Bayes (NB), stochastic gradient descent (SGD) and random forest (RF) were employed to develop predictive models using clinical variables and were assessed using the two test cohorts.

RESULTS

The SVM model indicated good discernment of AL, and might have increased the implementation of temporary ileostomy in patients with AL in the training cohort (p < 0.001). Following the assessment of the two test cohorts, the SVM model could identify AL in a favorable manner, which performed with positive predictive values of 0.150 (0.091-0.234) and 0.151 (0.091-0.237), and negative predictive values of 0.977 (0.958-0.988) and 0.986 (0.969-0.994), respectively. It is important to note that the implementation of temporary ileostomy in patients without AL would have been significantly reduced (p < 0.001) and which would have been significantly increased in patients with AL (p < 0.05).

CONCLUSION

The model (https://alrisk.21cloudbox.com/) indicated good discernment of AL, which may be used to assist the surgeon's decision-making of performing temporary ileostomy.

A preoperative prediction model based on Lymphocyte-C-reactive protein ratio predicts postoperative anastomotic leakage in patients with colorectal carcinoma: a retrospective study

Background & Aims

Lymphocyte-C-reactive Protein Ratio (LCR) has been demonstrated as a promising new marker for predicting surgical and oncological outcomes in colorectal carcinoma (CRC). However, anastomotic leakage (AL) is also likely related to this inflammatory marker. Herein, we aimed to identify preoperative predictors of AL and build and develop a novel model able to identify patients at risk of developing AL.

Methods

We collected 858 patients with CRC undergoing elective radical operation between 2007 and 2018 at a single center were retrospectively reviewed. We performed univariable and multivariable analyses and built a multivariable model that predicts AL based on preoperative factors. Propensity adjustment was used to correct the bias introduced by non-random matching of the LCR. The model's performance was evaluated by using the area under the receiver operator characteristic curves (AUROCs), decision curve analysis (DCA), Brier scores, D statistics, and R2 values.

Results

Age, nutrition risk screening 2002 (NRS2002) score, tumor location and LCR, together with hemoglobin < 90 g/l, were independent predictors of AL. The models built on these variables showed good performance (internal validation: c-statistic = 0.851 (95%CI 0.803–0.965), Brier score = 0.049; temporal validation: c-statistic = 0.777 (95%CI 0.823–0.979), Brier score = 0.096). A regression equation to predict the AL was also established by multiple linear regression analysis: [Age(≥ 60 year) × 1.281] + [NRS2002(≥ 3) × 1.341] + [Tumor location(pt.) × 1.348]-[LCR(≤ 6000) × 1.593]-[Hemoglobin(< 90 g/L) × 1.589]-6.12.

Conclusion

Preoperative LCR is an independent predictive factor for AL. A novel model combining LCR values, age, tumor location, and NRS2002 provided an excellent preoperative prediction of AL in patients with CRC. The nomogram can help clinical decision-making and support future research.

A preoperative prediction model for anastomotic leakage after rectal cancer resection based on 13.175 patients

INTRODUCTION

This study aims to develop a robust preoperative prediction model for anastomotic leakage (AL) after surgical resection for rectal cancer, based on established risk factors and with the power of a large prospective nation-wide population-based study cohort.

MATERIALS AND METHODS

A development cohort was formed by using the DCRA (Dutch ColoRectal Audit), a mandatory population-based repository of all patients who undergo colorectal cancer resection in the Netherlands. Patients aged 18 years or older were included who underwent surgical resection for rectal cancer with primary anastomosis (with or without deviating ileostomy) between 2011 and 2019. Anastomotic leakage was defined as clinically relevant leakage requiring reintervention. Multivariable logistic regression was used to build a prediction model and cross-validation was used to validate the model.

RESULTS

A total of 13.175 patients were included for analysis. AL was diagnosed in 1319 patients (10%). A deviating stoma was constructed in 6853 patients (52%). The following variables were identified as significant risk factors and included in the prediction model: gender, age, BMI, ASA classification, neo-adjuvant (chemo)radiotherapy, cT stage, distance of the tumor from anal verge, and deviating ileostomy. The model had a concordance-index of 0.664, which remained 0.658 after cross-validation. In addition, a nomogram was developed.

CONCLUSION

The present study generated a discriminative prediction model based on preoperatively available variables. The proposed score can be used for patient counselling and risk-stratification before undergoing rectal resection for cancer.

Training prediction models for individual risk assessment of postoperative complications after surgery for colorectal cancer

BACKGROUND

The occurrence of postoperative complications and anastomotic leakage are major drivers of mortality in the immediate phase after colorectal cancer surgery. We trained prediction models for calculating patients' individual risk of complications based only on preoperatively available data in a multidisciplinary team setting. Knowing prior to surgery the probability of developing a complication could aid in improving informed decision-making by surgeon and patient and individualize surgical treatment trajectories.

METHODS

All patients over 18 years of age undergoing any resection for colorectal cancer between January 1, 2014 and December 31, 2019 from the nationwide Danish Colorectal Cancer Group database were included. Data from the database were converted into Observational Medical Outcomes Partnership Common Data Model maintained by the Observation Health Data Science and Informatics initiative. Multiple machine learning models were trained to predict postoperative complications of Clavien-Dindo grade ≥ 3B and anastomotic leakage within 30 days after surgery.

RESULTS

Between 2014 and 2019, 23,907 patients underwent resection for colorectal cancer in Denmark. A Clavien-Dindo complication grade ≥ 3B occurred in 2,958 patients (12.4%). Of 17,190 patients that received an anastomosis, 929 experienced anastomotic leakage (5.4%). Among the compared machine learning models, Lasso Logistic Regression performed best. The predictive model for complications had an area under the receiver operating characteristic curve (AUROC) of 0.704 (95%CI 0.683-0.724) and an AUROC of 0.690 (95%CI 0.655-0.724) for anastomotic leakage.

CONCLUSIONS

The prediction of postoperative complications based only on preoperative variables using a national quality assurance colorectal cancer database shows promise for calculating patient's individual risk. Future work will focus on assessing the value of adding laboratory parameters and drug exposure as candidate predictors. Furthermore, we plan to assess the external validity of our proposed model.

Conventional regression analysis and machine learning in prediction of anastomotic leakage and pulmonary complications after esophagogastric cancer surgery

Background and Objectives

With the current advanced data‐driven approach to health care, machine learning is gaining more interest. The current study investigates the added value of machine learning to linear regression in predicting anastomotic leakage and pulmonary complications after upper gastrointestinal cancer surgery.

Methods

All patients in the Dutch Upper Gastrointestinal Cancer Audit undergoing curatively intended esophageal or gastric cancer surgeries from 2011 to 2017 were included. Anastomotic leakage was defined as any clinically or radiologically proven anastomotic leakage. Pulmonary complications entailed: pneumonia, pleural effusion, respiratory failure, pneumothorax, and/or acute respiratory distress syndrome. Different machine learning models were tested. Nomograms were constructed using Least Absolute Shrinkage and Selection Operator.

Results

Between 2011 and 2017, 4228 patients underwent surgical resection for esophageal cancer, of which 18% developed anastomotic leakage and 30% a pulmonary complication. Of the 2199 patients with surgical resection for gastric cancer, 7% developed anastomotic leakage and 15% a pulmonary complication. In all cases, linear regression had the highest predictive value with the area under the curves varying between 61.9 and 68.0, but the difference with machine learning models did not reach statistical significance.

Conclusion

Machine learning models can predict postoperative complications in upper gastrointestinal cancer surgery, but they do not outperform the current gold standard, linear regression

Risk Nomogram Does Not Predict Anastomotic Leakage After Colon Surgery Accurately: Results of the Multi-center LekCheck Study

PURPOSE

Anastomotic leakage (AL) is a dreaded complication after colorectal surgery. Preoperatively identifying high-risk patients can help to reduce the incidence of this complication. For this reason, AL risk nomograms have been developed. The objective of this study was to test the AL risk nomogram developed by Frasson, et al. for validity and to identify risk-factors for AL.

METHODS

From the international multi-center LekCheck study database, patients who underwent colonic surgery with the formation of an anastomosis were included. Data were prospectively collected between 2016 and 2019 at 14 hospitals. Univariate and multivariable regression analyses, and area under receiver operating characteristic curve analysis (AUROC) were performed.

RESULTS

A total of 643 patients were included. The median age was 70 years and 51% were male. The majority underwent surgery for malignancies (80.7%). The overall AL rate was 9.2%. The risk nomogram was not predictive for AL in the population tested (AUROC 0.572). Low preoperative haemoglobin (p = 0.006), intraoperative hypothermia (p = 0.02), contamination of the operative field (p = 0.004), and use of epidural analgesia (p = 0.02) were independent risk-factors for AL.

CONCLUSION

The AL risk nomogram could not be validated using the international LekCheck study database. In the future, intraoperative predictive factors for AL, as identified in this study, should also be included in AL risk predictors.

A Diagnostic Nomogram for Predicting the Risk of Anastomotic Leakage in Elderly Patients With Rectal Cancer: A Single-center Retrospective Cohort Study

BACKGROUND

Laparoscopic resection for rectal cancer has been gaining popularity over the past 2 decades. Whether elderly patients had more benefits from laparoscopy-assisted anterior resection (LAR) need further investigation when comparing with open anterior resection (OAR).

OBJECTIVES

This study aimed to evaluate the clinical outcomes and prognosis of LAR in elderly patients (65 y and above) with rectal cancer and investigate the factors associated with the anastomotic leakage (AL). Besides, the study sought to create a nomogram for precise prediction of AL after anterior resection for rectal cancer.

MATERIALS AND METHODS

A total of 343 rectal cancer patients over 65 years old who underwent LAR or OAR at a single center between January 2013 to January 2021 were retrospectively reviewed. Univariate analysis was conducted to explore potential risk factors for AL, and a nomogram for AL was created based on the multivariate logistic regression model.

RESULTS

A total of 343 patients were included in this study, 271 patients in LAR group and 72 patients in OAR group. Most of the variables were comparable between the 2 groups. The mean operative time was longer in the LAR group than that in the OAR group (191.66±58.33 vs. 156.85±53.88 min, P<0.0001). The LAR group exhibited a significantly lower intraoperative blood loss than the OAR group (85.17±50.03 vs. 131.67±79.10 mL; P<0.0001). Moreover, laparoscopic surgery resulted in shorter postoperative hospital stay, lower rates of diverting stoma and receiving sphincter sparing surgery in comparison with open surgery. The overall rates of complications were 25.1% and 40.3% in the LAR and OAR groups (P=0.011), respectively. And the reoperation rates in the OAR group (0%) was lower than in the LAR group (1.5%), but the difference did not reach statistical significance (P=0.300). Sex, location of tumor, diverting stoma and combined organ resection were identified as independent risk factors for AL based on multivariate analysis. Such factors were selected to develop a nomogram. After a median follow-up of 37.0 months, our study showed no significant difference in overall survival or disease free survival between the 2 groups for treatment of rectal cancer.

CONCLUSIONS

This study suggests that LAR is an alternative minimally invasive surgical procedure in patients above 65 years with better short-term outcomes and acceptable long-term outcomes compared with OAR. In addition, our nomogram has satisfactory accuracy and clinical utility may benefit for clinical decision-making.

Machine learning-based random forest predicts anastomotic leakage after anterior resection for rectal cancer

Background

Anastomotic leakage (AL) is one of the commonest and most serious complications after rectal cancer surgery. The previous analyses on predictors for AL included small-scale patients, and their prediction models performed unsatisfactorily.

Methods

Clinical data of 5,220 patients who underwent anterior resection for rectal cancer were scrutinized to create a prediction model via random forest classifier. Additionally, data of 836 patients served as the test dataset. Patients diagnosed with AL within 6 months' follow-up were recorded. A total of 20 candidate factors were included. Receiver operating characteristic (ROC) curve was conducted to determine the clinical efficacy of our model, and compare the predictive performance of different models.

Results

The incidence of AL was 6.2% (326/5,220). A multivariate logistic regression analysis and the random forest classifier indicated that sex, distance of tumor from the anal verge, bowel stenosis or obstruction, preoperative hemoglobin, surgeon volume, diabetes, neoadjuvant chemoradiotherapy, and surgical approach were significantly associated with AL. After propensity score matching, the temporary stoma was not identified as a protective factor for AL (P=0.58). Contrastingly, the first year of performing laparoscopic surgery was a predictor (P=0.009). We created a predictive random forest classifier based on the above predictors that demonstrated satisfactory prediction efficacy. The area under the curve (AUC) showed that the random forest had higher efficiency (AUC =0.87) than the nomogram (AUC =0.724).

Conclusions

Our findings suggest that eight factors may affect the incidence of AL. Our random forest classifier is an innovative and practical model to effectively predict AL, and could provide rational advice on whether to perform a temporary stoma, which might reduce the rate of stoma and avoid the ensuing complications.

Cost analysis in a randomized trial of early closure of a temporary ileostomy after rectal resection for cancer (EASY trial)

Background

Hospital costs associated with the treatment of rectal cancer are considerable and the formation of a temporary stoma accounts for additional costs. Results from the EASY trial showed that early closure of a temporary ileostomy was associated with significantly fewer postoperative complications but no difference in health-related quality of life up to 12 months after rectal resection. The aim of the present study was to perform a cost analysis within the framework of the EASY trial.

Methods

Early closure (8–13 days) of a temporary stoma was compared to late closure (> 12 weeks) in the randomized controlled trial EASY (NCT01287637). The study period and follow-up was 12 months after rectal resection. Inclusion of participants was made after index surgery. Exclusion criteria were diabetes mellitus, steroid treatment, signs of postoperative complications or anastomotic leakage. Clinical effectiveness and resource use were derived from the trial and unit costs from Swedish sources. Costs were calculated for the year 2016 and analysed from the perspective of the healthcare sector.

Results

Fifty-five patients underwent early closure, and 57 late closure in eight Swedish and Danish hospitals between 2011 and 2014. The difference in mean cost per patient was 4060 US dollar (95% confidence interval 1121; 6999, p value < 0.01) in favour of early closure. A sensitivity analysis, taking protocol-driven examinations into account, resulted in an overall difference in mean cost per patient of $3608, in favour of early closure (95% confidence interval 668; 6549, p value 0.02). The predominant cost factors were reoperations, readmissions and endoscopic examinations.

Conclusions

The significant cost reduction in this study, together with results of safety and efficacy from the randomized controlled trial, supports the routine use of early closure of a temporary ileostomy after rectal resection for cancer in selected patients without signs of anastomotic leakage.

Clinical trial

Registered at clinicaltrials.gov, clinical trials identifier NCT01287637.