Nomogram to predict prolonged postoperative ileus after gastrectomy in gastric cancer

Artificial intelligence algorithms for predicting post-operative ileus after laparoscopic surgery

Objective

By constructing a predictive model using machine learning and deep learning technologies, we aim to understand the risk factors for postoperative intestinal obstruction in laparoscopic colorectal cancer patients, and establish an effective artificial intelligence-based predictive model to guide individualized prevention and treatment, thus improving patient outcomes.

Methods

We constructed a model of the artificial intelligence algorithm in Python. Subjects were randomly assigned to either a training set for variable identification and model construction, or a test set for testing model performance, at a ratio of 7:3. The model was trained with ten algorithms. We used the AUC values of the ROC curves, as well as accuracy, precision, recall rate and F1 scores.

Results

The results of feature engineering composited with the GBDT algorithm showed that opioid use, anesthesia duration, and body weight were the top three factors in the development of POI. We used ten machine learning and deep learning algorithms to validate the model, and the results were as follows: the three algorithms with best accuracy were XGB (0.807), Decision Tree (0.807) and Neural DecisionTree (0.807); the two algorithms with best precision were XGB (0.500) and Decision Tree (0.500); the two algorithms with best recall rate were adab (0.243) and Decision Tree (0.135); the two algorithms with highest F1 score were adab (0.290) and Decision Tree (0.213); and the three algorithms with best AUC were Gradient Boosting (0.678), XGB (0.638) and LinearSVC (0.633).

Conclusion

This study shows that XGB and Decision Tree are the two best algorithms for predicting the risk of developing ileus after laparoscopic colon cancer surgery. It provides new insight and approaches to the field of postoperative intestinal obstruction in colorectal cancer through the application of machine learning techniques, thereby improving our understanding of the disease and offering strong support for clinical decision-making.

Nomogram for predicting prolonged postoperative ileus after laparoscopic low anterior resection for rectal cancer

BACKGROUND

Prolonged postoperative ileus (PPOI) is a common complication after colorectal surgery that increases patient discomfort, hospital stay, and financial burden. However, predictive tools to assess the risk of PPOI in patients undergoing laparoscopic low anterior resection have not been developed. Thus, the purpose of this study was to develop a nomogram to predict PPOI after laparoscopic low anterior resection for rectal cancer.

METHODS

A total of 548 consecutive patients who underwent laparoscopic low anterior resection for mid-low rectal cancer at a single tertiary medical center were retrospectively enrolled between January 2019 and January 2023. Univariate and multivariate logistic regression analysis was performed to analyze potential predictors of PPOI. The nomogram was constructed using the filtered variables and internally verified by bootstrap resampling. Model performance was evaluated by receiver operating characteristic curve and calibration curve, and the clinical usefulness was evaluated by the decision curve.

RESULTS

Among 548 consecutive patients, 72 patients (13.1%) presented with PPOI. Multivariate logistic analysis showed that advantage age, hypoalbuminemia, high surgical difficulty, and postoperative use of opioid analgesic were independent prognostic factors for PPOI. These variables were used to construct the nomogram model to predict PPOI. Internal validation, conducted through bootstrap resampling, confirmed the great discrimination of the nomogram with an area under the curve of 0.738 (95%CI 0.736-0.741).

CONCLUSIONS

We created a novel nomogram for predicting PPOI after laparoscopic low anterior resection. This nomogram can assist surgeons in identifying patients at a heightened risk of PPOI.

Nomogram for predicting prolonged postoperative ileus in colorectal cancer based on age and inflammatory markers

Aim: To explore the relationship between inflammatory markers and prolonged postoperative ileus (PPOI), and to establish a nomogram for predicting PPOI. Patients & methods: The data of 229 patients were analyzed retrospectively. Univariate and multivariate logistic regression analysis were used to analyze the risk factors affecting the occurrence of PPOI. The predictive model of PPOI was established and verified internally. Results: Postoperative PPOI occurred in 87 (38.0%) of all 229 patients. Our study showed that age, preoperative neutrophil-lymphocyte ratio and changes in neutrophil-lymphocyte ratio were independent risk factors for PPOI. Conclusion: The nomograms established based on these independent risk factors have good predictive efficacy and may be able to guide clinicians to individualize the diagnosis and treatment.

The effect of probiotics on surgical outcomes in patients with gastric cancer: a meta-analysis of randomized controlled trials

The effect of probiotics on postoperative infectious complications and nutritional status in patients with gastric cancer is still controversial, and a comprehensive search and analysis of the current relevant evidence is necessary. Our study aimed to define the effects of probiotics on surgical outcomes in gastric cancer patients undergoing surgery. Up to June 20, 2023, Embase, PubMed, Web of Science, and Cochrane databases were searched for randomized controlled trials of probiotics in gastric cancer patients undergoing surgery. Relative risk (RR) or mean difference (MD) was used to calculate the effect sizes using RevMan 5.3. A total of nine studies reporting on 861 participants were included. Perioperative supplementation with probiotics did not improve weight loss (MD 0.73 kg; 95% CI: -0.56, 2.02) or serum prealbumin levels (MD 9.48 mg/L 95% CI: -3.43, 22.40), but did reduce the incidence of postoperative infectious complications (RR 0.46, 95% CI 0.28, 0.77), shorten the time to first exhaust (MD -11.27 h; 95% CI: -16.83, -5.70), the time to first defecation (MD -15.71 h; 95% CI: -25.62, -5.79), and the length of hospital stay (MD -0.94 days; 95% CI: -1.33, -0.55), and increase serum albumin levels (MD 0.73 g/L; 95% CI: 0.01, 1.46) in gastric cancer patients undergoing surgery. Probiotics are effective in preventing postoperative infectious complications, promoting postoperative recovery, and improving nutritional status in gastric cancer patients undergoing surgery. Our study highlights the importance of probiotics for healthcare systems and offers a potential strategy to improve the prognosis and reduce the medical burden of gastric cancer patients undergoing surgery.

Small extrachromosomal circular DNAs as biomarkers for multi-cancer diagnosis and monitoring

BACKGROUND

Small extrachromosomal circular DNAs (eccDNAs) have the potential to be cancer biomarkers. However, the formation mechanisms and functions of small eccDNAs selected in carcinogenesis are not clear, and whether the small eccDNA profile in the plasma of cancer patients represents that in cancer tissues remains to be elucidated.

METHODS

A novel sequencing workflow based on the nanopore sequencing platform was used to sequence naturally existing full-length small eccDNAs in tissues and plasma collected from 25 cancer patients (including prostate cancer, hepatocellular carcinoma and colorectal cancer), and from an independent validation cohort (including 7 cancer plasma and 14 healthy plasma).

RESULTS

Compared with those in non-cancer tissues, small eccDNAs detected in cancer tissues had a significantly larger number and size (P = 0.040 and 2.2e-16, respectively), along with more even distribution and different formation mechanisms. Although small eccDNAs had different general characteristics and genomic annotation between cancer tissues and the paired plasma, they had similar formation mechanisms and cancer-related functions. Small eccDNAs originated from some specific genes had great multi-cancer diagnostic value in tissues (AUC ≥ 0.8) and plasma (AUC > 0.9), especially increasing the accuracy of multi-cancer prediction of CEA/CA19-9 levels. The high multi-cancer diagnostic value of small eccDNAs originated from ALK&ETV6 could be extrapolated from tissues (AUC = 0.804) to plasma and showed high positive predictive value (100%) and negative predictive value (82.35%) in a validation cohort.

CONCLUSIONS

As independent and stable circular DNA molecules, small eccDNAs in both tissues and plasma can be used as ideal biomarkers for cost-effective multi-cancer diagnosis and monitoring.

A novel prognostic model to predict OS and DFS of stage II/III gastric adenocarcinoma patients in China

Background

The prognosis of advanced gastric adenocarcinoma (GAC) after radical gastrectomy varies greatly. We aimed to build and validate a novel individualized nomogram based on inflammation index and tumor markers for patients with stage II/III GAC.

Methods

A total of 755 individuals with stage II/III GAC who had undergone radical gastrectomy at the First Affiliated Hospital of Zhengzhou University between 2012 and 2017 were included in this retrospective study. The patients were randomly divided into a training cohort (n ​= ​503) and a validation cohort (n ​= ​252). Univariate and multivariate analyses were used to determine independent prognostic factors of overall survival (OS) and disease-free survival (DFS). A nomogram was developed based on these independent factors. The concordance index (C-index) and calibration curves were used to evaluate the predictive accuracy of the nomogram.

Results

Univariate and multivariate analyses demonstrated that older age, poor differentiation, advanced stage, elevated neutrophil-to-lymphocyte ratio (NLR), lower hemoglobin, and high carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) levels were significantly associated with lower OS and DFS and were independent prognostic factors in stage II/III GAC. The nomogram developed based on these factors in the training cohort showed excellent calibration and discrimination (OS: C-index ​= ​0.739, 95% CI ​= ​0.706-0.772; DFS: C-index ​= ​0.735, 95% CI ​= ​0.702-0.769). In the internal validation cohort, the nomogram was also well-calibrated for the prediction of OS and DFS; it was superior to the 8th edition UICC/AJCC TNM staging system (for OS: C-index ​= ​0.746 vs. 0.679, respectively; for DFS: C-index ​= ​0.736 vs. 0.675, respectively; P ​< ​0.001).

Conclusion

The nomogram model could reliably predict OS and DFS in stage II/III gastric cancer patients with radical gastrectomy. It may help physicians make better treatment decisions.

Nomogram for prediction of prolonged postoperative ileus after colorectal resection

BACKGROUND

Prolonged postoperative ileus (PPOI) is a major complication in patients undergoing colorectal resection. The aim of this study was to analyze the risk factors contributing to PPOI, and to develop an effective nomogram to determine the risks of this population.

METHODS

A total of 1,254 patients with colorectal cancer who underwent radical colorectal resection at Fujian Cancer Hospital from March 2016 to August 2021 were enrolled as a training cohort in this study. Univariate analysis and multivariate logistic regressions were performed to determine the correlation between PPOI and clinicopathological characteristics. A nomogram predicting the incidence of PPOI was constructed. The cohort of 153 patients from Fujian Provincial Hospital were enrolled as a validation cohort. Internal and external validations were used to evaluate the prediction ability by area under the receiver operating characteristic curve (AUC) and a calibration plot.

RESULTS

In the training cohort, 128 patients (10.2%) had PPOI after colorectal resection. The independent predictive factors of PPOI were identified, and included gender, age, surgical approach and intraoperative fluid overload. The AUC of nomogram were 0.779 (95% CI: 0.736-0.822) and 0.791 (95%CI: 0.677-0.905) in the training and validation cohort, respectively. The two cohorts of calibration plots showed a good consistency between nomogram prediction and actual observation.

CONCLUSIONS

A highly accurate nomogram was developed and validated in this study, which can be used to provide individual prediction of PPOI in patients after colorectal resection, and this predictive power can potentially assist surgeons to make the optimal treatment decisions.

Construction of a novel signature and prediction of the immune landscape in gastric cancer based on necroptosis-related genes

Necroptosis, a type of programmed cell death, has become a potential therapeutic target for solid tumors. Nevertheless, the potential roles of necroptosis-related genes (NRGs) in gastric cancer (GC) remain unknown. The objective of the present study was to create a necroptosis-related prognostic signature that can provide more accurate assessment of prognosis in GC. Using The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) data, we identified differentially expressed NRGs. Univariate analysis and Lasso regression were performed to determine the prognostic signature. Risk scores were calculated and all GC patients were divided into high- and low-risk score group according to the median risk score value. The robustness of this signature was externally validated with data from GSE84437 cohort (n = 431). Survival analysis revealed high-risk score patients had a worse prognosis. Results evidenced that the signature was an independent prognosis factor for survival. Single-sample sequence set enrichment analysis (ssGSEA) exhibited different enrichment of immune cells and immune-related pathways in the two risk groups. Furthermore, a predictive nomogram was generated and showed excellent predictive performance based on discrimination and calibration. In addition, the risk score positively correlated with tumor mutational burden and was associated with sensitivity to multiple anti-cancer drugs. Overall, our work demonstrates a close relationship between necroptosis and the prognosis of GC. The signature we constructed with potential clinical application value, can be used for prognosis prediction and being a potential therapeutic responses indicator in GC patients.

Identification and validation of a novel ferroptosis-related gene model for predicting the prognosis of gastric cancer patients

Background

Ferroptosis is a novel form of regulated cell death that plays a critical role in tumorigenesis. The purpose of this study was to establish a ferroptosis-associated gene (FRG) signature and assess its clinical outcome in gastric cancer (GC).

Methods

Differentially expressed FRGs were identified using gene expression profiles from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) database. Univariate and least absolute shrinkage and selection operator (LASSO) Cox regression analyses were performed to construct a prognostic signature. The model was validated using an independent GEO dataset, and a genomic-clinicopathologic nomogram integrating risk scores and clinicopathological features was established.

Results

An 8-FRG signature was constructed to calculate the risk score and classify GC patients into two risk groups (high- and low-risk) according to the median value of the risk score. The signature showed a robust predictive capacity in the stratification analysis. A high-risk score was associated with advanced clinicopathological features and an unfavorable prognosis. The predictive accuracy of the signature was confirmed using an independent GSE84437 dataset. Patients in the two groups showed different enrichment of immune cells and immune-related pathways. Finally, we established a genomic-clinicopathologic nomogram (based on risk score, age, and tumor stage) to predict the overall survival (OS) of GC patients.

Conclusions

The novel FRG signature may be a reliable tool for assisting clinicians in predicting the OS of GC patients and may facilitate personalized treatment.

It's not just an ileus: disparities associated with ileus following ventral hernia repair

PURPOSE

We sought to identify risk factors associated with postoperative ileus following ventral hernia repair.

METHODS

Utilizing the Nationwide Inpatient Sample (NIS) from 2008 to 2012, we identified adult patients that underwent either open or laparoscopic hernia repair for umbilical and ventral hernias with a diagnosis of umbilical/ventral hernia. We excluded cases with diagnosis of obstruction and bowel gangrene that underwent bowel resection, or with missing data. Risk variables of interest were age, sex, race, income status, insurance status, elective admission, comorbidity status (using the validated van Walraven Score), complications (mechanical, respiratory, postoperative infection, cardiovascular, intraoperative), morbid obesity, procedure type, conversion to open, mesh use, hospital type (rural, urban non-teaching, urban teaching), bed size, and region (northeast, midwest, south, west). Univariate analysis comparing patients with ileus vs control was performed. We then performed multivariable analysis using logistic regression, adjusting for all the risk variables, with ileus as the dependent variable.

RESULTS

30,912 patients were identified that met criteria. Of these, 2660 (8.61%) had postoperative ileus during their stay at the hospital. Univariate analysis showed all risk variables were associated with development of ileus with the exception of income status (p = 0.2903), elective admission (p = 0.7989), mesh use (p = 0.3620), and hospital bed size (p = 0.08351). Median length of stay was 7 days in the ileus cohort vs 3 days in control (p < 0.0001). Median total charges (adjusted to 2012 dollars) was $54,819 vs $35,058 (p < 0.0001). We then performed logistic regression adjusting for all risk variables and found that age (OR 1.66, p < 0.0001), male sex (OR 1.51, p < 0.0001), Black race (OR 1.49, p < 0.0001), comorbidity status (OR 1.12, p < 0.0001), laparoscopic cases converted to open (OR 1.55, p < 0.0001), postoperative complications (mechanical: OR 2.32, p < 0.0001, respiratory: OR 1.54, p < 0.0001, postoperative infection: OR 2.12, p < 0.0001, cardiovascular: OR 1.57, p = 0.0006, intraoperative: OR 1.29, p = 0.0200) were independently associated with increased risk of ileus. However, laparoscopic vs open (OR 0.76, p < 0.0001), elective admission (OR 0.91, p = 0.0378), and northeast vs south hospital region (OR 0.74, p < 0.0001) were independently associated with decreased risk of ileus.

CONCLUSION

We performed a large observational study looking for risk factors associated with ileus following ventral hernia repair. Race and region of treatment are independent risk factors associated with ileus following ventral hernia repair, and a potential source of disparities in care and increased admission length and higher cost of care. Further prospective studies are warranted.