Anastomotic Leak is Increased With Clostridium difficile Infection After Colectomy: Machine Learning-Augmented Propensity Score Modified Analysis of 46 735 Patients

Collagenase-producing bacteria are common in anastomotic leakage after colorectal surgery: a systematic review

PURPOSE

Some gut bacteria can produce enzymes (collagenases) that can break down collagen in the intestinal wall. This could be a part of the pathophysiology of anastomotic leakage (AL). This systematic review aimed to investigate if such bacteria were present more frequently in AL patients versus non-AL patients following colorectal surgery.

METHODS

This systematic review was reported according to the PRISMA and AMSTAR guidelines. Before the literature search, a study protocol was registered at PROSPERO (CRD42022363454). We searched PubMed, EMBASE, Google Scholar, and Cochrane CENTRAL on April 9th, 2023, for randomized and observational human studies of AL following colorectal surgery with information on gastrointestinal bacteria. The primary outcome was bacteria with the potential to produce collagenase. The risk of bias was assessed with the Newcastle-Ottawa Scale, as all studies were observational.

RESULTS

We included 15 studies, with a total of 52,945 patients, of which 1,747 had AL, and bacteriological information from feces, mucosa, the resected specimen, or drain fluid was presented. In 10 of the 15 studies, one or more collagenase-producing bacteria were identified in the patients with AL. Neither the bacteria nor the collagenase production were quantified in any of the studies. The studies varied greatly in terms of sample material, analytical method, and time of collection. Studies using DNA sequencing methods did not report findings of collagenase-producing bacteria.

CONCLUSION

Collagenase-producing bacteria are more common in patients with AL following colorectal surgery than in patients without AL, but the significance is unclear. From the current studies, it is not possible to determine the pathogenicity of the individual gut bacteria.

Combined application of inflammation-related biomarkers to predict postoperative complications of rectal cancer patients: a retrospective study by machine learning analysis

BACKGROUND

Postoperative complications in patients of rectal cancer pose challenges to postoperative recovery. Accurately predicting these complications is crucial for developing effective treatment plans for patients.

METHODS

In this retrospective study, 493 patients with rectal cancer who underwent radical resection between January 2020 and December 2021 were examined. We evaluated logistic regression, support vector machines, regression trees, and random forests to predict the incidence of postoperative complications in patients and evaluate the performance of the model. The results will be analyzed to make recommendations for reducing complications.

RESULTS

Among the four machine learning models, random forest demonstrated the highest results. The performance of this model was showed with an AUC of 0.880 (95% CI 0.807-0.949), an accuracy of 88.0% (95% CI 0.815-0.929), a sensitivity of 96.6%, and a specificity of 45.8%. Notably, factors such as inflammation related prognostic index, prognostic nutritional index, tumor location, and T stage were found to significantly increase the probability of postoperative complications.

CONCLUSION

Our study provided evidence that machine learning models can effectively evaluate early postoperative complications of the patients after surgery.

Predictors and clinical impact of postoperative diarrhea after colorectal cancer surgery: a prospective, multicenter, observational study (SHISA-1602)

PURPOSE

Postoperative diarrhea, including high-output stoma (HOS), frequently occurs after colorectal surgery; its risk factors and clinical implications on subsequent complications remain unknown. This study aimed to evaluate the risk factors and clinical implications of postoperative diarrhea after primary colorectal cancer (CRC) surgery.

METHODS

This prospective observational study included patients with CRC who underwent radical surgery at six hospitals between June 2016 and December 2017. The patients were categorized into three groups (non-stoma, colostoma, and ileostoma groups).

RESULTS

A total of 178 patients participated in the study. In the non-stoma group, the incidence of postoperative diarrhea was 18.4% (27/147). The incidence of HOS was 28.6% (4/14) in the ileostoma group, and 0% in the colostoma group. Multivariable analyses of the incidence of diarrhea in the non-stoma group indicated that habitual smoking and hypertension were significantly associated with postoperative diarrhea (P = 0.012 and P = 0.0274, respectively). Postoperative diarrhea was more likely to occur in patients with rectal cancer than in those with colon cancer (P = 0.0501). In the non-stoma and ileostoma groups, the probability of the occurrence of other complications with Clavien-Dindo (C-D) grades II or higher was significantly higher in patients with C-D grade I diarrhea, including HOS, than in patients without diarrhea (39.3% vs. 14.6%, P = 0.0061).

CONCLUSIONS

Smoking and hypertension are the independent predictors of postoperative diarrhea after an elective CRC surgery. Rectal cancer surgery seems to be associated with postoperative diarrhea more than colon cancer surgery does. Mild postoperative diarrhea may lead to more severe complications.

Racial and Socioeconomic Disparities in Out-Of-Hospital Cardiac Arrest Outcomes: Artificial Intelligence-Augmented Propensity Score and Geospatial Cohort Analysis of 3,952 Patients

Introduction

Social disparities in out-of-hospital cardiac arrest (OHCA) outcomes are preventable, costly, and unjust. We sought to perform the first large artificial intelligence- (AI-) guided statistical and geographic information system (GIS) analysis of a multiyear and multisite cohort for OHCA outcomes (incidence and poor neurological disposition).

Method

We conducted a retrospective cohort analysis of a prospectively collected multicenter dataset of adult patients who sequentially presented to Houston metro area hospitals from 01/01/07-01/01/16. Then AI-based machine learning (backward propagation neural network) augmented multivariable regression and GIS heat mapping were performed.

Results

Of 3,952 OHCA patients across 38 hospitals, African Americans were the most likely to suffer OHCA despite representing a significantly lower percentage of the population (42.6 versus 22.8%; p < 0.001). Compared to Caucasians, they were significantly more likely to have poor neurological disposition (OR 2.21, 95%CI 1.25–3.92; p=0.006) and be discharged to a facility instead of home (OR 1.39, 95%CI 1.05–1.85; p=0.023). Compared to the safety net hospital system primarily serving poorer African Americans, the university hospital serving primarily higher income commercially and Medicare insured patients had the lowest odds of death (OR 0.45, p < 0.001). Each additional $10,000 above median household income was associated with a decrease in the total number of cardiac arrests per zip code by 2.86 (95%CI -4.26- -1.46; p < 0.001); zip codes with a median income above $54,600 versus the federal poverty level had 14.62 fewer arrests (p < 0.001). GIS maps showed convergence of the greater density of poor neurologic outcome cases and greater density of poorer African American residences.

Conclusion

This large, longitudinal AI-guided analysis statistically and geographically identifies racial and socioeconomic disparities in OHCA outcomes in a way that may allow targeted medical and public health coordinated efforts to improve clinical, cost, and social equity outcomes.

TAVR and cancer: machine learning-augmented propensity score mortality and cost analysis in over 30 million patients

Introduction

Cardiovascular disease (CVD) and cancer are the top mortality causes globally, yet little is known about how the diagnosis of cancer affects treatment options in patients with hemodynamically compromising aortic stenosis (AS). Patients with cancer often are excluded from aortic valve replacement (AVR) trials including trials with transcatheter AVR (TAVR) and surgical AVR (SAVR). This study looks at how cancer may influence treatment options and assesses the outcome of patients with cancer who undergo SAVR or TAVR intervention. Additionally, we sought to quantitate and compare both clinical and cost outcomes for patients with and without cancer.

Methods

This population-based case-control study uses the most recent year available National Inpatient Sample (NIS (2016) from the United States Department of Health and Human Services’ Agency for Healthcare Research and Quality (AHRQ). Machine learning augmented propensity score adjusted multivariable regression was conducted based on the likelihood of undergoing TAVR versus medical management (MM) and TAVR versus SAVR with model optimization supported by backward propagation neural network machine learning.

Results

Of the 30,195,722 total hospital admissions, 39,254 (0.13%) TAVRs were performed, with significantly fewer performed in patients with versus without cancer even in those of comparable age and mortality risk (23.82% versus 76.18%, p < 0.001) despite having similar hospital and procedural mortality. Multivariable regression in patients with cancer demonstrated that mortality was similar for TAVR, MM, and SAVR, though LOS and cost was significantly lower for TAVR versus MM and comparable for TAVR versus SAVR. Patients with prostate cancer constituted the largest primary cancer among TAVR patients including those with metastatic disease. There were no significant race or geographic disparities for TAVR mortality.

Discussion

Comparison of aortic valve intervention in patients with and without cancer suggests that interventions are underutilized in the cancer population. This study suggests that patients with cancer including those with metastasis have similar inpatient outcomes to patients without cancer. Further, patients who have symptomatic AS and those with higher risk aortic valve disease should be offered the benefit of intervention. Modern techniques have reduced intervention-related adverse events, provided improved quality of life, and appear to be cost effective; these advantages should not necessarily be denied to patients with co-existing cancer.