Complications and failure-to-rescue after pancreatectomy and hospital participation in the targeted American College of Surgeons National Surgical Quality Improvement Program registry

BACKGROUND

More than 700 hospitals participate in the American College of Surgeons National Surgical Quality Improvement Program, but most pancreatectomies are performed in 165 centers participating in the pancreas procedure-targeted registry. We hypothesized that these hospitals ("targeted hospitals") might provide more specialized care than those not participating ("standard hospitals").

METHODS

The 2014 to 2019 pancreas-targeted and standard American College of Surgeons National Surgical Quality Improvement Program registry were reviewed regarding patient demographics, comorbidities, and perioperative outcomes using standard univariate and multivariable logistic regression analyses. Primary outcomes included 30-day mortality and serious morbidity.

RESULTS

The registry included 30,357 pancreatoduodenectomies (80% in targeted hospitals) and 14,800 distal pancreatectomies (76% in targeted hospitals). Preoperative and intraoperative characteristics of patients treated at targeted versus standard hospitals were comparable. On multivariable analysis, pancreatoduodenectomies performed at targeted hospitals were associated with a 39% decrease in 30-day mortality (odds ratio, 0.61; 95% confidence interval, 0.50-0.75), 17% decrease in serious morbidity (odds ratio, 0.83; 95% confidence interval, 0.77-0.89), and 41% decrease in failure-to-rescue (odds ratio, 0.59; 95% confidence interval, 0.47-0.74). These differences did not apply to distal pancreatectomies. Participation in the targeted registry was associated with higher rates of optimal surgery for both pancreatoduodenectomy (odds ratio, 1.33; 95% confidence interval, 1.25-1.41) and distal pancreatectomy (odds ratio, 1.17; 95% confidence interval, 1.06-1.30).

CONCLUSION

Mortality and failure-to-rescue rates after pancreatoduodenectomy in targeted hospitals were nearly half of rates in standard American College of Surgeons National Surgical Quality Improvement Program hospitals. Further research should delineate factors underlying this effect and highlight opportunities for improvement.

Pancreatic Cyst Fluid Analysis

Pancreatic cyst fluid analysis can help diagnose pancreatic cyst type and the risk of high-grade dysplasia and cancer. Recent evidence from molecular analysis of cyst fluid has revolutionized the field with multiple markers showing promise in accurate diagnosis and prognostication of pancreatic cysts. The availability of multi-analyte panels has great potential for more accurate prediction of cancer.

Use of Artificial Intelligence Deep Learning to Determine the Malignant Potential of Pancreatic Cystic Neoplasms With Preoperative Computed Tomography Imaging

BACKGROUND

Society consensus guidelines are commonly used to guide management of pancreatic cystic neoplasms (PCNs). However, downsides of these guidelines include unnecessary surgery and missed malignancy. The aim of this study was to use computed tomography (CT)-guided deep learning techniques to predict malignancy of PCNs.

MATERIALS AND METHODS

Patients with PCNs who underwent resection were retrospectively reviewed. Axial images of the mucinous cystic neoplasms were collected and based on final pathology were assigned a binary outcome of advanced neoplasia or benign. Advanced neoplasia was defined as adenocarcinoma or intraductal papillary mucinous neoplasm with high-grade dysplasia. A convolutional neural network (CNN) deep learning model was trained on 66% of images, and this trained model was used to test 33% of images. Predictions from the deep learning model were compared to Fukuoka guidelines.

RESULTS

Twenty-seven patients met the inclusion criteria, with 18 used for training and 9 for model testing. The trained deep learning model correctly predicted 3 of 3 malignant lesions and 5 of 6 benign lesions. Fukuoka guidelines correctly classified 2 of 3 malignant lesions as high risk and 4 of 6 benign lesions as worrisome. Following deep learning model predictions would have avoided 1 missed malignancy and 1 unnecessary operation.

DISCUSSION

In this pilot study, a deep learning model correctly classified 8 of 9 PCNs and performed better than consensus guidelines. Deep learning can be used to predict malignancy of PCNs; however, further model improvements are necessary before clinical use.

Challenges of Modeling Outcomes for Surgical Infections: A Word of Caution

Background: We developed a novel analytic tool for colorectal deep organ/space surgical site infections (C-OSI) prediction utilizing both institutional and extra-institutional American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) data. Methods: Elective colorectal resections (2006-2014) were included. The primary end point was C-OSI rate. A Bayesian-Probit regression model with multiple imputation (BPMI) via Dirichlet process handled missing data. The baseline model for comparison was a multivariable logistic regression model (generalized linear model; GLM) with indicator parameters for missing data and stepwise variable selection. Out-of-sample performance was evaluated with receiver operating characteristic (ROC) analysis of 10-fold cross-validated samples. Results: Among 2,376 resections, C-OSI rate was 4.6% (n = 108). The BPMI model identified (n = 57; 56% sensitivity) of these patients, when set at a threshold leading to 80% specificity (approximately a 20% false alarm rate). The BPMI model produced an area under the curve (AUC) = 0.78 via 10-fold cross- validation demonstrating high predictive accuracy. In contrast, the traditional GLM approach produced an AUC = 0.71 and a corresponding sensitivity of 0.47 at 80% specificity, both of which were statstically significant differences. In addition, when the model was built utilizing extra-institutional data via inclusion of all (non-Mayo Clinic) patients in ACS-NSQIP, C-OSI prediction was less accurate with AUC = 0.74 and sensitivity of 0.47 (i.e., a 19% relative performance decrease) when applied to patients at our institution. Conclusions: Although the statistical methodology associated with the BPMI model provides advantages over conventional handling of missing data, the tool should be built with data specific to the individual institution to optimize performance.