Diagnostic significance of the CT rim sign in cases of gangrenous cholecystitis

Prediction of extensive necrotic change in acute gangrenous cholecystitis

PURPOSE

Gangrenous cholecystitis (GC) is a severe type of acute cholecystitis that implies higher mortality and morbidity rates than uncomplicated cholecystitis. The characteristics of GC are various for each case. However, preoperative predictors of GC with extensive necrotic change have not been investigated well.

METHODS

A total of 239 patients who were pathologically diagnosed with GC underwent laparoscopic cholecystectomy at our hospital between January 2013 and December 2021. Of these, 135 patients were included in this study and were subdivided into the extensive necrosis group (patients with necrotic change extending to the neck of the gallbladder, n = 18) and the control group (patients with necrotic change limited to the fundus or body, not extending to the neck, n = 117) according to each operation video. Patient characteristics and perioperative factors predicting extensive necrotic change were investigated.

RESULTS

Pericholecystic fat stranding (83.3 vs. 53.8%, p = 0.018) and absence of wall enhancement on preoperative CT images (50.0 vs. 24.7%, p = 0.026) were significantly associated with extensive necrosis. Seven of 18 patients in the extensive necrosis group showed necrotic changes beyond the infundibulum. The absence of wall enhancement on preoperative CT images (71.4 vs. 28.8%, p = 0.018) was significantly associated with necrotic changes beyond the infundibulum.

CONCLUSIONS

Pericholecystic fat stranding and absence of wall enhancement on preoperative enhanced CT are predictors of extensive necrotic change in patients with GC. In addition, the absence of wall enhancement also predicts the presence of necrotic changes beyond the infundibulum.

Diagnosing gangrenous cholecystitis on computed tomography using deep learning: A preliminary study

Aim

To compare deep learning and experienced physicians in diagnosing gangrenous cholecystitis using computed tomography images and explore the feasibility of diagnostic assistance for acute cholecystitis requiring emergency surgery.

Methods

This retrospective study included 25 patients with pathologically confirmed gangrenous cholecystitis and 129 patients with noncomplicated acute cholecystitis who underwent computed tomography between 2016 and 2021 at two institutions. All available computed tomography images at the time of the initial diagnosis were used for the analysis. A deep learning model based on a convolutional neural network was trained using 1,517 images of 112 patients (18 patients with gangrenous cholecystitis and 94 patients with acute cholecystitis) and tested with 68 images of 42 patients (seven patients with gangrenous cholecystitis and 35 patients with acute cholecystitis). Three blinded, experienced physicians independently interpreted the test images. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were compared between the convolutional neural network and the reviewers.

Results

The convolutional neural network (sensitivity, 0.70; 95% confidence interval [CI], 0.44–0.87, specificity, 0.93; 95% CI, 0.88–0.96, accuracy, 0.89; 95% CI, 0.81–0.95, area under the receiver operating characteristic curve, 0.84; 95% CI, 0.68–1.00) had achieved a better diagnostic performance than the reviewers (ex. sensitivity, 0.55; 95% CI, 0.30–0.77, specificity, 0.67; 95% CI, 0.62–0.71, accuracy, 0.65; 95% CI, 0.57–0.72, area under the receiver operating characteristic curve, 0.63; 95% CI, 0.44–0.82; P = 0.048 for area under the receiver operating characteristic curve versus convolutional neural network).

Conclusions

Deep learning had a better diagnostic performance than experienced reviewers in diagnosing gangrenous cholecystitis and has potential applicability for assisting in identifying indications for emergency surgery in the future.