Machine learning­based radiomics models accurately predict Crohn's disease­related anorectal cancer.

The radiological diagnosis of Crohn's disease (CD)-related anorectal cancer is difficult; it is often found in advanced stages and has a poor prognosis because of the difficulty of curative surgery. However, there are no studies on predicting the diagnosis of CD-related cancer. The present study aimed to develop a predictive model to diagnose CD cancerous lesions more accurately in a way that can be interpreted by clinicians. Patients with CD who developed anorectal CD lesions at Hyogo Medical University (Nishinomiya, Japan) between March 2009 and June 2022 were included in the present study. T2-weighted and T1-weighted magnetic resonance (MR) images were utilized for our analysis. Images of anorectal lesions were segmented using open-source 3D Slicer software, and radiomic features were extracted using PyRadiomics. Six machine learning models were investigated and compared: i) Support vector machine; ii) naive Bayes; iii) random forest; iv) light gradient boosting machine; v) extremely randomized trees; vi) and regularized greedy forest (RGF). SHapley Additive exPlanations (SHAP) values were calculated to assess the extent to which each radiomic feature contributed to the model's predictions compared to baseline, represented as the average of the model's predictions for all test data. The T2-weighted images of 28 patients with anorectal cancer and 40 non-cancer patients were analyzed and the contrast-enhanced T1-weighted images of 22 cancer and 40 non-cancer patients. The model with the highest area under the curve (AUC) was the RGF-based model constructed using T2-weighted image features, achieving an AUC of 0.944 (accuracy, 0.862; recall, 0.830). The SHAP-based model explanation suggested a strong association between the diagnosis of CD-related anorectal cancer and features such as complex lesion texture; greater pixel separation within the same coronal cross-section; larger, randomly distributed clumps of pixels with the same signal intensity; and a more spherical lesion shape on T2-weighted images. The MRI radiomics-based RGF model demonstrated outstanding performance in predicting CD-related anorectal cancer. These results may affect the diagnosis and surveillance strategies of CD-related colorectal cancer.

The Comprehensive Complication Index in Ulcerative Colitis: A Comparison with the Clavien-Dindo Classification.

Introduction: The comprehensive complication index (CCI), which weights all postoperative complications according to severity and integrates them into a single formula, has been reported as a new evaluation system. We aimed to compare the CCI with the Clavien-Dindo Classification (CDC) to patients with ulcerative colitis (UC). Methods: Patients who underwent initial surgery for UC from April 2012 to March 2020 were included. The patients were classified into a length of stay (LOS) >30 days group or an LOS ≤30 days group. We performed a multivariate analysis of risk factors for LOS >30 days in the model with the factors identified in the univariate analysis plus the CCI (the CCI model) and plus CDC (the CDC model). An ROC curve was used to test the difference in the area under the curve (AUC) between the CCI model and the CDC model. Results: The median LOS was 21 days (IQR: 16-29 days), and the rate of LOS >30 days was 119/588 (20.2%). In the CCI model, age at the time of surgery (odds ratio [OR] = 1.24, 95% confidence interval [CI] 1.07-1.45, p = 0.01), ASA score ≥3 (OR = 1.94, 95% CI:1.00-3.76, p = 0.04), and CCI (OR = 1.07, 95% CI: 1.05-1.09; p < 0.01) were identified as independent risk factors for LOS >30 days. The AUC value of the CCI model (0.86) was significantly better in relation to LOS >30 days than that of the CDC model (0.82) (p = 0.02). Conclusion: The CCI was a better measure of LOS than was the CDC and was found to be a useful indicator in UC.

The Impacts of Laparoscopic Restorative Proctocolectomy for Ulcerative Colitis: Systematic Review and Meta-Analysis.

Introduction: Laparoscopic surgery (LAP) is now recognized as the standard procedure for colorectal surgery. However, the standard surgery for ulcerative colitis (UC) is total proctocolectomy with ileal pouch anal anastomosis (IPAA), which may be an overly complex procedure to complete laparoscopically. We conducted this systematic review and meta-analysis to evaluate the efficacy as well as the advantages and disadvantages of LAP-IPAA in patients with UC stratified by the outcome of interest. Method: We performed a systematic literature review by searching the PubMed/MEDLINE, the Cochrane Library, and the Japan Centra Reuvo Medicina databases from inception until January 2023. Meta-analyses were performed for surgical outcomes, including morbidity and surgical course, to evaluate the efficacy of LAP-IPAA. Results: A total of 707 participants, including 341 LAP and 366 open surgery (OPEN) patients in 9 observational studies and one randomized controlled study, were included. From the results of the meta-analyses, the odds ratio (OR) of total complications in LAP was 1.12 (95% CI: 0.58-2.17, p = 0.74). The OR of mortality for LAP was 0.38 (95% CI: 0.08-1.92, p = 0.24). Although the duration of surgery was extended in LAP (mean difference (MD) 118.74 min (95% CI: 91.67-145.81), p < 0.01) and hospital stay were not shortened, the duration until oral intake after surgery was shortened in LAP (MD -2.10 days (95% CI: -3.52-0.68), p = 0.004). Conclusions: During IPAA for UC, a similar morbidity rate was seen for LAP and OPEN. Although LAP necessitates extended surgery, there may be certain advantages to this procedure, including easy visibility during the surgical procedure or a shortened time to oral intake after surgery.

Associations between Pouchitis and Fecal Calprotectin after Restorative Proctocolectomy in Patients with Ulcerative Colitis.

Introduction: Recently, fecal calprotectin has been identified and used as an assessment tool for the confirmation of disease activity in ulcerative colitis. Although a meta-analysis suggested the usefulness of fecal calprotectin for the assessment of pouchitis, the number of participants was still insufficient. Therefore, we prospectively measured fecal calprotectin levels during pouchoscopy and analyzed their associations with pouchitis. Methods: Patients who underwent pouchoscopy after total proctocolectomy and ileal pouch-anal anastomosis for ulcerative colitis were included. Fecal samples were collected for the measurement of calprotectin during pouchoscopy. Patients either with or without suspicious pouchitis were included. Pouchitis was defined as a modified pouchitis disease activity index (m-PDAI) score of ≥5. The associations between the development of pouchitis and the m-PDAI score and fecal calprotectin and serum markers, including C-related protein, albumin, and white blood cells, were assessed. Results: A total of 170 patients were included. Seventy-two patients were diagnosed with pouchitis with an m-PDAI score of 7.3 ± 1.5. The values of fecal calprotectin were 1,500 ± 1,544 µg/g in patients with pouchitis and 259 ± 402 µg/g in patients without pouchitis (p < 0.01). The correlation coefficient between calprotectin and the m-PDAI score was significant (r2 = 0.279, p < 0.001). The cutoff value of fecal calprotectin in receiver operating characteristic analysis was 246 µg/g (area under curve 0.85, sensitivity 83.9%, specificity 71.0%). Fecal samples were able to be collected from 6 patients. The levels of fecal calprotectin significantly decreased from 2,101.3 ± 880.3 µg/g to 284.2 ± 96.9 µg/g in response to the treatment. Conclusions: Elevated fecal calprotectin appeared to be significantly correlated with pouchitis. We should consider the alteration of this marker during treatments in further studies.

Clinical features and surgical selection in colitis-associated colorectal cancer with ulcerative colitis.

PURPOSE: The aim of this study was to compare the clinical characteristics of ulcerative colitis (UC) patients who underwent surgery for cancer/dysplasia with those who underwent surgery for refractory disease and to discuss the preoperative preparation for successful hand-sewn IPAA. METHODS: Patients who underwent surgery for UC between January 2014 and December 2021 at Hyogo Medical University were included in the study. A total of 443 UC surgical cases were included in the study, which comprised 188 cancer/dysplasia patients and 255 refractory patients. Clinical records were compared retrospectively. RESULTS: The proportion of surgical UC cases with cancer/dysplasia has been on the rise, accounting for approximately 40% in recent years. The duration of disease (months) was 186 (2-590) in the cancer/dysplasia group and 48 (1-580) in the refractory group (p = 0.02). UC severity (mild/moderate/severe) was 119/69/0 in the cancer/dysplasia group and 18/157/80 in the refractory group (p < 0.01). The four nutrition factors of weight (55.2 (32.7-99.6) kg: 49.9 (20.3-85.2) kg), body mass index (21.0 (13.9-32.5) kg/m2: 18.3 (11.4-34.1)kg/m2), serum albumin level (4.3 (2.7-5.0)g/dl: 3.4 (1.4-5.2)g/dl) and prognostic nutrition index (49.2 (33.2-61.2): 40.9 (17.4-61.1)) were significantly higher in the cancer/dysplasia group (p < 0.01). The degree of obesity was also significantly higher in the cancer/dysplasia group (p < 0.01). CONCLUSION: UC patients with cancer/dysplasia were more likely than refractory patients to have mild inflammation; they also had a longer duration of UC disease and better nutritional status.

Risk factors for the postoperative recurrence of ulcerative colitis-associated colorectal cancer.

PURPOSE: Although ulcerative colitis-associated colorectal cancer (UC-CRC) has been described, there are few reports regarding recurrent cases of UC-CRC. In this study, we investigated the risk factors for UC-CRC recurrence. METHODS: Recurrence-free survival (RFS) was determined for 144 stage I to III cancer patients among 210 UC-CRC patients from August 2002 to August 2019. The Kaplan‒Meier method was used to obtain the cumulative RFS rate, and the Cox proportional hazard model was used to extract recurrence risk factors. The interaction term between cancer stage and prognostic factors specific to UC-CRC was evaluated using the Cox model. The Kaplan‒Meier method was applied by cancer stage to the UC-CRC-specific prognostic factors for which interaction effects were indicated. RESULTS: There were 18 cases of recurrence involving patients with stage I to III cancer, and the recurrence rate was 12.5%. The cumulative 5-year RFS rate was 87.5%. Multivariable analysis showed that age at surgery (hazard ratio (HR): 0.95, 95% CI: 0.91-0.99, p = 0.02), undifferentiated carcinoma (HR: 4.42, 95% CI: 1.13-17.24, p = 0.03), lymph node metastasis (HR: 4.11, 95% CI: 1.08-15.69, p = 0.03), and vascular invasion (HR: 8.01, 95% CI: 1.54-41.65, p = 0.01) were significant risk factors for recurrence. Patients with stage III CRC in the young adult (age < 50 years) group had a significantly worse prognosis than those in the adult (age ≥ 50 years) group (p < 0.01). CONCLUSION: Age at surgery was identified as a risk factor for UC-CRC recurrence. Young adult patients with stage III cancer may have a poor prognosis.

Have advances in medical therapy for ulcerative colitis impacted surgical treatment?

Aim: The purpose of this study was to examine how the clinical characteristics, indications for surgery, and postoperative course of ulcerative colitis (UC) surgical patients changed before and after the introduction of biological agents. Methods: Patients who underwent surgery for UC at Hyogo Medical University between 2000 and 2019 were included in the study; those who underwent surgery between 2000 and 2009 were included in the early group (n = 864), and those who underwent surgery between 2010 and 2019 were included in the late group (n = 834); each study factor was retrospectively compared. Results: The mean ages at surgery (y) were 39.7 ± 15.1 y in the early group and 46.7 ± 17.8 y in the late group (P < .01). Antitumor necrosis factor agents were used (%) in 2 (0.2) patients in the early group and 317 (38.0) patients in the late group (P < .01). The number of cancer or dysplasia patients for whom surgery was indicated was significantly higher in the late group (11%/26%, P = .01). With regard to surgery in elderly individuals, the number of elderly patients (65 y and older) undergoing surgery was significantly higher in the late group (8.0%/18.6%, P < .01). For emergency surgery, the mortality rates were 16.7% (2/12) in the early surgery group and 15.7% (8/51) in the late surgery group (P = .61). Conclusion: The characteristics of UC patients requiring surgery in Japan have changed. The distribution of surgical indications changed, and the number of patients with cancer and dysplasia requiring surgery increased. The prognosis of elderly patients who underwent emergency surgery was poor.

Rectal carcinoma with dual differentiation toward enteroblastic and neuroendocrine features arising in a patient with ulcerative colitis: a case report.

BACKGROUND: Colorectal carcinoma with enteroblastic differentiation is a rare subtype of colorectal carcinomas expressing at least one characteristic immunohistochemical marker among α-fetoprotein, glypican-3, and spalt-like transcription factor 4. On the other hand, colorectal carcinoma with neuroendocrine differentiation is also a unique subtype of colorectal carcinomas showing expression of at least one distinctive marker among chromogranin A, synaptophysin, and CD56. CASE PRESENTATION: We experienced an extremely rare case of rectal carcinoma with dual differentiation toward enteroblastic and neuroendocrine features in a 53-year-old male patient with long-standing ulcerative colitis (UC). Most of the tumor cells were positive for enteroblastic differentiation markers and approximately a half of them for neuroendocrine differentiation markers. Some tumor cells showed only enteroblastic differentiation, and some did only neuroendocrine feature, but some showed both enteroblastic and neuroendocrine differentiation. CONCLUSION: Colorectal carcinoma with dual differentiation toward enteroblastic and neuroendocrine features has not been reported yet. Neoplastic transformation from pluripotent stem cells in dysplastic epithelium of long-standing UC patients may be associated with such dual differentiation features.

Being Elderly Is Associated with Infectious and Fatal Postoperative Complications in Ulcerative Colitis Patients.

INTRODUCTION: It is well known that the infectious complications and mortality rates are increased among elderly individuals with ulcerative colitis (UC) during medical treatment. However, there have been few reports on surgery in elderly individuals with UC, and some cohort studies have reported surgical complication and mortality rates similar to those in nonelderly individuals. METHODS: UC patients who underwent colectomy at the Hyogo College of Medicine between April 2012 and March 2020 were included in this study. The patients were classified as elderly (≥65) or nonelderly (<65). Characteristics and postoperative complications were analyzed and compared between the groups; possible risk factors for infectious and fatal complications were also analyzed. RESULTS: In all, 136/599 (22.7%) elderly patients were included. The incidence of infectious and fatal complications was 177/599 (29.5%) and 18/599 (3%), respectively. These complication rates were significantly higher in the elderly than the nonelderly group (p < 0.01). Age ≥65 years at surgery (OR = 2.92, 95% CI: 1.87-4.57, p < 0.01) was identified as an independent risk factor for infectious complications. Age ≥65 years at surgery (OR = 8.03, 95% CI: 2.16-29.83, p < 0.01), American Society of Anesthesiologists score ≥3 (OR = 6.00, 95% CI: 1.40-25.6 p = 0.02), and urgent/emergent surgery (OR = 16.24, 95% CI: 1.70-154.95, p = 0.02) were identified as independent risk factors for fatal complications. DISCUSSION/CONCLUSION: Age ≥65 years was identified as a risk factor for infectious and fatal complications. It is important to avoid urgent/emergent surgery in elderly patients with an ASA score >3 by emphasizing surgical and medical collaboration and optimizing the timing of surgery.

Association between preoperative biologic use and surgical morbidity in patients with Crohn's disease.

PURPOSE: We evaluated the influence of preoperative treatments with biologics on surgical morbidity in patients with Crohn's disease (CD). METHODS: We reviewed the surveillance data of patients with CD who underwent surgery between April 2018 and April 2021. The possible risk factors for morbidity were analyzed. RESULTS: A total of 305 surgically treated patients were included. Anti-TNF alpha agents and ustekinumab were used in 92 and 27 patients, respectively, within 12 weeks before surgery. There were no cases of mortality. In total, 70/305 (23.0%) patients developed a complication, and 42/305 (13.8%) patients developed a surgical site infection (SSI) (17 incisional SSIs and 35 organ/space SSIs). Current smoking status (OR 3.44), emergent/urgent surgery (OR 6.85), and abdominoperineal resection (APR) (OR 14.93) were identified as risk factors for total complications. Penetrating disease (OR 14.55) was identified as a risk factor for incisional SSIs. Current smoking status (OR 7.09), an American Society of Anesthesiologists (ASA) score greater than 3 (OR 5.85), a postoperative blood sugar level over 155 mg/dL (OR 4.37), and APR (OR 207.95) were identified as risk factors for organ/space SSIs. CONCLUSIONS: No correlation between preoperative treatment with biologics and surgical mortality or morbidity was found. However, we should perform further analyses on a larger number of patients because the analyses may be limited by selection bias for treatment and several confounding factors.