Search

Background Although imaging, endoscopy, and inflammatory biomarkers are associated with future Crohn disease (CD) outcomes, common laboratory studies may also provide prognostic opportunities. We evaluated machine learning models incorporating routinely collected laboratory studies to predict surgical outcomes in U.S. Veterans with CD. Methods Adults with CD from a Veterans Health Administration, Veterans Integrated Service Networks (VISN) 10 cohort examined between 2001 and 2015 were used for analysis. Patient demographics, medication use, and longitudinal laboratory values were used to model future surgical outcomes within 1 year. Specifically, data at the time of prediction combined with historical laboratory data characteristics, described as slope, distribution statistics, fluctuation, and linear trend of laboratory values, were considered and principal component analysis transformations were performed to reduce the dimensionality. Lasso regularized logistic regression was used to select features and construct prediction models, with performance assessed by area under the receiver operating characteristic using 10-fold cross-validation. Results We included 4950 observations from 2809 unique patients, among whom 256 had surgery, for modeling. Our optimized model achieved a mean area under the receiver operating characteristic of 0.78 (SD, 0.002). Anti-tumor necrosis factor use was associated with a lower probability of surgery within 1 year and was the most influential predictor in the model, and corticosteroid use was associated with a higher probability of surgery. Among the laboratory variables, high platelet counts, high mean cell hemoglobin concentrations, low albumin levels, and low blood urea nitrogen values were identified as having an elevated influence and association with future surgery. Conclusions Using machine learning methods that incorporate current and historical data can predict the future risk of CD surgery.

Automated image analysis methods have shown potential for replicating expert interpretation of histology and endoscopy images, which traditionally require highly specialized and experienced reviewers. Inflammatory bowel disease (IBD) diagnosis, severity assessment, and treatment decision-making require multimodal expert data interpretation and integration, which could be significantly aided by applications of machine learning analyses. This review introduces fundamental concepts of machine learning for imaging analysis and highlights research and development of automated histology and endoscopy interpretation in IBD. Proof-of-concept studies strongly suggest that histologic and endoscopic images can be interpreted with similar accuracy as knowledge experts. Encouraging results support the potential of automating existing disease activity scoring instruments with high reproducibility, speed, and accessibility, therefore improving the standardization of IBD assessment. Though challenges surrounding ground truth definitions, technical barriers, and the need for extensive multicenter evaluation must be resolved before clinical implementation, automated image analysis is likely to both improve access to standardized IBD assessment and advance the fundamental concepts of how disease is measured.

Background 5-aminosalicylate (5-ASA) medications have a long history of use for the treatment of inflammatory bowel disease and continue to be widely prescribed today. The effectiveness of 5-ASAs in ulcerative colitis is clear; however, studies have shown little benefit for induction or maintenance treatment of Crohn disease (CD). We aimed to quantify usage and examine trends in 5-ASA prescription rates in patients with CD. Methods Using a retrospective design, we queried a national database of commercially insured patients (Truven-Health databases) between 2009 and 2014 to identify patients with CD aged 18 to 65 years. Prescription rates for 5-ASA medications including sulfasalazine, mesalamine, olsalazine, and balsalazide were calculated for each calendar year. Regression models were used to examine year-to-year trends in prescription rates and identify patient factors associated with 5-ASA use. Results We identified 132,804 patients with CD, of whom 37.3% (n = 49,529) received a 5-ASA prescription during the study period. From 2009 to 2014, the overall prescription rates of 5-ASAs declined from 42.9% to 30.0% (P < 0.001). Patient factors independently associated with 5-ASA use included younger age, male sex, multimorbidity, and a health maintenance organization insurance plan, while controlling for the region of residence. Conclusions About 1 in 3 privately insured patients with CD received 5-ASA prescriptions despite their questionable effectiveness; however, in an encouraging trend, prescription rates significantly decreased from 2009 to 2014. This high prescription rate may reflect a gap in providers’ knowledge regarding the available evidence—an opportunity for cost savings with improved health care delivery.

Background Extraintestinal manifestations (EIMs) occur commonly in inflammatory bowel disease (IBD), but population-level understanding of EIM behavior is difficult. We present a natural language processing (NLP) system designed to identify both the presence and status of EIMs using clinical notes from patients with IBD. Methods In a single-center retrospective study, clinical outpatient electronic documents were collected in patients with IBD. An NLP EIM detection pipeline was designed to determine general and specific symptomatic EIM activity status descriptions using Python 3.6. Accuracy, sensitivity, and specificity, and agreement using Cohen’s kappa coefficient were used to compare NLP-inferred EIM status to human documentation labels. Results The 1240 individuals identified as having at least 1 EIM consisted of 54.4% arthritis, 17.2% ocular, and 17.0% psoriasiform EIMs. Agreement between reviewers on EIM status was very good across all EIMs (κ = 0.74; 95% confidence interval [CI], 0.70-0.78). The automated NLP pipeline determining general EIM activity status had an accuracy, sensitivity, specificity, and agreement of 94.1%, 0.92, 0.95, and κ = 0.76 (95% CI, 0.74-0.79), respectively. Comparatively, prediction of EIM status using administrative codes had a poor sensitivity, specificity, and agreement with human reviewers of 0.32, 0.83, and κ = 0.26 (95% CI, 0.20-0.32), respectively. Conclusions NLP methods can both detect and infer the activity status of EIMs using the medical document an information source. Though source document variation and ambiguity present challenges, NLP offers exciting possibilities for population-based research and decision support in IBD.

Background Patients with inflammatory bowel disease (IBD) are at increased risk for pneumonia, and corticosteroids are reported to amplify this risk. Less is known about the impact of corticosteroid-sparing IBD therapies on pneumonia risk or the efficacy of pneumococcal vaccination in reducing all-cause pneumonia in real-world IBD cohorts. Methods We performed a population-based study using an established Veterans Health Administration cohort of 29,957 IBD patients. We identified all patients who developed bacterial pneumonia. Cox survival analysis was used to determine the association of corticosteroids at study entry and as a time-varying covariate, corticosteroid-sparing agents (immunomodulators and antitumor necrosis-alpha [TNF] inhibitors), and pneumococcal vaccination with the development of all-cause pneumonia. Results Patients with IBD who received corticosteroids had a greater risk of pneumonia when controlling for age, gender, and comorbidities (hazard ratio [HR] 2.21; 95% confidence interval [CI], 1.90–2.57 for prior use; HR = 3.42; 95% CI, 2.92–4.01 for use during follow-up). Anti-TNF inhibitors (HR 1.52; 95% CI, 1.02–2.26), but not immunomodulators (HR 0.91; 95% CI, 0.77–1.07), were associated with a small increase in pneumonia. A history of pneumonia was strongly associated with subsequent pneumonia (HR = 4.41; 95% CI, 3.70–5.27). Less than 15% of patients were vaccinated against pneumococcus, and this was not associated with a reduced risk of pneumonia (HR = 1.02; 95% CI, 0.80–1.30) in this cohort. Conclusion In a large US cohort, corticosteroids were confirmed to increase pneumonia risk. Tumor necrosis-alpha inhibitors were associated with a smaller increase in the risk of pneumonia. Surprisingly, pneumococcal vaccination did not reduce all-cause pneumonia in this population, though few patients were vaccinated.

Background Intraabdominal abscess management decisions in the treatment of Crohn disease (CD) can be challenging. Our aim was to determine the effect of clinical, medication use, and imaging disease characteristics on the need for future surgical management. Methods A retrospective chart review was performed in patients with CD hospitalized for abscess confirmed by imaging between 2008 and 2016. Selection criteria included nonoperative management with intravenous antibiotics at the index hospitalization and a minimum of 2 years of follow up. Demographic, disease, and medication history were extracted from electronic medical records. Radiographic disease features were assessed by an expert abdominal radiologist, blinded to clinical data. The primary outcome was resection of the bowel segment involving the abscess within 2 years of index hospitalization. Cox proportional hazards regression and statistical methods were performed using SAS 9.4. Results Of the 121 patients meeting the selection criteria, 36.4% avoided surgery after 2 years of follow up. On adjusted multivariable analysis, disease-activity factors including bowel wall thickness (HR 3.08, 95% CL 1.20-6.21), disease length (HR 2.67, 95% CL 1.40-6.20), bowel dilation (HR 2.19, 95% CL 1.02-4.68), and abscess size of greater than 6 cm (HR 2.47, 95%CL 1.17-5.21) were independent risk factors for future surgery in patients not undergoing immediate bowel resection for abscess management. Biologic use and percutaneous drainage were not risk factors for ultimate surgical management. Conclusions Radiographic CD features and abscess size over 6 cm are predictors of ultimately requiring bowel resection. Radiographic measures may help stratify patients to immediate surgery or conservative management for intraabdominal CD-related abscesses.

Background and aims Vedolizumab (VDZ) is effective for Crohn's disease (CD) but costly and is slow to produce remission. Early knowledge of whether vedolizumab is likely to succeed is valuable for physicians, patients, and insurers. Methods Phase 3 clinical trial data on VZD for CD were used to predict outcomes. Random forest modeling on the training cohort was used to predict the outcome of corticosteroid-free biologic remission at week 52 on the testing cohort. Models were constructed using baseline data, or data through week 6 of VDZ therapy. Results The clinical trial included 594 subjects who received VDZ with baseline active inflammation [elevated C-reactive protein (>5 mg/L)]. Subjects with missing predictor variables (N = 120) or missing outcome data (N = 2) were excluded to produce a modeling dataset of 472 subjects. The Area Under the Receiver Operating Characteristic curve (AuROC) for corticosteroid-free biologic remission at week 52 using baseline data was only 0.65 (95% CI: 0.53 - 0.77), but was 0.75 (95% CI: 0.64 - 0.86) with data through week 6 of VDZ . Patients predicted to be in corticosteroid-free biologic remission at week 52 by the model achieved this endpoint 35.8% of the time, whereas patients predicted to fail only succeeded 6.7% of the time. Conclusions An algorithm using laboratory data through week 6 of VDZ therapy was able to identify which CD patients with baseline inflammation would achieve corticosteroid-free biologic remission on VDZ at week 52. A majority of patients can be identified by week 6 as very unlikely to achieve remission.

Background Inflammatory bowel disease (IBD) is a chronic disease characterized by unpredictable episodes of flares and periods of remission. Tools that accurately predict disease course would substantially aid therapeutic decision-making. This study aims to construct a model that accurately predicts the combined end point of outpatient corticosteroid use and hospitalizations as a surrogate for IBD flare. Methods Predictors evaluated included age, sex, race, use of corticosteroid-sparing immunosuppressive medications (immunomodulators and/or anti-TNF), longitudinal laboratory data, and number of previous IBD-related hospitalizations and outpatient corticosteroid prescriptions. We constructed models using logistic regression and machine learning methods (random forest [RF]) to predict the combined end point of hospitalization and/or corticosteroid use for IBD within 6 months. Results We identified 20,368 Veterans Health Administration patients with the first (index) IBD diagnosis between 2002 and 2009. Area under the receiver operating characteristic curve (AuROC) for the baseline logistic regression model was 0.68 (95% confidence interval [CI], 0.67–0.68). AuROC for the RF longitudinal model was 0.85 (95% CI, 0.84–0.85). AuROC for the RF longitudinal model using previous hospitalization or steroid use was 0.87 (95% CI, 0.87–0.88). The 5 leading independent risk factors for future hospitalization or steroid use were age, mean serum albumin, immunosuppressive medication use, and mean and highest platelet counts. Previous hospitalization and corticosteroid use were highly predictive when included in specified models. Conclusions A novel machine learning model substantially improved our ability to predict IBD-related hospitalization and outpatient steroid use. This model could be used at point of care to distinguish patients at high and low risk for disease flare, allowing individualized therapeutic management.

Background Evaluating structural damage using imaging is essential for the evaluation of small intestinal Crohn’s disease (CD), but it is limited by potential interobserver variation. We compared the agreement of enterography-based bowel damage measurements collected by experienced radiologists and a semi-automated image analysis system. Methods Patients with small bowel CD undergoing a CT-enterography (CTE) between 2011 and 2017 in a tertiary care setting were retrospectively reviewed. CT-enterography studies were reviewed by 2 experienced radiologists and separately underwent automated computer image analysis using bowel measurement software. Measurements included maximum bowel wall thickness (BWT-max), maximum bowel dilation (DIL-max), minimum lumen diameter (LUM-min), and the presence of a stricture. Measurement correlation coefficients and paired t tests were used to compare individual operator measurements. Multivariate regression was used to model identification of strictures using semi-automated measures. Results In 138 studies, the correlation between radiologists and semi-automated measures were similar for BWT-max (r = 0.724, 0.702), DIL-max (r = 0.812, 0.748), and LUM-min (r = 0.428, 0.381), respectively. Mean absolute measurement difference between semi-automated and radiologist measures were no different from the mean difference between paired radiologists for BWT-max (1.26 mm vs 1.12 mm, P = 0.857), DIL-max (2.78 mm vs 2.67 mm, P = 0.557), and LUM-min (0.54 mm vs 0.41 mm, P = 0.596). Finally, models of radiologist-defined intestinal strictures using automatically acquired measurements had an accuracy of 87.6%. Conclusion Structural bowel damage measurements collected by semi-automated approaches are comparable to those of experienced radiologists. Radiomic measures of CD will become an important new data source powering clinical decision-making, patient-phenotyping, and assisting radiologists in reporting objective measures of disease status.

BACKGROUND Corticosteroids are effective rescue therapies for patients with inflammatory bowel disease (IBD), but have significant side effects, which may be amplified in the growing population of elderly patients with IBD. We aimed to compare the use of steroids and steroid-sparing therapies (immunomodulators and biologics) and rates of complications among elderly (≥65) and younger patients in a national cohort of veterans with IBD. METHODS We used national Veterans Health Administrative data to conduct a retrospective study of veterans with IBD between 2002 and 2010. Medications and the incidence of complications were obtained from the Veterans Health Administrative Decision Support Systems. Multivariate logistic regression accounting for facility-level clustering was used to identify predictors of use of steroid-sparing medications. RESULTS We identified 30,456 veterans with IBD. Of these, 94% were men and 40% were more than 65, and 32% were given steroids. Elderly veterans were less likely to receive steroids (23.8% versus 38.3%, P < 0.001) and were less likely to be prescribed steroid-sparing medications (25.5% versus 46.9%, respectively, P < 0.001). In multivariate analysis controlling for sex, age

Article first published online 27 March 2015., Background: Decisions between medical and surgical management of Crohn's disease (CD) incorporate risk assessments for potential complications of each therapy. Analytic morphomics is a novel method of image analysis providing quantifiable measurements of body tissue composition, characterizing body fat more comprehensively than body mass index alone. The aim of this study was to determine the risk factors associated with postoperative complications in CD, incorporating fat composition analysis using analytic morphomics. Methods: We performed a retrospective review of adults undergoing bowel resection for CD between 2004 and 2011 at a single center. Computed tomography obtained within 30 days prior to surgery underwent morphomic analysis for fat characterization. Postoperative infectious complications were defined as the need for a postoperative abdominal drain, intravenous antibiotics, or reoperation within 30 days. Bivariate and multivariate analyses using logistic regression were used to generate a prediction model of infectious complications. Results: A total of 269 subjects met selection criteria; 27% incurred postoperative infectious complications. Bivariate analysis showed hemoglobin, albumin, surgical urgency, high-dose prednisone use, and subcutaneous-to-visceral fat volume distribution as predictors of complications. Body mass index, anti-tumor necrosis factor alpha therapies, and immunomodulator use were not predictors of complication. Multivariate modeling demonstrated a c-statistic of 0.77 and a negative predictive value of 81.1% with surgical urgency (odds ratio = 2.78; 95% confidence interval, 1.46–6.02; P = 0.004), subcutaneous-to-visceral fat distribution (odds ratio = 2.01; 95% confidence interval, 1.20–3.19; P = 0.006), and hemoglobin (odds ratio = 0.69; 95% confidence interval, 0.55–0.85; P = 0.001) as predictors of infectious complication. Conclusions: Fat subtype and distribution are predictive of postoperative infectious complications after bowel resection for CD. Analytic morphomics provides additional body composition detail not captured by body mass index.