Methods: This article describes the development a computed-tomography enterography (CTE)–based radiomic model (RM). This retrospective multicenter study included 167 CD patients who underwent preoperative CTE and bowel resection. 1454 radiomic features were extracted from venous-phase CTE and a machine learning–based RM was developed based on the reproducible features using logistic regression. The RM was validated in an independent external test cohort recruited from 3 centers.
In the training cohort, the area under the ROC curve (AUC) of RM for distinguishing moderate–severe from none–mild intestinal fibrosis was 0.888.
In the test cohort, the RM had an AUC of 0.816.
RM was more accurate than visual interpretations by either radiologist (radiologist 1, AUC = 0.554; radiologist 2, AUC = 0.598; both, P < .001) in the test cohort
My take: This CT approach with RM allowed for accurate characterization of intestinal fibrosis in CD. The images look pretty cool too.
With new tools at our disposal in diagnosing inflammatory bowel disease, we need to decide how and when to use them. Potential new modalities include stool inflammatory markers, video capsule endoscopy, CT enterography (CTE), and MR enterography (MRE). Several studies have shown that the information that these studies yield may change management. The latest of these studies (Inflamm Bowel Dis 2012; 18: 219-25) looked at how the knowledge of CTE effected management and physician confidence with Crohn’s disease.
The authors prospectively assessed 273 patients with established or suspected Crohn’s disease. In their analysis, 70 patients (48%) of established cases had altered management because of CTE and 69 (54%) of suspected cases. These changes were considered to be independent of clinical, serological or histologic findings. Changes included medication modification in 45 (16.2%), excluding Crohn’s disease in 46 (16.8%), surgery referral in 10 (3.7%), alternate diagnosis established in 9 (3.2%), & canceling surgery in 7 (2.6%). The authors considered excluding active small disease as an important management plan change; this occurred in 18 patients (6.6%).
The authors state that their current practice is to use MRE for serial imaging rather than CTE, to minimize risks from radiation; though CTE is often the initial imaging.
My take on this article is that information from imaging often increases the certainty about the diagnosis and gives a more complete picture of the severity. It is likely that more information leads to more aggressive therapy. At the same time, in pediatric gastroenterology, the trend towards using more effective therapy earlier in the course of the disease has developed even in the absence of extensive imaging (see previous: Only one chance to make first impression). Whether more imaging in pediatric patients would be worthwhile is not known.
-JPGN 2008; 47: 31. Capsule endoscopy may reclassify pediatric IBD
-NEJM 2010; 363: 1, 4. Safety of CT. Can have overdose of radiation and even standard doses could cause complications. Also, a big issue is downstream unnecessary testing due to incidental findings.
-Clin Gastro 2008; 6:283. Use of CT enterography.
-JPGN 2010; 51: 603. MRE for suspected IBD. Useful in Crohn’s disease.
-IBD 2004;10: 278-285. WCE for Crohn’s (review) Capsule can help differentiate UC from Crohn’s.