What’s New in Gastrointestinal Imaging? – May 2022

1 month ago

Richard Fagan – ACORE May

Colonoscopy versus catheter angiography for lower gastrointestinal bleeding after localization on CT angiography. Tse JR, Felker ER, etal. J Am Coll Radiol. 2022;19:513-520


This study compared outcomes of catheter angiography (CA) versus colonoscopy after successful CT angiographic (CTA) localization of lower GI bleed. The primary outcomes that the authors assessed was confirmation of active bleeding during CA or colonoscopy. Secondary outcomes were ability to deliver hemostatic therapy, time to procedure, rebleeding time and adverse outcomes. The study group included 44 patients who underwent catheter angiography, and 27 who underwent colonoscopy. The catheter angiography group showed higher yield of confirmation of bleeding site with 55% compared to colonoscopy with 26%. The therapeutic yields were not significantly different with catheter angiography achieving hemostasis in 70% versus 56% with colonoscopy. Catheter angiography had a significantly faster time to procedure with 5.1 +/- 3.4 hours, compared to colonoscopy with 15.5 +/- 13.6 hours. Shorter time to procedure was found to be the only statistically significant predictor of confirmation and therapeutic yield. The authors note that a 1 hour increase in time to intervention was associated with an 11% decrease in confirmation yield and an 8% decrease in therapeutic yield. The type of procedure, patient hemoglobin level, and hemodynamic stability were not significant predictors of confirmation of therapeutic yield. There was no statistically significant difference found for adverse events or rebleeding between patients who underwent catheter angiography and patients who underwent colonoscopy. This study specifically found no difference in renal injury following catheter angiography compared to colonoscopy.


Detection of high-risk sessile lesions: multitarget stool DNA versus CT colonography. Deiss-Yehiely  N, Graffy PM, Et al. Amer Journal Roentgenology. 2022; 218:670-677


Sessile serrated lesions account for 15-30% of all colorectal cancers, yet diagnosis of these lesions is a challenge given their flat contour and predominantly right sided location within the colon. These lesion when large or dysplastic are high risk lesions for progression to colorectal cancer. While colonoscopy is the gold standard for detection and removal of pre-cancerous lesions, many patients desire a noninvasive screening test. Currently, these noninvasive screening tests for high risk sessile serrated lesions (SSLs) have unclear benefit. In this study the authors aimed to compare the ability of the multitarget stool DNA (mt-sDNA) test and CT colonography to detect high risk SSLs. 7974 asymptomatic patients who underwent noninvasive CRC screening at a single center were retrospectively studied. The primary outcome was the histologically proven test yield of high risk SSLs, defined as lesions 10 mm or greater, or with dysplasia. The authors found that 13.1% of mt-sDNA tests yielded a positive result, while 12.2% of CT colonography exams using a threshold size of 6-mm, and 6.5% of CT colonography exams using a threshold of 10-mm yielded a positive result. The PPV for mt-sDNA was found to be 5.5%, versus 14.4% for 6-mm and 25.9% for 10-mm cutoff with CT colonography. The overall screening yield of high risk SSLs was found to be 0.7% for mt-sDNA versus 1.7% and 1.6% for CT colonography at 6-mm and 10-mm respectively. These results are significant because they suggest that CT colonography screening detects high risk SSLs at a higher rate than mt-sDNA screening, and patients who opt for noninvasive colorectal cancer screening should be informed of this screening option.


Saline-aided ultrasound versus upper gastrointestinal series in neonates and infants with suspected upper gastrointestinal obstruction: a prospective multicenter comparative study. Chen S, Chen D, et al. Amer Journal of Roentgenology 2022; 218:526-533


Upper gastrointestinal series are the standard for diagnosis of upper GI obstruction in neonates and infants, but these studies lead to concern over radiation exposure to developing organs. Ultrasonography can be used to diagnose many GI pathologies in the pediatric population, and when used with saline as a contrast medium, ultrasonography can be used to follow fluid flow and diagnose and grade obstructions. The authors of this study aimed to assess the diagnostic performance of saline-aided ultrasound for the presence, level, and cause of upper GI obstructions in neonates and infants compared to upper GI series. 209 neonates with suspected upper GI obstruction underwent both saline-aided ultrasound and upper GI series. Patients were excluded if US showed evidence of malrotation or pyloric stenosis. Saline-aided ultrasound showed strong interobserver agreement for presence (k=0.87) and level of obstruction (k=0.85). For presence of upper GI obstruction accuracy, sensitivity, and specificity were 94.7%, 98.4%, and 89.4% for saline-aided ultrasound, compared with 89.5%, 95.2%, and 81.2% for upper GI series. For obstruction level accuracy, sensitivity, and specificity were 90.3%, 97.2%, and 56.3% for saline-aided ultrasound compared with 87.1%, 92.6%, and 50.0% for upper GI series. Only the accuracy for presence of obstruction showed a significant difference between the two groups. Saline-aided ultrasound was also able to detect cause of upper GI obstruction with accuracy ranging from 75.0% to 95.2%. These findings demonstrate that saline-guided ultrasound has high diagnostic performance in upper GI obstruction and compares favorably with upper GI series with the added benefit of a reducing exposure to ionizing radiation.


Discriminating rectal cancer grades using restriction spectrum imaging. Xiong Z, Geng Z, et al. Abdominal Radiology 2022; 47:2014-2022


Accurate tumor localization and grading are key for disease management in rectal cancer. Currently, grading rectal cancer requires biopsy, which is costly and carries risk of adverse events. This presents the need for noninvasive imaging techniques for grading tumors. Diffusion-weighted MRI utilizes features of water molecule interactions to characterize tissue microstructure, and has shown utility as an imaging biomarker for tumor aggressiveness and treatment response. Several different parameters for DWI are utilized, including diffusion kertosis imaging (DKI) and apparent diffusion coefficient (ADC) parameters. Restriction spectrum imaging is a novel technique utilizing diffusion weighted imaging that separates water diffusion into three compartments based on diffusivity value. The restricted compartment is highly specific for aggressive tumors given the high cellularity of these tumors. This study analyzes the performance of a compartment RSI model for patients with differing histopathological grades of rectal cancer (low grade being G1 and G2 and high grade being G3). The RSI model was then compared to diffusion kertosis imaging (DKI) and apparent diffusion coefficient (ADC) parameters. 58 patients with differing grades of rectal cancer as confirmed by biopsy were analyzed using a three-compartment RSI model and compared to ADC and DKI parameters on DWI. The restricted compartment from RSI was significantly correlated with tumor grade, with significant difference between G1 and G3, and between G2 and G3, as well as a significant difference between low grade (G1+G2) and high grade (G3) with an AUC of 0.753. When compared with DKI and ADC, RSI showed a significantly superior performance. These findings demonstrate the potential of RSI  as a powerful tool in the management of rectal cancer.

Bowel peristalsis artifact on dual-energy CT: In vitro study on influence of different dual-energy CT platforms and enteric contrast agents. Obmann M, Sun Y, et al. Amer Journal Roentgenology. 2022; 218:290-299.


Dual energy CT is a powerful tool for bowel wall assessment, however, the intrinsic peristalsis of bowel wall can introduce artifact that limits the use of CT. This study aimed to evaluate the influence of different dual energy CT scanners and enteric contrast agents on the severity of bowel wall peristalsis in vitro. The authors used a 3-cm diameter corrugated hollow tube which was constantly oscillated to simulate bowel peristalsis. This construct was scanned on four different DECT platforms (spectral detector, rapid peak kilovoltage switching, split filter, and dual source) while filling the tube with air, water, iodinated, and an experimental dark contrast. The image reconstructions were then rated for artifact severity. For 120-kVp-like images, mean peristalsis artifact scores were lower for split-filter and dual-source scanners than for spectral-detector and rapid kilovoltage-switching scanners. Compared with 120-kVp images, peristalsis artifacts on iodine images were less severe for spectral-detector and rapid-kilovoltage-switching but more severe for dual-source and split-filter systems. As for the contrast mediums, the experimental dark contrast was rated less severe for peristalsis artifact than the other bowel contrast agents. These findings can help determine the best protocol and equipment for reducing peristalsis artifact in clinical imaging.



  • Share