What’s new in Gastrointestinal Imaging – July 2020

DIAGNOSTIC VALUE OF SPECIFIC MODALITIES

Role of Ultrasound and CT in the Workup of Right Upper Quadrant Pain in Adults in the Emergency Department: A Retrospective Review of More Than 2800 Cases

Hiatt KD, Ou JJ, Childs DD. Role of Ultrasound and CT in the Workup of Right Upper Quadrant Pain in Adults in the Emergency Department: A Retrospective Review of More Than 2800 Cases. Am J Roentgenol [Internet]. 2020 Jun 1 [cited 2020 Jun 20];214(6):1305–10.

https://www.ajronline.org/doi/10.2214/AJR.19.22188

Keywords: acute cholecystitis, right upper quadrant pain, ultrasound, CT, incidental findings

Pediatric patients with right upper quadrant pain undergo ultrasound for further workup. However, when adult patients present with right upper quadrant pain to the emergency department, the concern for acute cholecystitis can be addressed with either US or CT. Hiatt et al performed a retrospective review of the imaging associated with 2859 such adult patient presentations over a 5-year period; 18% met the definition of cholecystitis. There was no statistically significant difference between US (sensitivity 61%, specificity 91%) and CT (sensitivity 55%, specificity 92%). However, CT showed an acute non-gallbladder abnormality in 103 of 322 cases (32%), while US could only detect an acute non-gallbladder abnormality otherwise missed on CT in 1 of 238 cases (0.4%). Hiatt et al concluded that CT is noninferior to US in diagnosis of cholecystitis in adult patients and additionally offers the advantage of depicting acute non-gallbladder pathologies.

Diagnostic Approach to Benign and Malignant Calcifications in the Abdomen and Pelvis

Zulfiqar M, Shetty A, Tsai R, Gagnon MH, Balfe DM, Mellnick VM. Diagnostic approach to benign and malignant calcifications in the abdomen and pelvis. Radiographics [Internet]. 2020 May 1 [cited 2020 Jun 21];40(3):731–53. https://pubs.rsna.org/doi/abs/10.1148/rg.2020190152

Keywords: calcifications, RadioGraphics, diagnostic approach

Zulfiqar et al, among whom Dr. Vincent Mellnick is one of our very own senior editors, present the imaging features of various abdominopelvic calcified lesions in this RadioGraphics article and a diagnostic approach to these calcifications. These calcified lesions can be benign, premalignant, or malignant. Zulfiqar et al address how to diagnose these calcifications based on clinical information (i.e. recent treatment), location (mesenteric, peritoneal, retroperitoneal, organ-based, vascular, and musculoskeletal), and morphology (round, curvilinear, sheet-like).

Colon Adenocarcinoma Associated With Clostridium Septicum Endophthalmitis Detected on ¹⁸F-FDG PET/CT

Plouznikoff N, Artigas C, Woff E, Ene D, Flamen P. Colon Adenocarcinoma Associated With Clostridium Septicum Endophthalmitis Detected on 18F-FDG PET/CT. Clin Nucl Med [Internet]. 2020 Jun 1 [cited 2020 Jun 21];45(6):459–60.

http://journals.lww.com/10.1097/RLU.0000000000003025

Keywords: PET, Clostridium septicum, correlations, colon cancer

Plouznikoff et al show a case of occult cecal adenocarcinoma detected on PET/CT after the patient presented with endogenous endolphthalmitis. Endogenous endolphthalmitis due to Clostridium septicum is a rare ocular infection, but is highly associated with GI and hematological malignancies (other infections with C. septicum such as aortitis have also been described to be associated with colorectal cancer). Plouznikoff et al suggest that PET/CT should be considered in all patients with endogeneous endolphthalmitis to exclude occult malignancy.

Reduced Field-of-View Diffusion-Weighted Imaging in Histological Characterization of Rectal Cancer: Impact of Different Region-of-Interest Positioning Protocols on Apparent Diffusion Coefficient Measurements

Yang P, Xu C, Hu X, Shen Y, Hu D, Kamel I, et al. Reduced Field-of-View Diffusion-Weighted Imaging in Histological Characterization of Rectal Cancer: Impact of Different Region-of-Interest Positioning Protocols on Apparent Diffusion Coefficient Measurements. Eur J Radiol. 2020 Jun 1;127:109028.

https://www.sciencedirect.com/science/article/pii/S0720048X20302175

Keywords: rectal cancer, MRI, ROI methods, diagnostic value

Histological characterization of rectal cancer can be done noninvasively by using a region-of-interest (ROI) method on preoperative rectal cancer MRI. This is done using the reduced-field-of-view (rDWI) technique on the apparent diffusion coefficient (ADC) map provided in the MRI examination. Per Yang et al, to date, no investigations had reported the variable influence of different region-of-interest (ROI) methods on the ultimate histological characterization of rectal cancer. They compared four ROI methods (whole tumor volume [WTV], single-slice [SS], three-slices observer-based sampling [TSOB], and three-slices predefined sampling [TSPD]) in 49 patients. They also collected the corresponding times taken for these ADC measurements under each ROI protocol. Results showed that the WTV method of ROI had the best diagnostic value for assessment of the well-differentiated tumors, T2 stage tumors, and N0 status. However, WTV required the longest measurement time (66 s), while the SS method had the shortest measurement time (13 s). Yang et al concluded that while WTV has the longest measurement time, ADC measurements based on rDWI using the WTV method had the best diagnostic value for histological characterization of rectal cancer.

IMAGING OF SPECIFIC REGIONS OF THE ABDOMEN

Morison’s pouch: anatomical review and evaluation of pathologies and disease spread on cross-sectional imaging

Nassar S, Menias CO, Nada A, Blair KJ, Shaaban AM, Mellnick VM, et al. Morison’s pouch: anatomical review and evaluation of pathologies and disease spread on cross-sectional imaging. Abdom Radiol [Internet]. 2020 Jun 11 [cited 2020 Jun 20];1–12.

http://link.springer.com/10.1007/s00261-020-02597-1

Keywords: anatomy, peritoneum, Morison’s pouch

Nassar et al, including our senior editors Dr. Mellnick and Dr. Elsayes, describe the complex anatomy of the peritoneum in this paper. They focus on reviewing the anatomy of Morison’s pouch, and the incidence and development of associated diseases, with the goal of providing the radiologist with the tools necessary to diagnose these pathologies.

Multimodality Imaging of Gastric Pathologic Conditions: A Primer for Radiologists

Anderson AC, Millet JD, Manganaro MS, Wasnik AP. Multimodality imaging of gastric pathologic conditions: A primer for radiologists [Internet]. Vol. 40, Radiographics. Radiological Society of North America Inc.; 2020 [cited 2020 Jun 21]. p. 707–8.

https://pubs.rsna.org/doi/abs/10.1148/rg.2020190065

Keywords: RadioGraphics, stomach, diagnostic approach

Anderson et al show many imaging features of the non-postsurgical pathologic conditions of the stomach, including neoplastic, infectious, inflammatory, and emergent diseases in this review paper in RadioGraphics. Of note, this article is part of the “fundamentals” section of RadioGraphics.

Imaging of Abdominal Wall Masses, Masslike Lesions, and Diffuse Processes

Ballard DH, Mazaheri P, Oppenheimer DC, Lubner MG, Menias CO, Pickhardt PJ, et al. Imaging of abdominal wall masses, mass-like lesions, and diffuse processes. Radiographics [Internet]. 2020 May 1 [cited 2020 Jun 21];40(3):684–706.

https://pubs.rsna.org/doi/abs/10.1148/rg.2020190170

Keywords: abdominal wall mass, diagnostic approach, RadioGraphics

Ballard et al, including our senior editors Dr. Mellnick and Dr. Lubner, describe the variety of abdominal wall masses that can be seen incidentally on cross-sectional imaging in this RadioGraphics article. They additionally present a diagnostic algorithm to help distinguish between the different types of masses that may be seen. The highest level of this diagnostic algorithm involves differentiating whether the mass is a hernia versus fat-containing (lipomas), cystic (abscesses, seromas, hematomas), or solid (desmoid tumors, sarcomas, endometriomas, and metastases). Diffuse disease processes may also be present and have a separate diagnostic algorithm.

USE OF ALGORITHMS AND MACHINE LEARNING IN ABDOMINOPELVIC IMAGING

Reducing Radiation Dose and Contrast Medium Volume With Application of Dual-Energy CT in Children and Young Adults

Tabari A, Gee MS, Singh R, Lim R, Nimkin K, Primak A, et al. Reducing Radiation Dose and Contrast Medium Volume With Application of Dual-Energy CT in Children and Young Adults. Am J Roentgenol [Internet]. 2020 Jun 1 [cited 2020 Jun 20];214(6):1199–205.

https://www.ajronline.org/doi/10.2214/AJR.19.22231

Keywords: CT, radiation dose, image quality

Dual-source, dual-energy CT (DS-DECT) was shown to substantially reduce radiation dose and contrast volume compared with single-energy CT (SECT) performed on the same scanners (Somatom Flash or Force, Siemens Healthineers) in this retrospective study. Tabari et al studied 85 children and young adults who underwent contrast-enhanced DS-DECT (with CARE dose 4D), and 66 matched patients (across multiple demographic criteria) who underwent SECT (with CARE dose 4D and CARE kV). Results showed size-specific dose estimates (SSDE) were significantly reduced with DS-DECT to 4.0 mGy (from 6.1 mGy) for the chest and to 5.0 mGy (from 8.3 mGy) for the abdominopelvic region. Contrast material volume for DS-DECT was 19-22% lower when compared with SECT. Image quality was acceptable in all DECT examinations.

 Single-Energy Retrospective Metal Artifact Reduction Using Adaptive Thresholding for Metal Implants in the Abdomen and Pelvis

Dommaraju S, Nakhaei M, Zhang D, Camacho A, Boos J, Appel E, et al. Single-Energy Retrospective Metal Artifact Reduction Using Adaptive Thresholding for Metal Implants in the Abdomen and Pelvis. J Comput Assist Tomogr [Internet]. 2020 May 1 [cited 2020 Jun 20];44(3):443–9.

http://journals.lww.com/10.1097/RCT.0000000000001013

Keywords: metal artifact, CT, image quality

Dommaraju et al performed a prospective study with 90 patients who had various metals detected on abdomen/pelvis CT (Canon Medical, Aquilion ONE, and PRIME) over a 15-month period. Retrospectively, they applied a single-energy metal artifact reduction (SEMAR) algorithm and compared the density in the near and far-fields with respect to these metals with or without SEMAR application. Density standard deviation with or without SEMAR application was calculated to represent objective artifact severity. Additionally, two human observers were trained to evaluate the subjective severity of artifacts on a five-point scale. Results showed that the objective artifact severity significantly decreased in the near field of high-density metallic implants (density SD of 204 HU reduced to 66 HU after SEMAR) and also decreased in the near-field of low-density metallic implants (density SD of 86 HU reduced to 49 HU after SEMAR). Artifact severity in the far-field for all metallic implants was objectively similar after SEMAR. However, subjectively, artifact severity decreased for both high- and low-density metallic implants in both the near- and far-fields. Dommaraju et al concluded that retrospective application of the SEMAR algorithm could reduce both subjective and objective artifact severity from low- and high-density metals.

Fully Automated Segmentation of Connective Tissue Compartments for CT-Based Body Composition Analysis: A Deep Learning Approach

Nowak S, Faron A, Luetkens JA, Geißler HL, Praktiknjo M, Block W, et al. Fully Automated Segmentation of Connective Tissue Compartments for CT-Based Body Composition Analysis. Invest Radiol [Internet]. 2020 Jun 1 [cited 2020 Jun 21];55(6):357–66.

http://journals.lww.com/10.1097/RLI.0000000000000647

Keywords: CT, deep learning, segmentation

Nowak et al sought out to use retrospective CT examinations from 1143 patients to develop a fully automated and accurate segmentation tool for connective tissue compartments using the open-source convolutional neural network (CNN) DeepMedic. They state that this tool can be useful in providing information regarding body composition in patients with malignancies, such as their degree of cachexia. Since detection of body compartments, especially subcutaneous and visceral fat, can be affected by anasarca, Nowak et al chose cohorts of patients undergoing transcatheter aortic valve implementation and patients with cirrhosis to ensure a robust model. The data was divided into a training subset (70%), validation subset (15%), and test data (15%). Semi-manual threshold-based segmentation technique was used for ground truth labeling. The algorithm was tested further on 100 randomly selected patients who underwent CT for routine diagnostic purposes and on 50 additional native CT examinations. Segmentation of the subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), skeletal muscle (SM) compartments were performed. Results showed excellent generalizability to both the routine CT diagnostic patients and on native CT scans. Nowak et al concluded that the open-source CNN DeepMedic can perform fully automated determination of body composition.