thomps132@gmail.com

1) Know the ins and outs of performing emergency gastrointestinal examinations. 2) Know the common presentations of emergency esophageal and abdominal disorders. 3) Know how to diagnose the common emergency gastrointestinal disorders demonstrated on fluoroscopic examinations.

Nicolas.Murray@vch.ca

Evaluate the diagnostic performance and efficiency of MRI in suspected acute appendicitis compared to ultrasound (US) and computed tomography (CT).

Single institution, IRB-approved, retrospective study of adult patients presenting to emergency department with suspected acute appendicitis from May 2017 to May 2018. Diagnostic characteristics of US, MRI, and CT were analyzed using a contingency table. Diagnostic efficiency was examined by average patient wait time, defined between times of initial imaging and final management decision.

599 patients met the eligibility criteria, with 445 US (54.7%), 137 MRI (16.9%), and 231 CT scans (28.4%) performed. Sensitivity, specificity and diagnostic yield of MRI were respectively 91.7% (95%CI, 73.0%-99.0%), 85.0% (95%CI, 77.0%-91.0%), and 88.3% (95%CI, 81.9%-92.7%), not significantly different than CT with respective values of 94.3% (95%CI, 84.3%-98.8%), 88.8% (95%CI, 83.2%-93.0%) and 93.1% (95%CI, 89.0%-95.7%). Using an intention-to-diagnose approach, diagnostic properties of US were significantly lower than both MRI and CT (p<0.01) with sensitivity of 61.5% (95%CI, 51.5%-70.9%), specificity of 18.2% (95%CI, 14.2%-22.7%), and diagnostic yield of 29.7% (95%CI, 25.6%-34.1%). Mean wait time for patients undergoing MRI as initial investigation (n=21, 3.5%, 100.6 minutes) was not significantly different from patients examined initially by CT (n=133, 22.2%, 104.3 minutes, p=0.78) or US (n=238, 39.7%, 125.6 minutes, p=0.29). All imaging routes where patients experienced multiple modalities had significantly longer wait times than routes involving one modality (p<0.01).

Diagnostic performance of MRI is comparable to CT and superior than US. With favourable patient wait times, MRI can be considered as initial investigation modality in suspected acute appendicitis.

The greater role of MRI as first-line investigation modality in suspected acute appendicitis will reduce exposure to ionizing radiation without compromising diagnostic performance or timeliness.

haroon.radiology@gmail.com

To assess the utility of intraluminal air in an inflamed, apparently non-perforated appendix in predicting gangrenous changes or occult perforation. Determine, if obstructive appendicolith has an added value in predicting the same.

This retrospective study was done on adult patients (n=554) of histopathologically proven appendicitis who underwent enhanced MDCT prior to surgery, presenting at our hospital over a consecutive period of 3 years. Patients with obvious CT signs of perforation were excluded to create a cohort of acute uncomplicated appendicitis. These CT were reviewed by an Emergency Radiology Fellow and 2 Emergency Radiologists for presence or absence of intraluminal gas and obstructive appendicoliths. These findings were compared with surgical/pathological results regarding presence or absence of gangrenous/perforated appendicitis. Statistical analysis was performed with the help of contingency tables and sensitivity, specificity, positive and negative predictive values were determined and correlation was tested with Chi-squared test and p value < 0.05 was considered statistically significant.

Of the total 554 cases of acute uncomplicated appendicitis on imaging, 130 had intraluminal gas (90 gangrenous), 178 had obstructive appendicoliths (74 gangrenous) and 66 were with both gas and appendicoliths (50 gangrenous). Sensitivity, specificity, positive and negative predictive values for intraluminal gas and presence or absence of gangrene were 69%, 90%, 69% and 90% respectively. These values for obstructive appendicolith were 42%, 72%, 42% and 72% respectively. These values for the presence of both intraluminal gas and appendicolith were 39%, 96%, 75% and 84% respectively. These values for the presence of either intraluminal gas or appendicoliths were 86%, 69%, 46% and 94% respectively. All these results were significant with p value < 0.05.

Presence of intraluminal gas in otherwise acute uncomplicated appendicitis on imaging is a reliable sign of underlying gangrenous changes or image-occult perforation. Presence of obstructive appendicolith, although less reliable sign as an independent risk factor as compared to intraluminal gas, it notably adds to the predective value.

These CT signs are reliable in predicting the risk of gangrene and perforation and can help surgeons to avoid delays in surgery thereby reducing incidence of complications.

tugce.trabzonlu@northwestern.edu

To evaluate the performance of computed tomography (CT) without oral and intravenous (iv) contrast material for the diagnosis of acute diverticulitis by comparing dual energy CT (DECT) contrast enhanced and virtual non-contrast (VNC) images.

In this retrospective analysis, we reviewed CT studies with oral and IV contrast obtained with DECT scanner for abdominal pain. Cohort included 153 patients with 306 sets of CT images with a radiological diagnosis of acute diverticulitis (n=76) and control cases without evidence of diverticulitis (n=77) scanned between October 2018 and March 2019. In the first session, the virtual non-contrast images were randomized and analyzed for the presence of diverticulitis. The findings and presence of complication (perforation, abscess formation or fistula) were also noted. In the second session, true contrast enhanced images were randomized and analyzed. Diagnostic performance of VNC images were compared with contrast enhanced CT studies. Sensitivity, specificity and accuracy were calculated.

Out of 153 patients, 76 (49.7%) had acute diverticulitis and 77 (50.3%) did not have findings of acute diverticulitis on contrast enhanced computed tomography (CECT). 18 of 76 (23.7%) patients with acute diverticulitis had findings of complicated diverticulitis on CECT. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of VNC images were 96.1% (95% Cl= 88.9-99.2%), 97.4% (95% Cl= 90.9-99.7%), 97.3% (Cl= 90.3-99.3%), 96.2% (Cl= 89.2- 98.7%) and 96.7% (95% Cl= 92.5-98.9%) respectively. The complications of acute diverticulitis was detected in 11 of 18 (61.1%) patients with VNC images.

When compared to routine CT imaging with iv and oral contrast, non-contrast images have high diagnostic accuracy for acute colonic diverticulitis. However, for the assessment of the signs of complicated diverticulitis, non-contrast CT had much lower diagnostic performance.

Non-contrast CT can be beneficial for diagnosing uncomplicated diverticulitis. However, the use of contrast remains necessary when complicated diverticulitis is present.

To assess the prevalence and diagnostic yield of CT in the detection of diverticulitis and alternative diagnoses (AD) in a large cohort of patients with suspected colonic diverticulitis (CD).

We retrospectively included 1069 patients (560 women) undergoing CT for the evaluation of suspected CD. The final clinical diagnosis derived from the discharge report was used to determine the prevalence of CD and AD and to calculate the diagnostic accuracy of CT. Differences in the prevalence of diagnoses by age (<45;45-69;>=70 y/o) were compared using Cochran-Armitage test with a p-value <0.05 indicating statistical significance.

Prevalence of CD was 52.4% (561/1069) and of AD 40% (427/1069). In the remaining 7.6% (81/1069) no final clinical diagnoses was established. Overall, CT had a sensitivity and specificity of 99.1%/99.8% for diagnosing CD and for AD 92.7%/96.4%, respectively. The prevalence of diverticulitis was significantly lower in patients >=70 y/o (43%; 128/298) when compared with patients <45 y/o (54.1%;100/185) and 45-69 y/o (56.8%;333/586) (p<=0.0004). The most frequent alternative diagnoses were appendicitis (12.6%; 54/427), infectious colitis (10.5%;45/427), infectious enteritis (8.2%;35/427), urolithiasis (6.1%;26/427), and pyelonephritis (4.9%;21/427). Prevalence of specific AD varied significantly according to age (p<0.05). Appendicitis was significantly more frequent in patients <45 y/o (5.4%;10/185), whereas ischemic colitis, hemorrhage and pneumonia were more frequent in patients >=70 y/o. In the latter group colorectal carcinoma was also a frequent AD (10/298).

In the clinical setting of suspected diverticulitis the prevalence of acute diverticulitis and alternative diagnoses varies according to age. CT provides high diagnostic accuracy in the diagnosis of both, diverticular disease and alternative conditions.

Clinicians must be aware that in about 40% of patients with suspected diverticulitis alternative diagnoses are the causes for their symptoms, and that there is an age-related prevalence of AD.

moshiri@uw.edu

1) Learn essential criteria for diagnosis of a normal viable first trimester pregnancy. 2) Learn essential criteria for differentiating an ectopic pregnancy from intrauterine pregnancy, and various ectopic pregnancies. 3) Learn appropriate use of such terms as 'pregnancy of unknown location', findings suspicious for early pregnancy failure', etc as outline by SRU lexicon and criteria.

Imaging evaluation of first trimester pregnancy especially in an emergent setting can pose dilemmas since in early pregnancy a gestational sac may not be clearly visible. Differentiating an IUP from an ectopic pregnancy is crucial as the latter requires clinical intervention. There is some overlap of serum BhCG levels with IUP, ectopic pregnancy, and spontaneous pregnancy loss. in 2012, SRU consensus panel published their agreed upon criteria and lexicon for reporting first trimester ultrasound exams. We will review the clinical application of these criteria and the lexicon, and review appearance of various types of ectopic pregnancies.

savvas.nicolaou@vch.ca

1) Explain the principles of Dual Energy CT/Spectral imaging. 2) Describe and apply 3-material decomposition. 3) Evaluate application of 3-material decomposition in select cases (organ perfusion in trauma, bowel ischemia, active bleeding, renal stone analysis).

elina.khasanova@vch.ca

To assess the impact of virtual monoenergetic imaging (VMI) and color-coded iodine-overlay images (IOI) on reader confidence and image quality (IQ) in the detection of hypoperfused bowel compared to simulated 120-kVp images (s-120-kVp).

Institutional review board approval was obtained. Acute bowel ischemia was reported in 80 patients imaged with triphasic CT studies with the portal venous phase acquired with dual energy analysis (90-150 snVp; 3rd generation dual source CT) from 01/02/2016 to 31/12/2018. Of 80 patients, 26 (33%) had bowel ischemia confirmed intra-operatively, 11(14%) deceased within 72 hours, 43 (53%) did not qualify for surgery. S-120-kVp, VMI (40, 50, 60 keV) and IOI (40%, 50%, 60% of iodine overlay color-coded saturation) datasets were created for each patient. Quantitative assessment (HU and CNR) on ischemic bowel, normal bowel, and portal vein was performed only on the surgically proven cases (n=26). Two emergency radiologists independently evaluated subjective image quality (IQ) and diagnostic confidence (DC). Time-to-diagnosis (TTD) was recorded on VMI and IOI datasets with the highest IQ and DC and s-120-kVp dataset. One-way ANOVA and Kruskal-Wallis/Wilcoxon rank sign tests were used for statistical analysis.

There was a significant increase in absolute attenuation difference between normal and ischemic bowel in 40, 50, 60 keV datasets (mean±SD 66±5.3, 62±4.3, 54±5.5 HU) compared to s-120-kVp (38±5.6HU). Both readers deemed 50-keV as the best VMI dataset for subjective IQ including image sharpness and resolution (p= 0.0017), DC (p= 0.0003). IOI-50% demonstrated subjective IQ (p= 0.0021) and DC (p= 0.0041). TTD for 50-keV, IOI -50%, and s-120-kvp datasets resulted 37±4 seconds (sec), 39±10 sec, 107±7 sec).

Low energy imaging (50-keV) and color-coded IOI (50% saturation) significantly improved bowel wall conspicuity with increased attenuation differences and higher diagnostic confidence between ischemic and non-ischemic bowel compared to simulated 120-kVp. In addition, 50 keV and 50% IOI datasets allowed shorter TTD.

Hypoperfused bowel often goes unrecognized especially for short ischemic segments that blend in with normal bowel loops. Low monoenergetic images and color-coded overlay iodine maps increase bowel wall attenuation differences improving hypoenhanced bowel segments identification.

stefromano@libero.it

Purpose of the study was to retrospectively analyze the processed imaging findings from the Dual Energy CT examinations of patients with acute abdominal symptoms compared with the native axial and multiplanar reconstructions in evaluating the cause of disease.

A retrospective analysis on 122 patients who underwent DECT examination in emergency for acute abdominal symptoms were considered. All examinations were performed using a dual energy dual source CT 128 detector rows scanner (Drive, Siemens), after administration of i.v. contrast medium (Iomeron 400, Bracco) with mono or double phase acquisition. Native and processed images (i.e. iodine map, fusion series, virtual non contrast) were analysed in evaluation of: mesenteric vessels opacification of major and secundary branches; bowel wall thickening; bowel wall enhancement; abdominal addictional findings (free peritoneal air and fluid, mesenteric stranding, bowel lumen dilatation). Double readers / blinded final diagnosis analysis were performed; a cross-check of imaging and surgical/endoscopic from both native and processed images were made.

In 94/122 patients a correct diagnosis with correlative native imaging findings have been noted. In 39 patients in which the native images were already effective, post processed imaging findings did not add any new informations, whereas among the 28 patients with inconclusive findings at the native scans regarding the final diagnosis, processed images (iodine map) seemed to show altered findings, most of them in inflammatory and ischemic bowel disease.

DECT could be of help in case of controversial and not defined imaging findings , but the relative absolu value of the iodine map in evaluating the bowel wall trophism seemed to be reconsidered.

Clinical relevance of this study is mainly pertinent on the DECT in evaluating the bowel wall enhancement in acute conditions

gulerezgi@yahoo.com

To evaluate abdominopelvic imaging findings of cancer patients treated with immune checkpoint inhibitors (ICIs) who presented to the emergency department (ED).

A retrospective database search was performed to identify patients treated with ICIs who presented to the ED and underwent abdominopelvic imaging between January 2010 and November 2018. Images were reviewed to assess tumor burden and to detect immune-related adverse events (irAEs). Clinical indications for imaging and management were documented from medical record.

A hundred patients (62 men, median age: 63 years) with 138 abdominopelvic exams including 123 CT scans, 9 X-rays, 5 US, and 1 HIDA exams were identified. The most common cancer types included lung (40%), melanoma (17%), and kidney (12%). The date of ED visit occurred a median of 70 days after starting ICI (IQR 75-25 139-31). Common imaging indications included abdominal pain (72%), constipation (5%), and hematuria (5%). Forty-nine (35%) ED abdominopelvic scans showed worsening tumor burden at a median of 42 days (IQR 75-25 92-22) following initiation of ICIs. Twenty-five (18%) scans detected a cause of acute abdomen in patients treated with ICIs. In 5 (4%) exams, both an etiology of acute abdomen and worsening tumor burden were identified. Ten (10%) out of 100 patients demonstrated an irAE at a median of 74 days (IQR 75-25 284-42) following initial dose of ICI. irAEs detected on imaging included colitis/enteritis (6/10), pneumatosis coli (2/10), acute hepatitis (1/10), and acute acalculous cholecystitis (1/10). Steroid therapy was started in 2 cases following ICI discontinuation. The other most common causes of acute abdomen included gastrointestinal tract emergencies (10/100), and collections/abscesses (5/100).

Thirty-five percent of abdominopelvic imaging of patients on ICI therapy who presented to the ED demonstrated worsening tumor burden. Abdominopelvic irAEs were detected on imaging in 10% of patients and colitis/enteritis was the most common irAE.

Abdominopelvic imaging at the ED detected the etiology of clinical presentation in 57% of patients treated with ICIs.

camilo_campo@hms.harvard.edu

Computed tomography (CT) has the highest sensitivity and specificity for urolithiasis in patients with acute flank pain. However, studies have suggested that ultrasound (US) should be the initial imaging test in the Emergency Department (ED) for acute flank pain. The purpose of this study is to assess the value of renal US in predicting follow-up imaging for patients with acute flank pain and to calculate the additional time required to obtain a renal US prior to CT.

This was an IRB-approved, HIPAA-compliant retrospective study of all patients that underwent renal US in the ED from March 2018-March 2019 for acute flank pain. Data points collected were: presence of calculi and/or hydronephrosis on US, whether patients underwent follow-up imaging within 24 hours of US, presence of calculi and/or hydronephrosis on follow-up imaging, acute extra-renal findings, need for intervention, and history of urolithiasis. The time interval between US and follow-up CT was recorded based on the time that the tests were ordered.

271 patients underwent renal US in the ED for acute flank pain. 76 of 271 patients (28%) underwent follow-up imaging within 24 hours of initial US: 72 underwent CT abdomen/pelvis and 4 underwent magnetic resonance urogram (MR). Of the initial 271 US, 138 (51%) were positive for calculi and/or hydronephrosis on US. Of the 76 patients who underwent follow-up CT or MR, 40 (52%) had been positive for calculi and/or hydronephrosis on initial US and 36 (47%) had been negative on US. Of the 76 that underwent follow-up imaging, 10 had acute extra-renal findings, and 17 had subsequent intervention. For patients that underwent follow-up CT, the mean time interval between US and CT was 170 min.

Few cases of acute flank pain underwent follow-up imaging. Of these, approximately half had a positive US (54%) and half had a negative US (47%). Therefore, it is likely that clinical judgement plays a large role in predicting the need for follow-up imaging. We also found a larger time interval between US and follow-up CT for patients with positive versus negative US, which may reflect that patients with positive US were given the opportunity to pass the stones before follow-up imaging.

Renal US is often the initial test done in the ED for acute flank pain. We evaluate how renal US predicts follow-up imaging and the additional time required to obtain a renal US prior to CT.

jhines@northwell.edu

1) Identify acute and emergent conditions of the kidney, with emphasis on renal obstruction, infection and hemorrhage. 2) Discuss cross-sectional imaging findings typically found with acute kidney disease, with an emphasis on CT. 3) Explain how the radiologist can integrate imaging findings with clinical history in order to help guide management of the patient with acute kidney disease.