Management of Patients With Common Bile Duct Dilatation Without a Sonographic Evident Cause: Evaluating the Yield of Subsequent Magnetic Resonance Imaging and Findings Correlated With Causative Pancreaticobiliary Pathology.

OBJECTIVES: The aims of the study were to evaluate the yield of magnetic resonance imaging (MRI) in patients with common bile duct (CBD) dilatation without a sonographic evident cause and to investigate sonographic and laboratory findings at presentation that might predict identification of underlying pancreaticobiliary pathology. METHODS: Included were consecutive patients in whom MRI was performed for further evaluation of CBD dilatation detected on ultrasound (US), without a sonographic evident cause, from January 1, 2014, to December 31, 2014. Magnetic resonance imaging and clinical data were retrospectively reviewed. Patients were divided into two groups: patients with and without identified underlying causative pancreaticobiliary pathology. Ultrasound findings and laboratory results at presentation were compared between groups to identify findings suggestive of underlying pancreaticobiliary pathology. RESULTS: Fifty-seven patients, with a mean age of 54 ± 16 years including 37 females (65%), underwent MRI. Specific pancreaticobiliary causes for CBD dilatation were identified in 38 patients (66%, 31 benign and 7 malignant). In the remaining patients, no cause was identified in 17 (30%) and MRIs were nondiagnostic in 2 patients. Magnetic resonance imaging accuracy for correctly identifying the underlying cause of CBD dilatation was 91%. Patients with associated intrahepatic bile duct dilatation on US and elevated alanine aminotransferase and alkaline phosphatase were more likely to have underlying pancreaticobiliary pathology (P < 0 .05). No patient with initial negative MRI had pancreaticobiliary malignancy. CONCLUSIONS: Magnetic resonance imaging seems to be an accurate noninvasive method for identifying the underlying cause in most patients with CBD dilatation on US and in excluding pancreaticobiliary malignancy. Patients with associated intrahepatic bile duct dilatation and/or elevated liver enzymes are at higher risk of harboring underlying pancreaticobiliary pathology.

The "Speckle Sign" in the Diagnosis of Posterior Compartment Endometriosis.

OBJECTIVE: To determine usefulness of the "speckle sign" in the diagnosis of deep invasive endometriosis. MATERIALS AND METHODS: This HIPAA-compliant, institutional review board-approved retrospective study with informed consent waived included 25 women (mean age 20-69 years) with histopathologically confirmed posterior cul-de-sac endometriosis between 2013 and 2018. Transvaginal ultrasound exams of these patients were reviewed by 2 expert radiologists searching for the "speckle sign," defined as irregular obliteration of the posterior cul-de-sac and bright (hyperechoic) internal echoes. The frequency of additional findings such as "kissing ovaries," endometriomas in the adnexa, bowel tethering in the posterior pelvic compartment, retroflexed uterus, adenomyosis, and pelvic free fluid were also analyzed. Data regarding clinical features, histopathologic findings and management were collected through a review of the medical record. RESULTS: Reader one identified posterior compartment endometriosis in 20/25 patients, and reader two in 22/25 patients, with 96% agreement. Adnexal endometriomas were found in 21/25 patients for both readers (k = 0.70) and were bilateral in 23% of patients. The ovaries were adherent to each other in the midline ("kissing ovaries") in 50% of patients; the bowel was tethered anteriorly in 20%; the presence of adenomyosis was seen in about 27%, and a retroflexed uterus was seen in 24% of patients. CONCLUSIONS: The speckle sign could be helpful in making the diagnosis of posterior compartment endometriosis, and the sign is often found in conjunction with other imaging features of endometriosis.

Correction to: Effects of Physician Experience, Specialty Training, and Self-referral on Inappropriate Diagnostic Imaging.

There were some errors in the variables in this paper.

Effects of Physician Experience, Specialty Training, and Self-referral on Inappropriate Diagnostic Imaging.

BACKGROUND: Although previous research has demonstrated high rates of inappropriate diagnostic imaging, the potential influence of several physician-level characteristics is not well established. OBJECTIVE: To examine the influence of three types of physician characteristics on inappropriate imaging: experience, specialty training, and self-referral. DESIGN: A retrospective analysis of over 70,000 MRI claims submitted for commercially insured individuals. Physician characteristics were identified through a combination of administrative records and primary data collection. Multi-level modeling was used to assess relationships between physician characteristics and inappropriate MRIs. SETTING: Massachusetts PARTICIPANTS: Commercially insured individuals who received an MRI between 2010 and 2013 for one of three conditions: low back pain, knee pain, and shoulder pain. MEASUREMENTS: Guidelines from the American College of Radiology were used to classify MRI referrals as appropriate/inappropriate. Experience was measured from the date of medical school graduation. Specialty training comprised three principal groups: general internal medicine, family medicine, and orthopedics. Two forms of self-referral were examined: (a) the same physician who ordered the procedure also performed it, and (b) the physicians who ordered and performed the procedure were members of the same group practice and the procedure was performed outside the hospital setting. RESULTS: Approximately 23% of claims were classified as inappropriate. Physicians with 10 or less years of experience had significantly higher odds of ordering inappropriate MRIs. Primary care physicians were almost twice as likely to order an inappropriate MRI as orthopedists. Self-referral was not associated with higher rates of inappropriate MRIs. LIMITATIONS: Classification of MRIs was conducted with claims data. Not all self-referred MRIs could be detected. CONCLUSIONS: Inappropriate imaging continues to be a driver of wasteful health care spending. Both physician experience and specialty training were highly associated with inappropriate imaging.

Pancreatic perfusion modulation following glucose stimulation assessed by noninvasive arterial spin labeling (ASL) MRI.

BACKGROUND: More than 100 million adults in the US suffer from prediabetes or type-2 diabetes. Noninvasive imaging of pancreas endocrine function might provide a surrogate marker of ß-cell functional integrity loss linked to this disease. PURPOSE: To noninvasively assess pancreatic blood-flow modulation following a glucose challenge using arterial spin labeling (ASL) MRI. STUDY TYPE: Prospective. SUBJECTS: Fourteen adults (30 ± 7 years old, 3M/11F, body mass index [BMI] = 24 ± 3 kg.m-2 ). FIELD STRENGTH/SEQUENCE: 3T MRI / background-suppressed pseudocontinuous PCASL preparation with single-shot fast-spin-echo (FSE) readout before and after an oral glucose challenge using either fruit juice (n = 7) or over-the-counter glucose gel (n = 7). ASSESSMENT: Subjects were fasting prior to initiation of oral stimulation, then dynamic perfusion measurements were performed every 2 minutes for 30 minutes. We quantified absolute blood flow at each timepoint. STATISTICAL TESTS: Repeated-measures analysis of variance (ANOVA) followed by paired t-tests to assess for a significant effect of glucose challenge on measured perfusion. RESULTS: Measured basal blood flow was 187 ± 53 mL/100g/min. A significant blood flow increase of +38 ± 26% was observed 10 minutes poststimulation (P < 0.05) and continuing until the end of the experiment. The gel stimulation provided the most consistent results, with an early rise followed by an additional later increase consistent with the known pancreatic insulin response to elevated blood glucose. Across-subject variations in blood flow increase were partially attributable to basal flow, with a negative correlation of r = -0.84 between basal and maximal relative flow increase in the gel group. DATA CONCLUSION: ASL can be used to measure pancreatic flow in response to a glucose challenge, which could be linked to insulin release and secretion. This paradigm might be useful to characterize disorders of glucose regulation. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2020;51:854-860.

Pancreatic Ductal Adenocarcinoma and Its Variants: Pearls and Perils.

Pancreatic ductal adenocarcinoma (PDAC), an epithelial neoplasm derived from the pancreatic ductal tree, is the most common histologic type of pancreatic cancer and accounts for 85%-95% of all solid pancreatic tumors. As a highly lethal malignancy, it is the seventh leading cause of cancer death worldwide and is responsible for more than 300 000 deaths per year. PDAC is highly resistant to current therapies, affording patients a 5-year overall survival rate of only 7.2%. It is characterized histologically by its highly desmoplastic stroma embedding tubular and ductlike structures. On images, it typically manifests as a poorly defined hypoenhancing mass, causing ductal obstruction and vascular involvement. Little is known about the other histologic subtypes of PDAC, mainly because of their rarity and lack of specific patterns of disease manifestation. According to the World Health Organization, these variants include adenosquamous carcinoma, colloid carcinoma, hepatoid carcinoma, medullary carcinoma, signet ring cell carcinoma, undifferentiated carcinoma with osteoclast-like giant cells, and undifferentiated carcinoma. Depending on the subtype, they can confer a better or even worse prognosis than that of conventional PDAC. Thus, awareness of the existence and differentiation of these variants on the basis of imaging and histopathologic characteristics is crucial to guide clinical decision making for optimal treatment and patient management.

Feasibility, safety, and utility of brain MRI for patients with non-MRI-conditioned CIED.

Feasibility, safety, and utility of brain MRI for patients with non-MRI-conditioned cardiac implantable electrical devices (CIEDs) remains controversial. While a growing number of studies have shown safe employment in select patients under strict protocols, there is an increasing clinical need for further off-label investigations. To assess the feasibility and utility of brain MRI in neurological and neurosurgical patients with non-MRI-conditioned CIEDs using off-label protocol. We retrospectively evaluated 126 patients with non-MRI-conditioned CIEDs referred to our hospital between 2014 to 2018 for MRI under an IRB-approved protocol. A total of 126 off-label brain MRI scans were performed. The mean age was 67.5 ± 13.0. Seventy percent of scans were performed on female patients. Indications for MRI are neurosurgical (45.2%), neurological (51.6%), and others (3.2%). MRI utilization for tumor cases was highest for tumor cases (68.3%), but employment was valuable for vascular (12.7%), deep brain stimulators (3.2%), and other cases (15.9%). In the tumor category, (37.2%) of the scans were performed for initial diagnosis and pre-surgical planning, (47.7%) for post-intervention evaluation/surveillance, (15.1%) for stereotactic radiosurgery treatment (CyberKnife). No clinical complications were encountered. No functional device complications of the CIED were identified during and after the MRI in 96.9% of the studies. A 49.6% of the off-label brain MRI scans performed led to a clinically significant decision and/or intervention for the patients. A 42.9% of obtained MRI studies did not change the plan of care. A 7.9% of post-scan decision-making data was not available. We demonstrate that off-label brain MRI scans performed on select patients under a strict protocol is feasible, safe, and relevant. Almost 50% of scans provided critical information resulting in clinical intervention of the patients.

Accuracy of a novel noninvasive secretin-enhanced MRCP severity index scoring system for diagnosis of chronic pancreatitis: correlation with EUS-based Rosemont criteria.

PURPOSE: To evaluate the accuracy of a secretin-enhanced MRCP Chronic Pancreatitis Severity Index (CPSI) in the diagnosis of chronic pancreatitis (CP) based on endoscopic ultrasound (EUS) Rosemont criteria. METHODS: In this retrospective study, 31 patients (20 women; median age 48 years, range 18-77) with known/suspected CP evaluated with both EUS and secretin-enhanced MRCP were included. CP severity was graded using a ten-point-scale secretin-enhanced MRCP-based CPSI scoring system which considered ductal, parenchymal and secretin-based dynamic abnormalities. Cases were categorized as normal, mild, moderate or severe CP. Correlation between CPSI and the EUS Rosemont criteria was performed using Cohen's kappa coefficient. Comparative evaluation of test performance was obtained using ROC analysis. RESULTS: Using EUS Rosemont criteria, eight patients had features consistent/suggestive of CP, 20 patients were normal and three were indeterminate. On CPSI, five patients were normal, 12 had mild and 14 had moderate/severe CP. There was only fair agreement (k = 0.272) between CPSI and Rosemont criteria categories. CPSI showed 87.5% sensitivity, 69.6% specificity and 74.2% accuracy (cutoff value = 3.5 points; area under the curve = 0.804; p = 0.0026) for CP diagnosis based on EUS Rosemont criteria. CONCLUSION: CPSI showed relatively high diagnostic accuracy for diagnosis of CP based on Rosemont criteria. The CPSI scoring system can be proposed as a noninvasive alternative to the EUS Rosemont criteria for CP diagnosis.

Pancreatic perfusion and arterial-transit-time quantification using pseudocontinuous arterial spin labeling at 3T.

PURPOSE: To demonstrate the feasibility of noninvasively measuring pancreatic perfusion using pseudocontinuous arterial spin labeling (ASL) and to derive quantitative blood-flow and transit-time measurements in healthy volunteers. METHODS: A pseudocontinuous ASL sequence with background suppression and a single-slice single-shot fast-spin-echo readout was acquired at 3 T in 10 subjects with a single standard postlabeling delay (PLD) of 1.5 s and in 4 additional subjects with 4 PLD from 0.7 to 2 s. An imaging synchronized breathing approach was used to minimize motion artifacts during the 3 min of acquisition. Scan-rescan reproducibility was assessed in 3 volunteers with single-delay ASL. Quantitative blood flow and arterial transit time (ATT) were derived and the impact of ATT correction was studied using either subject-specific ATT in the second group or an average ATT derived from the group with multidelay ASL for subjects with single-delay ASL. RESULTS: Successful ASL acquisitions were performed in all volunteers. An average pancreatic blood flow of 201 ± 40 mL/100 g/min was measured in the single-delay group using an assumed ATT of 750 ms Average ATT measured in the multidelay group was 1029 ± 89 ms Using the longer, measured ATT reduced the measured flow to 162 ± 12 and 168 ± 28 mL/100 g/min with subject-specific or average ATT correction, respectively. ASL signal heterogeneities were observed at shorter PLD, potentially linked to its complex vascular supply and islet distribution. CONCLUSIONS: ASL enables reliable measurement of pancreatic perfusion in healthy volunteers. It presents a valuable alternative to contrast-enhanced methods and may be useful for diagnosis and characterization of several inflammatory, metabolic, and neoplastic diseases affecting the pancreas.

Incidental pulmonary embolism in pancreatic ductal adenocarcinoma: Impact of tumor and AJCC stages at initial staging CT.

BACKGROUND/OBJECTIVES: To determine the prevalence of incidental pulmonary embolism (PE) detected during initial staging CT among patients with newly diagnosed pancreatic ductal adenocarcinoma (PDAC) and assess their association with underlying tumor burden. MATERIALS AND METHODS: This retrospective cohort study evaluated staging chest CT scans (2013-2017) to identify PE among patients with treatment naïve, biopsy-proven PDAC. Data included age, sex, T stage, AJCC stage, presence/absence of metastases and their location at diagnosis. The association of PE with tumor (T1-T4) and AJCC stage were assessed using Pearson Chi-square and Fischer's exact test. A threshold p-value of <0.05 indicated statistical significance. RESULTS: A total of 174 patients (90 female, mean age, 68 years; range: 34-93) were identified, of which 10 patients harbored incidental PE (prevalence, 5.7%). In the PE group, two patients presented with distant metastasis (liver, 20%), while eight patients had T4 tumors (80%). No statistical association was detected between PE and age, sex, and the presence/absence or location of distant metastasis (p = 0.065, p = 0.59, p = 0.687 and p = 0.933, respectively). Patients with T4 tumors and higher AJCC stages (stage III/IV) were significantly more likely to present with PE than those with lower T stage (p = 0.045) and AJCC stage (stage I/II; p = 0.017). CONCLUSION: The prevalence of incidental PE among PDAC patients undergoing initial CT staging is 5.7%. Patients with T4 and AJCC stages III/IV are at higher risk of PE. Caution should be exercised during radiographic interpretation of initial staging chest CTs, as incidental PE may be lurking and require treatment.

DWI of Pancreatic Ductal Adenocarcinoma: A Pilot Study to Estimate the Correlation With Metastatic Disease Potential and Overall Survival.

OBJECTIVE: The purpose of this study is to analyze the relationship between the apparent diffusion coefficient (ADC) of pancreatic ductal adenocarcinoma (PDAC) and the presence or development of metastasis and overall survival (OS). MATERIALS AND METHODS: Of 290 consecutive patients with histopathologically proven PDAC from January 2013 to December 2014, staging DWI was performed for 124 patients. Image quality was adequate in 112 studies. Sixty-five patients were treatment naïve, but 17 of the 65 were excluded because of the presence of other associated pancreatic pathologic abnormalities. Data for the remaining 48 patients (24 men and 24 women; median age, 65.5 years; interquartile range, 56-77 years) were obtained during a 4-year follow-up period (mean [± SD], 397 ± 415.1 days). The correlation between ADC and the presence or development of metastasis was assessed using descriptive statistics. OS was determined and mortality analysis was performed using Pearson correlation and Kaplan-Meier curves. RESULTS: Of 48 patients, 10 had metastases at staging MRI, and 12 later developed metastatic disease. Among the latter, the mean time from staging MRI to metastasis was 258 ± 274.1 days. Most (86%) metastases were hepatic (n = 19). During the follow-up period, the remaining 26 patients (54%) never developed metastases. Patients with metastatic disease (n = 22) had significantly lower mean ADCs than did those without metastases (1.27 × 10-3 vs 1.43 × 10-3 mm2/s; p = 0.047). The ADC of PDAC had a positive correlation with survival: patients with PDAC with lower ADCs (< 1.36 × 10-3 mm2/s) had significantly worse 4-year OS rates than did patients with higher ADC values (p = 0.036). CONCLUSION: Pretreatment ADC values of PDAC may be significantly lower in patients who have or will develop metastatic disease and may correlate with worse OS.

Magnitude and financial implications of inappropriate diagnostic imaging for three common clinical conditions.

OBJECTIVE: To quantify the level of adherence to imaging guidelines for three common clinical indications for a commercially insured population. DESIGN: Retrospective analysis of administrative claims data for commercially insured individuals with diagnostic imaging claims (MRI and X-ray) for either uncomplicated low back pain, non-traumatic knee pain or non-traumatic shoulder pain. SETTING: The State of Massachusetts for 2010 and 2013. PARTICIPANTS: Adults with no chronic conditions and without evidence of prior management in the 12 months preceding to the initial office visit for each of the clinical indications. MAIN OUTCOMES MEASURES: Imaging procedures performed within 30 days of the initial office visit were classified as appropriate or inappropriate according to adherence to imaging guidelines from American College of Radiology. RESULTS: More than 60% of lumbar spine MRI's were deemed inappropriate in 2010 and in 2013. Over 30% of MRI's for shoulder pain and knee pain were inappropriate in 2010 and in 2013. Patients age 18-59 with inappropriate imaging claims had significantly lower rates of surgical procedures within 90 days of imaging than those with appropriate imaging. Inappropriate imaging accounted for over 20% of annual imaging costs for the three clinical indications. CONCLUSIONS: Reducing inappropriate imaging procedures can lead to substantial savings through the elimination of unnecessary and low value procedures. Increased awareness of and adherence to best practice guidelines should be a focus of efforts to cut waste in our healthcare system.

Prediction of the histopathologic findings of intrahepatic cholangiocarcinoma: qualitative and quantitative assessment of diffusion-weighted imaging.

OBJECTIVE: To correlate qualitative and quantitative diffusion weighted imaging (DWI) characteristics of intrahepatic cholangiocarcinoma (ICC) with histopathologic tumour grade and fibrosis content. METHODS: Fifty-one patients (21M/30F; mean age 61y) with ICC and MRI including DWI were included in this IRB-approved multicentre retrospective study. Qualitative tumour features were assessed. Tumour apparent diffusion coefficient (ADC) mean, minimum, and normalized (nADCliver) values were computed. Tumour grade [well(G1), moderately(G2), or poorly differentiated(G3)] and tumour fibrosis content [minimal(1), moderate(2), or abundant(3)] were categorized pathologically. Imaging findings and ADC values were compared with pathologic measures. Utility of ADC values for predicting tumour grade was assessed using ROC analysis. RESULTS: 51 ICCs (mean size 6.5±1.1 cm) were assessed. 33/51(64%) of ICCs demonstrated diffuse hyperintensity and 15/51(29%) demonstrated target appearance on DWI. Infiltrative morphology (p=0.02) and tumour size (p=0.04) were associated with G3. ADCmean and nADCmean of G3 (1.32±0.47x10-3 mm2/sec and 0.97±0.95) were lower than G1+G2 (1.57±0.39x10-3 mm2/sec and 1.24±0.49; p=0.03 and p=0.04). ADCmean and nADCmean were inversely correlated with tumour grade (p<0.025). No correlation was found between ADC and tumour fibrosis content. AUROC, sensitivity and specificity of nADCmean for G3 versus G1+G2 were 0.71, 89.5% and 55.5%. CONCLUSION: ADC quantification has reasonable accuracy for predicting ICC grade. KEY POINTS: • ADC quantification was useful for predicting ICC tumour grade. • Infiltrative tumour morphology and size were associated with poorly differentiated ICCs. • ADC values depended more on ICC tumour grade than fibrosis content. • Ability to predict ICC tumour grade non-invasively could impact patient management.

Structured vs narrative reporting of pelvic MRI for fibroids: clarity and impact on treatment planning.

OBJECTIVES: To evaluate clarity and usefulness of MRI reporting of uterine fibroids using a structured disease-specific template vs. narrative reporting for planning of fibroid treatment by gynaecologists and interventional radiologists. METHODS: This is a HIPAA-compliant, IRB-approved study with waiver of informed consent. A structured reporting template for fibroid MRIs was developed in collaboration between gynaecologists, interventional and diagnostic radiologists. The study population included 29 consecutive women who underwent myomectomy for fibroids and pelvic MRI prior to implementation of structured reporting, and 42 consecutive women with MRI after implementation of structured reporting. Subjective evaluation (on a scale of 1-10, 0 not helpful; 10 extremely helpful) and objective evaluation for the presence of 19 key features were performed. RESULTS: More key features were absent in the narrative reports 7.3 ± 2.5 (range 3-12) than in structured reports 1.2 ± 1.5 (range 1-7), (p < 0.0001). Compared to narrative reports, gynaecologists and radiologists deemed structured reports both more helpful for surgical planning (p < 0.0001) (gynaecologists: 8.5 ± 1.2 vs. 5.7 ± 2.2; radiologists: 9.6 ± 0.6 vs. 6.0 ± 2.9) and easier to understand (p < 0.0001) (gynaecologists: 8.9 ± 1.1 vs. 5.8 ± 1.9; radiologists: 9.4 ± 1.3 vs. 6.3 ± 1.8). CONCLUSION: Structured fibroid MRI reports miss fewer key features than narrative reports. Moreover, structured reports were described as more helpful for treatment planning and easier to understand. KEY POINTS: • Structured reports missed only 1.2 ± 1.5 out of 19 key features, as compared to narrative reports that missed 7.3 ± 2.5 key features for planning of fibroid treatment. • Structured reports were more helpful and easier to understand by clinicians. • Structured template can provide essential information for fibroids treatment planning.

Optimizing Inpatient Body MRI Utilization: A Granular Look at Trends, Quality, Yield, and Timing.

OBJECTIVE: We sought to analyze body MRI utilization trends, quality, yield, and timing among inpatients in a tertiary care academic medical center. MATERIALS AND METHODS: By use of billing data from fiscal years (FYs) 2006-2015, the volume of admissions was compared with the total number of inpatient body MRI examinations. MRI examinations per admissions and discharge were adjusted using the Centers for Medicare & Medicaid Services case mix index by FY. Linear regression was used to assess trends. In addition, each inpatient body MRI examination performed in FY 2015 was evaluated and graded on its quality and yield and was judged as to whether it could have been performed on an outpatient basis. RESULTS: There was an increase in the number of inpatient body MRI examinations, from 637 examinations in FY 2006 to 871 examinations in FY 2015 (p = 0.005). By adjusting for case mix, the upward trend for body MRI use persisted (p = 0.012). Regarding quality, 2.3% of all inpatient body MRI examinations were nondiagnostic, 40.4% were limited quality, and 57.3% were of diagnostic quality. Concerning yield, 20.8% of all examinations had no yield, 5.1% of examinations had no yield but incidental findings, and 74.1% of examinations had a positive yield. Finally, regarding timing, 30.2% of examinations could have been performed as outpatient examinations. CONCLUSION: At our institution, the number of inpatient body MRI examinations has increased significantly over the past 10 years. Many of the examinations, however, are poor quality, often give redundant information, and could be performed in the outpatient setting.

Nonepithelial Neoplasms of the Pancreas, Part 2: Malignant Tumors and Tumors of Uncertain Malignant Potential From the Radiologic Pathology Archives.

Almost all neoplasms of the pancreas are derived from pancreatic epithelial components, including the most common pancreatic mass, primary pancreatic ductal adenocarcinoma (PDAC). Nonepithelial neoplasms comprise only 1%-2% of all pancreatic neoplasms. Although some may arise directly from intrapancreatic elements, many originate from mesenchymal, hematopoietic, or neural elements in the retroperitoneal peripancreatic space and grow into the pancreas. Once these tumors reach a certain size, it can be challenging to identify their origin. Because these manifest at imaging as intrapancreatic masses, awareness of the existence and characteristic features of these nonepithelial neoplasms is crucial for the practicing radiologist in differentiating these tumors from primary epithelial pancreatic tumors, an important distinction given the vastly different management and prognosis. In part 1 of this article, the authors reviewed benign nonepithelial neoplasms of the pancreas. This article focuses on malignant nonepithelial neoplasms and those of uncertain malignant potential that can be seen in the pancreas. The most common malignant or potentially malignant nonepithelial pancreatic tumors are of mesenchymal origin and include soft-tissue sarcomas, solitary fibrous tumor, and inflammatory myofibroblastic tumor. These tumors commonly manifest as large heterogeneous masses, often containing areas of necrosis and hemorrhage. The clinical features associated with these tumors and the imaging characteristics including enhancement patterns and the presence of fat or calcification help distinguish these tumors from PDAC. Hematopoietic tumors, including lymphoma and extramedullary plasmacytoma, can manifest as isolated pancreatic involvement or secondarily involve the pancreas as widespread disease. Hyperenhancing paragangliomas or hypervascular metastatic disease can mimic primary pancreatic neuroendocrine tumors or vascular anomalies.

Optimizing MRI Logistics: Prospective Analysis of Performance, Efficiency, and Patient Throughput.

OBJECTIVE: The objective of this study is to optimize MRI logistics through evaluation of MRI workflow and analysis of performance, efficiency, and patient throughput in a tertiary care academic center. SUBJECTS AND METHODS: For 2 weeks, workflow data from two outpatient MRI scanners were prospectively collected and stratified by value added to the process (i.e., value-added time, business value-added time, or non-value-added time). Two separate time cycles were measured: the actual MRI process cycle as well as the complete length of patient stay in the department. In addition, the impact and frequency of delays across all observations were measured. RESULTS: A total of 305 MRI examinations were evaluated, including body (34.1%), neurologic (28.9%), musculoskeletal (21.0%), and breast examinations (16.1%). The MRI process cycle lasted a mean of 50.97 ± 24.4 (SD) minutes per examination; the mean non-value-added time was 13.21 ± 18.77 minutes (25.87% of the total process cycle time). The mean length-of-stay cycle was 83.51 ± 33.63 minutes; the mean non-value-added time was 24.33 ± 24.84 minutes (29.14% of the total patient stay). The delay with the highest frequency (5.57%) was IV or port placement, which had a mean delay of 22.82 minutes. The delay with the greatest impact on time was MRI arthrography for which joint injection of contrast medium was necessary but was not accounted for in the schedule (mean delay, 42.2 minutes; frequency, 1.64%). Of 305 patients, 34 (11.15%) did not arrive at or before their scheduled time. CONCLUSION: Non-value-added time represents approximately one-third of the total MRI process cycle and patient length of stay. Identifying specific delays may expedite the application of targeted improvement strategies, potentially increasing revenue, efficiency, and overall patient satisfaction.

Optimizing MRI Logistics: Focused Process Improvements Can Increase Throughput in an Academic Radiology Department.

OBJECTIVE: The purpose of this study was to describe and evaluate the effect of focused process improvements on protocol selection and scheduling in the MRI division of a busy academic medical center, as measured by examination and room times, magnet fill rate, and potential revenue increases and cost savings to the department. MATERIALS AND METHODS: Focused process improvements, led by a multidisciplinary team at a large academic medical center, were directed at streamlining MRI protocols and optimizing matching protocol ordering to scheduling while maintaining or improving image quality. Data were collected before (June 2013) and after (March 2015) implementation of focused process improvements and divided by subspecialty on type of examination, allotted examination time, actual examination time, and MRI parameters. Direct and indirect costs were compiled and analyzed in consultation with the business department. Data were compared with evaluated effects on selected outcome and efficiency measures, as well as revenue and cost considerations. Statistical analysis was performed using a t test. RESULTS: During the month of June 2013, 2145 MRI examinations were performed at our center; 2702 were performed in March 2015. Neuroradiology examinations were the most common (59% in June 2013, 56% in March 2015), followed by body examinations (25% and 27%). All protocols and parameters were analyzed and streamlined for each examination, with slice thickness, TR, and echo train length among the most adjusted parameters. Mean time per examination decreased from 43.4 minutes to 36.7 minutes, and mean room time per patient decreased from 46.3 to 43.6 minutes (p = 0.009). Potential revenue from increased throughput may yield up to $3 million yearly (at $800 net revenue per scan) or produce cost savings if the facility can reduce staffed scanner hours or the number of scanners in its fleet. Actual revenue and expense impacts depend on the facility's fixed and variable cost structure, payer contracts, MRI fleet composition, and unmet MRI demand. CONCLUSION: Focused process improvements in selecting MRI protocols and scheduling examinations significantly increased throughput in the MRI division, thereby increasing capacity and revenue. Shorter scan and department times may also improve patient experience.

Pathologic Subtypes of Ampullary Adenocarcinoma: Value of Ampullary MDCT for Noninvasive Preoperative Differentiation.

OBJECTIVE: The purpose of this study was to evaluate the utility of ampullary MDCT in the noninvasive, preoperative differentiation of pancreatobiliary and intestinal subtypes of ampullary adenocarcinoma. MATERIALS AND METHODS: This retrospective study included 32 patients (20 men, 12 women; age range, 41-81 years) with resected ampullary adenocarcinoma who underwent preoperative contrast-enhanced ampullary MDCT. Two radiologists, blinded to pathologic diagnosis of adenocarcinoma subtype, evaluated the presence of seven MDCT features independently. MDCT findings and ampullary adenocarcinoma subtypes were correlated using chi-square and Fisher exact tests. Interobserver agreement was evaluated using the Cohen kappa statistic. RESULTS: When evaluated with ampullary MDCT, the intestinal and pancreatobiliary subtypes were significantly different in terms of lesion morphology (p < 0.0001), papillary shape (p < 0.0001), common bile duct (CBD) infiltration and dilatation (p = 0.003 and p = 0.0004, respectively), duodenopancreatic groove infiltration (p = 0.0009), and pancreaticoduodenal artery involvement (p = 0.004). Pancreatobiliary subtype tumors were more often infiltrative in morphology (18/18) and showed retracted papilla (14/18), CBD (18/18) and main pancreatic duct (MPD) infiltration (12/18), dilated CBD (18/18) and MPD (13/18), fixed duodenopancreatic groove appearance (15/18), and pancreaticoduodenal artery involvement (12/18). Intestinal subtype carcinomas were more frequently nodular (14/14) and had a bulging papilla (13/14), a free duodenopancreatic groove appearance (11/14), and no pancreaticoduodenal artery involvement (2/14). When all features were taken into account, MDCT showed sensitivity of 85.7% and specificity of 83.3% in differentiating intestinal and pancreatobiliary subtype tumors. Accuracy, positive predictive value, and negative predictive value of MDCT were 84.4%, 80%, and 88.2%, respectively. Interobserver agreement was almost perfect for the presence of each imaging feature (κ > 0.8). CONCLUSION: Ampullary MDCT can be useful to differentiate pancreatobiliary and intestinal subtypes of ampullary adenocarcinoma preoperatively, provided the duodenum is optimally distended at imaging.

Nonepithelial Neoplasms of the Pancreas: Radiologic-Pathologic Correlation, Part 1--Benign Tumors: From the Radiologic Pathology Archives.

Solid and cystic pancreatic neoplasms are being recognized more frequently with increasing utilization and spatial resolution of modern imaging techniques. In addition to the more common primary pancreatic solid (ductal adenocarcinoma) and cystic neoplasms of epithelial origin, nonepithelial neoplasms of the pancreas may appear as well-defined solid or cystic neoplasms. Most of these lesions have characteristic imaging features, such as a well-defined border, which allows differentiation from ductal adenocarcinoma. Solid masses include neurofibroma, ganglioneuroma, leiomyoma, lipoma, and perivascular epithelioid cell tumor (PEComa). Schwannomas and desmoid tumors can be solid or cystic. Cystic tumors include mature cystic teratoma and lymphangioma. Lipoma, PEComa, and mature cystic teratoma can contain fat, and ganglioneuroma and mature cystic teratoma may contain calcification. Although these unusual benign neoplasms are rare, the radiologist should at least consider them in the differential diagnosis of well-defined lesions of the pancreas. The goal of this comprehensive review is to improve understanding of these rare primary pancreatic mesenchymal tumors.