Computer-Generated R.E.N.A.L. Nephrometry Scores Yield Comparable Predictive Results to Those of Human-Expert Scores in Predicting Oncologic and Perioperative Outcomes.

PURPOSE: We sought to automate R.E.N.A.L. (for radius, exophytic/endophytic, nearness of tumor to collecting system, anterior/posterior, location relative to polar line) nephrometry scoring of preoperative computerized tomography scans and create an artificial intelligence-generated score (AI-score). Subsequently, we aimed to evaluate its ability to predict meaningful oncologic and perioperative outcomes as compared to expert human-generated nephrometry scores (H-scores). MATERIALS AND METHODS: A total of 300 patients with preoperative computerized tomography were identified from a cohort of 544 consecutive patients undergoing surgical extirpation for suspected renal cancer at a single institution. A deep neural network approach was used to automatically segment kidneys and tumors, and geometric algorithms were developed to estimate components of R.E.N.A.L. nephrometry score. Tumors were independently scored by medical personnel blinded to AI-scores. AI- and H-score agreement was assessed using Lin's concordance correlation and their predictive abilities for both oncologic and perioperative outcomes were assessed using areas under the curve. RESULTS: Median age was 60 years (IQE 51-68), and 40% were female. Median tumor size was 4.2 cm and 91.3% had malignant tumors, including 27%, 37% and 24% with high stage, grade and necrosis, respectively. There was significant agreement between H-scores and AI-scores (Lin's ⍴=0.59). Both AI- and H-scores similarly predicted meaningful oncologic outcomes (p <0.001) including presence of malignancy, necrosis, and high-grade and -stage disease (p <0.003). They also predicted surgical approach (p <0.004) and specific perioperative outcomes (p <0.05). CONCLUSIONS: Fully automated AI-generated R.E.N.A.L. scores are comparable to human-generated R.E.N.A.L. scores and predict a wide variety of meaningful patient-centered outcomes. This unambiguous artificial intelligence-based scoring is intended to facilitate wider adoption of the R.E.N.A.L. score.

Difference in the frequency of pouchitis between ulcerative colitis and familial adenomatous polyposis: is the explanation in peripouch fat?

AIM: Patients with ulcerative colitis (UC) have an unexplained higher incidence of pouchitis and a greater amount of peripouch fat compared with patients with familial adenomatous polyposis (FAP). The aims of this study were to compare the peripouch fat areas between patients with UC and patients with FAP, and to explore relationship between peripouch fat and pouchitis or chronic antibiotic-refractory pouchitis (CARP). METHOD: Patients with an abdominal CT image from our prospectively maintained Pouch Database were included. Abdominal fat and peripouch fat were measured on CT images at different levels or planes. Comparisons of peripouch fat and CARP were performed before and after propensity score matching. RESULTS: A total of 277 patients with UC and 40 patients with FAP were included. Compared with patients with FAP, patients with UC were found to have a higher incidence of pouchitis (58.5% vs 15.0%, P < 0.001) and CARP (24.5% vs 2.5%, P = 0.002) and a higher total peripouch fat area (P = 0.030) and mesenteric peripouch fat area (P = 0.022) at Level-3. Univariate and multivariate analyses showed that diagnosis (UC vs FAP) and peripouch fat areas at Level-3 and Level-5 were independent risk factors for CARP. With propensity score matching, 38 pairs of patients with UC and FAP were matched successfully. After matching, patients with UC were found to have higher total peripouch fat area and higher mesenteric peripouch fat area at Level-3, and a higher incidence of pouchitis (57.9% vs 13.2%, P < 0.001) and CARP (23.7% vs 2.6%, P = 0.007). CONCLUSION: Our study demonstrates that patients with UC have more peripouch fat than those with FAP, which may explain the difference in the frequency of pouchitis and CARP between these groups of patients.

Association between high visceral fat area and postoperative complications in patients with Crohn's disease following primary surgery.

AIM: The aim of this study was to determine the association between visceral fat area (VFA) on CT and postoperative complications after primary surgery in patients with Crohn's disease (CD). METHOD: Inclusion criteria were patients with a confirmed diagnosis of CD who had preoperative abdominal CT scan. The areas of total fat, subcutaneous fat and visceral fat were measured using an established image-analysis method at the lumbar 3 (L3) level on CT cross-sectional images. Visceral obesity was defined as a visceral fat area (VFA) of ≥ 130 cm(2) . Clinical variables, intra-operative outcomes and postoperative courses within 30 days were analysed. RESULTS: A total of 164 patients met the inclusion criteria. Sixty-three (38.4%) patients had postoperative complications. The mean age of the patients with complications (the study group) was 40.4 ± 15.4 years and of those without complications (the control group) was 35.8 ± 12.9 years (P = 0.049). There were no differences in disease location and behaviour between patients with or without complications (P > 0.05). In multivariable analysis, VFA [odds ratio (OR) = 2.69; 95% confidence interval (CI): 1.09-6.62; P = 0.032] and corticosteroid use (OR = 2.86; 95% CI: 1.32-6.21; P = 0.008) were found to be associated with postoperative complications. Patients with visceral obesity had a significantly longer operative time (P = 0.012), more blood loss (P = 0.019), longer bowel resection length (P = 0.003), postoperative ileus (P = 0.039) and a greater number of complications overall (P < 0.001). CONCLUSION: High VFA was found to be associated with an increased risk for 30-day postoperative complications in patients with CD undergoing primary surgery.

Comparison of radiation dose and image quality from single-energy and dual-energy CT examinations in the same patients screened for hepatocellular carcinoma.

AIM: To compare radiation dose surrogates [volume CT dose index (CTDIvol), dose-length product (DLP), size-specific dose estimate (SSDE), and effective dose] and image noise in a cohort of patients undergoing hepatocellular carcinoma screening who underwent both single-energy CT (SECT) and dual-energy CT (DECT). MATERIALS AND METHODS: In this institutional review board-approved, Health Insurance Portability and Accountability Act-compliant retrospective study, 74 adults (mean age 59.5 years) underwent 64 section SECT (120 kVp and weight-based reference mAs) and 128 section dual-source DECT (100/Sn 140 kVp and CTDIvol, adjusted to match the CDTIvol of the SECT protocol) on different occasions. Noise levels were measured in the liver, inferior vena cava (IVC), retroperitoneal (RP) fat, and aorta. Generalized linear models were constructed to compare dose and noise, adjusting for effective diameter. RESULTS: The total DLP (1371.11 mGy-cm, SD = 527.91) and effective dose (20.57 mSv, SD = 7.92) with SECT were significantly higher than the DLP (864.84 mGy-cm, SD = 322.10) and effective dose (12.97 mSv, SD = 4.83) with DECT (p < 0.001). The differences between SECT and DECT increased as the patient's effective diameter increased (p < 0.001). Noise levels in the liver (22.4 versus 21.9 HU), IVC (22.3 versus 23.4 HU), and RP fat (23.5 versus 23 HU) were similar for DECT and SECT (p > 0.05) but were significantly lower in the aorta for DECT (25.3 versus 26.4 HU; p = 0.006). CONCLUSION: DECT imaging of the abdomen can achieve noise levels comparable to those seen with SECT imaging without a dose penalty to patients.

Focal liver lesion detection and characterization with GD-EOB-DTPA.

Superior soft-tissue contrast affords magnetic resonance imaging (MRI) some advantages compared to computed tomography (CT) in both detection and characterization of focal liver lesions. Because of its relatively recently introduction into clinical practice, a growing number of articles in the literature have demonstrated the usefulness of the hepatobiliary-specific MRI contrast agent gadoxetic acid disodium (Gd-EOB-DTPA) in liver imaging. The purpose of this review is to demonstrate the typical enhancement patterns of the most common liver lesions using Gd-EOB-DTPA in daily clinical scenarios and briefly describe its mechanism of action. Radiologists interpreting liver MRI studies with this agent must be familiar with the appearance of focal lesions in the hepatocyte phase to avoid misinterpretation.

Imaging findings after minimally invasive nephron-sparing renal therapies.

With the trend towards minimally invasive and nephron-sparing surgery for renal masses, laparoscopic partial nephrectomy and energy ablative techniques have become common approaches to treat low-stage tumours. Complications following such techniques are occasional, especially for ablation techniques. This review illustrates the imaging of these complications and of tumour recurrence, with the conclusion that imaging plays an important role in their diagnosis, and in the infrequent situation that intervention is needed, helps to plan subsequent management.

Laparoscopic renal cryoablation in 32 patients.

OBJECTIVES: Laparoscopic renal cryoablation is a developmental minimally invasive nephron-sparing treatment alternative for highly select patients with small renal tumors. We present our evolving experience with this procedure. METHODS: Thirty-two patients (34 tumors) with a mean tumor size of 2.3 cm on preoperative computed tomography underwent laparoscopic renal cryoablation. As dictated by the tumor location, cryoablation was performed by either the retroperitoneal (n = 22) or the transperitoneal (n = 10) laparoscopic approach using real-time ultrasound monitoring. A double freeze-thaw cycle was routinely performed. RESULTS: The mean surgical time was 2.9 hours, cryoablation time 15.1 minutes, and blood loss 66.8 mL. For a mean intraoperative ultrasonographic tumor size of 2 cm, the mean cryolesion size was 3.2 cm. The hospital stay was less than 23 hours in 22 (69%) of 32 patients. Sequential magnetic resonance imaging scans demonstrated a gradual contraction in the mean diameter of the cryolesions. Of the 20 patients who underwent a 1-year follow-up magnetic resonance imaging scan, the cryoablated tumor was no longer visible in 5. Of note, 23 patients have now undergone a 3 to 6-month follow-up computed tomography-directed biopsy of the cryoablated tumor site; the biopsy was negative for cancer in all 23 patients. No evidence of local or port-site recurrence was found during a mean follow-up of 16.2 months. CONCLUSIONS: Critical long-term data regarding laparoscopic renal cryoablation, a developmental technique, are awaited. However, our initial experience is cautiously optimistic. Despite its significant potential for false-negative results, it is encouraging that the follow-up computed tomography-directed needle biopsies at 3 to 6 months were negative for cancer in 23 of 23 patients.

Feasibility of instructing radiology technologists in the performance of gastrointestinal fluoroscopy.

OBJECTIVE: We sought to determine if dedicated gastrointestinal technologists could be trained to properly perform esophagography and double-contrast barium enema examinations. SUBJECTS AND METHODS: Ninety-four patients undergoing double-contrast barium enema examinations and 123 patients undergoing esophagographic examinations were included in the study. The study was conducted over a 4-month period, with examinations performed by eight gastrointestinal technologists, 10 radiology residents, and four staff radiologists. Four random lists were generated for each set of examinations. Each staff gastrointestinal radiologist, who was unaware of who had performed the examination, independently scored the representative radiographs. RESULTS: For the double-contrast barium enema examinations, no statistically significant differences were found between the technologists and residents for amount of barium used, degree of distention, cecal opacification, and quality of spot radiographs. The technologist-performed examinations had a statistically significant lower mean fluoroscopy time (3.2 min, compared with 4.0 min for staff radiologists and 5.7 min for residents). For the esophagrams, no statistically significant differences between technologists and residents were found for single-contrast esophagrams; radiographs of the gastric cardia; assessment of motility, reflux, and transit of a solid bolus; and fluoroscopy time. Double-contrast esophagrams obtained by technologists received a better mean score than did those of the residents. CONCLUSION: Radiology technologists can be trained to perform high-quality esophagography and double-contrast barium enema examinations without an unacceptably high radiation dose.

Pyocystis: two case reports of patients in renal failure.

Two cases of pyocystis in patients in end-stage renal disease are reported. Pyocystis is a recognized complication of urinary diversion procedures, but also occurs in anuric or oliguric hemodialysis patients. The treatment differs from ordinary cystitis, requiring catheterization, bladder irrigation, and intravesical antibiotic instillation. When undiagnosed, pyocystis may progress to sepsis and death.

Adrenal mass evaluation in patients with lung carcinoma: a cost-effectiveness analysis.

OBJECTIVE: This study evaluates the cost-effectiveness of various imaging and biopsy strategies for characterizing adrenal masses in patients with newly diagnosed non-small cell carcinoma of the lung. MATERIALS AND METHODS: A decision-analysis model was used to compare the cost-effectiveness of nine strategies. Initial imaging included unenhanced CT using an adenoma or nonadenoma threshold of 0 or 10 H or in- and opposed-phase MR imaging. When initial imaging did not confirm an adenoma, CT-guided biopsy or subsequent imaging was performed. Medicare reimbursement was used as a surrogate of cost. Net costs were calculated as the difference in costs between two limbs of the decision tree. Net benefits were calculated as the difference between strategies and were calculated for life expectancy in years. MR imaging, CT, and biopsy accuracy, average life expectancy, and surgical mortality rates were based on the literature. RESULTS: The base case analysis determined that the most cost-effective strategy was CT with an adenoma or nonadenoma threshold of 10 H followed by MR imaging, if necessary. CT with a threshold of 0 H followed by biopsy, if necessary, was the least costly. The incremental cost-effectiveness ratio between these two strategies was $16,370 per year of life gained. CONCLUSION: Unenhanced CT using a 10 H threshold followed by MR imaging, if needed, was the most cost-effective strategy for evaluating an adrenal mass in a patient with newly diagnosed non-small cell lung cancer.

MR imaging of the kidneys after laparoscopic cryoablation.

OBJECTIVE: We describe the MR imaging findings of patients who underwent laparoscopic renal lesion cryoablation. MATERIALS AND METHODS: Twenty-one patients (men, 11; women, 10; age range, 36-84 years; average age, 65.5 years; SD, 11.9) with 23 small renal masses (< or =4 cm) underwent laparoscopic renal lesion cryoablation. Twenty patients (22 masses) underwent follow-up MR imaging on the first day after surgery, 12 (13 masses) at 1 month, 16 (18 masses) at 3 months, 14 (15 masses) at 6 months, and 12 (12 masses) at 12 months. Three radiologists retrospectively reviewed MR images for the signal intensity, characteristics, and size of cryolesions. CT-guided needle biopsy was performed 6 months after cryoablation (18 patients) and no evidence of malignancy was discovered. RESULTS: Including all lesions at all times on T1-weighted images, cryolesion signal intensity was isointense to renal parenchyma (47/76, 61.8%) or isointense with hyper- or hypointense foci (7/76, 9.2%). On T2-weighted images, almost all lesions (72/76, 94.7%) were isointense or hypointense, and there was a hypointense rim between the cryolesion and renal parenchyma in 38.2% of lesions (29/76). A thin peripheral rim of enhancement was noted in 19.7% (14/74) of lesions. Cryolesions decreased in size an average of 61.5% (SD, 22.82; n = 12) at 1 month, 78.7% (SD, 13.5; n = 17) at 3 months, 83.5% (SD, 24.3; n = 15) at 6 months, and 94.2% (SD, 8.1; n = 11) at 1 year after cryoablation (one patient was not scanned 1 day after cryoablation and was not included in our calculations). CONCLUSION: After renal cryoablation, MR imaging revealed common signal characteristics such as low-signal-intensity rims on T2-weighted images, enhancement patterns such as thin peripheral rims, and interval size changes.

MR characteristics of oxidized cellulose (Surgicel).

OBJECTIVE: Oxidized regenerated cellulose (Surgicel) is one of the most commonly used bioabsorbable topical hemostatic agents. Surgicel may mimic an abscess on both CT and sonography when a patient undergoes imaging early in the postoperative period. The objective of our study was to describe the appearance of Surgicel on postoperative MR imaging. CONCLUSION: Surgicel has a short relaxation time on T2-weighted images, resulting in low signal intensity in the early postoperative period. MR imaging may be helpful in differentiating Surgicel from an abscess and therefore in preventing unnecessary attempts at aspiration.

Laparoscopic renal cryoablation: initial clinical series.

OBJECTIVES: To present the technique and short-term results of retroperitoneal laparoscopic renal cryoablation. METHODS: Ten patients underwent laparoscopic renal cryoablation of 11 exophytic renal tumors ranging in size from 1.5 to 3 cm identified on computed tomography. Tumors were located at the upper (3), middle (5), or lower (3) pole of the kidney. Three patients had a solitary kidney. A 3-port retroperitoneal laparoscopic approach was used to create renal cryolesions. Puncture cryoablation was performed with a 4.8-mm cryoprobe. Real-time, endoscopic, steerable, color Doppler ultrasound was used to monitor the evolving cryolesion. All patients have completed a minimum follow-up of 3 months (mean 5.5, range 3 to 9). RESULTS: Cryoablation was technically successful in all 10 patients (11 tumors). Under ultrasound guidance, the ice ball was intentionally created up to 1 cm beyond the tumor edge with the aim of achieving negative margins. Mean surgical time was 2.4 hours, cryoablation (double freeze-thaw) time 12.9 minutes, cryoprobe tip temperature -186 degrees C, and blood loss 75 mL. Systemic temperature remained unaltered. Hospital stay was less than 23 hours in 9 of 10 patients. Follow-up magnetic resonance imaging at 1 day and 1, 2, and 3 months identified the punched-out, nonenhancing, spontaneously resorbing, renal cryolesion. Follow-up biopsies of the cryoablated tumor site were negative for cancer in the 3 patients who have undergone the biopsy. CONCLUSIONS: The initial series of laparoscopic renal cryoablation is presented. The retroperitoneoscopic approach, by avoiding the peritoneal cavity, minimizes the chances of the bowel coming in contact with the evolving cryolesion, and the potential sequelae thereof. Laparoscopic renal cryoablation is currently developmental and long-term data are awaited. Nevertheless, it is potentially an attractive addition to available nephron-sparing surgical techniques.

Hepatocellular carcinoma in North America: a multiinstitutional study of appearance on T1-weighted, T2-weighted, and serial gadolinium-enhanced gradient-echo images.

OBJECTIVE: The purpose of this study was to define the common appearances of hepatocellular carcinoma (HCC) in patients in North America by analyzing T1-weighted, T2-weighted, and serial gadolinium-enhanced gradient-echo images interpreted by radiologists at multiple institutions in North America. MATERIALS AND METHODS: One hundred thirteen consecutive patients with HCC from eight institutions were included in this retrospective case series. Inclusion criteria included MR imaging examinations performed on 1.5-T MR imagers using T1-weighted breath-hold spoiled gradient-echo images, T2-weighted images, and serial gadolinium-enhanced spoiled gradient-echo images. Diagnosis was established by histology in all patients. Images were analyzed retrospectively for lesion count, lesion diameter as less than or equal to 1.5 cm and greater than 1.5 cm, and signal intensity, by individual experienced radiologists at each institution. RESULTS: We found 354 HCC lesions in the 113 patients. Tumors were solitary in 63 patients, multifocal in 45 patients, and diffuse in five patients. Lesion appearance on combined T1-weighted, T2-weighted, and immediate gadolinium-enhanced spoiled gradient-echo images was as follows: 102 lesions (29%) were hypointense on T1-weighted images, were hyperintense on T2-weighted images, and exhibited diffuse heterogeneous enhancement; 52 lesions (15%) were isointense on both T1- and T2-weighted images and exhibited diffuse homogeneous enhancement (all of these lesions measured < or = 1.5 cm in diameter); 50 lesions (14%) were hypointense on T1-weighted images, were hyperintense on T2-weighted images, and exhibited diffuse homogeneous enhancement; 33 lesions (9%) were hypointense on T1-weighted images, were hyperintense on T2-weighted images, and exhibited predominantly peripheral rim enhancement; and 27 lesions (8%) were hypointense on T1-weighted images, were isointense on T2-weighted images, and exhibited diffuse homogeneous enhancement. The remaining 90 lesions showed less common patterns. The appearance of HCCs greater than 1.5 cm and of HCCs less than or equal to 1.5 cm was significantly different (p = .001). The appearance of histologically proven HCCs is separately described. CONCLUSION: The combination of hypointensity on T1-weighted images, hyperintensity on T2-weighted images, and diffuse heterogeneous enhancement was the most common appearance of HCC on MR images in a multiinstitutional patient population in North America. Small HCCs measuring less than or equal to 1.5 cm were frequently isointense on both T1-weighted and T2-weighted images and may be detected on immediate gadolinium-enhanced images only as diffuse homogeneously enhancing lesions.

Spiral noncontrast CT versus combined plain radiography and renal US after extracorporeal shock wave lithotripsy: cost-identification analysis.

PURPOSE: To investigate the costs of spiral computed tomography (CT) versus those of combined plain radiography and renal ultrasound (US) in screening for postprocedural complications after extracorporeal shock wave lithotripsy (ESWL). MATERIALS AND METHODS: Twenty-five adult patients who had undergone ESWL were prospectively examined with spiral CT, renal US, and plain abdominal radiography. Each examination was timed, and direct technical costs were calculated by using a procedural-based cost-accounting system. The combined cost of US and plain radiography was compared with the cost of spiral CT. RESULTS: The average time for spiral CT was 15.3 minutes compared with 37.2 minutes for combined US and plain radiography. The direct technical cost of spiral CT was $36.86 compared with $57.60 for combined US and plain radiography. Average examination times were varied to assess the effect on overall costs. Within reasonable time ranges, combined US and plain radiography cannot be cost equivalent to spiral CT. CONCLUSION: Spiral CT is faster and is associated with less direct technical cost than combined US and plain radiography when used to examine patients after ESWL, given the dependence of this model on time of examination. Further studies are needed to assess the relative accuracy of these alternative approaches.

MR evaluation of the anterior cruciate ligament: value of supplementing sagittal images with coronal and axial images.

OBJECTIVE: Most studies evaluating the anterior cruciate ligament have focused on sagittal MR images for the diagnosis of injury. Limitations of sagittal images have been reported, however, including nonvisualization and incomplete visualization of the ligament. This study was undertaken to assess the value of adding coronal and axial MR images to sagittal images in the evaluation of the anterior cruciate ligament. MATERIALS AND METHODS: We reviewed oblique sagittal T1-weighted, coronal T2-weighted, and axial T2-weighted images to determine the status of the anterior cruciate ligament in 325 patients. All patients had arthroscopy. Sagittal images were initially interpreted alone and then in combination with coronal and axial images. RESULTS: Sagittal T1-weighted images alone had a 94% sensitivity and an 84% specificity for determining the status of the anterior cruciate ligament. A multiplanar evaluation of the anterior cruciate ligament resulted in a change in MR interpretation in 21 patients (6%), which led to an improved sensitivity of 98% and a specificity of 93%. Diagnostic confidence was improved in an additional 14 patients (4%). CONCLUSION: Our results show that the efficacy of MR imaging for the detection of anterior cruciate ligament tears is greater when axial and coronal images are used in combination with sagittal images than when sagittal images are used alone.