© RSNA, 2008
]]>Materials and Methods: With institutional review board approval and HIPAA compliance, this work involved detailed x-ray imaging of discarded tissue from two mastectomy specimens, together with histologic examination of the skin and subcutaneous fat and evaluation of the skin during reduction mammoplasty surgery in two patients.
Results: Comparison of the histologic findings with the findings on x-ray images demonstrated that the lucencies seen by using mammography are cutaneous caves in the underside of the dermis into which insert columns of fat projecting up from the subcutaneous adipose tissue. This finding was confirmed with evaluation of the skin during reduction surgery.
Conclusion: As far as is known, this is the first description of a complex of anatomic structures that explain radiolucencies evident by using mammography.
© RSNA, 2008
]]>Materials and Methods: A total of 1011 women who were going to undergo mammography completed a HIPAA-compliant, Institutional Review Board–approved, validated survey in one of three languages: English, Spanish, or Haitian-Creole. Data were analyzed across three groups (ie, white, black, Hispanic) by using 2 analysis.
Results: Of 911 women included in the final analysis, 378 (41%) were white, 439 (48%) were black, and 94 (10%) were Hispanic. Income and education levels varied significantly across ethnicities (P < .0001). A higher percentage of white women (40%) correctly identified the sensitivity of mammography (P < .0001), while black (42%) and Hispanic (50%) participants thought mammography to have a higher sensitivity than it truly does. After a false-positive result, only 80% of black women and 71% of Hispanic women were likely to continue with screening in the future, compared with 93% of white women (P < .0001). Only 54% of black women and 59% of Hispanic women were willing to return for a second noninvasive procedure despite the possibility of a higher cancer detection rate, compared with 76% of white women (P < .0001). Black women (53%) were also more reluctant to be recalled for invasive studies, and Hispanic women (65%) were slightly less willing to take this risk than were white women (75%) (P < .0001).
Conclusion: Differences in ethnic background appear to influence women's understanding of mammography, compliance with recall, and preference for early detection of breast cancer.
© RSNA, 2008
]]>Materials and Methods: Patients and volunteers gave informed consent, and local medical ethics committee approval was obtained. Patient data were selected retrospectively and randomly from a database of MR studies obtained between July 2006 and July 2007. The 3D three-directional VE MR images were first validated in vitro and compared with 2D one-directional VE MR images. Mitral valve (MV) and tricuspid valve (TV) flow were assessed in 10 volunteers without valve insufficiency and 20 patients with valve insufficiency, with aortic systolic stroke volume (ASSV) as the reference standard.
Results: Phantom validation showed less than 5% error for both techniques. In volunteers, 3D three-directional VE MR images showed no bias for MV or TV flow when compared with ASSV, whereas 2D one-directional VE MR images showed significant bias for MV flow (15% overestimation, P < .01). TV flow showed 25% overestimation; however, this was insignificant because of the high standard deviation. Correlation with ASSV was strong for 3D three-directional VE MR imaging (r = 0.96, P < .01 for MV flow; r = 0.88, P < .01 for TV flow) and between MV and TV flow (r = 0.91, P < .01); however, correlation was weaker for 2D one-directional VE MR imaging (r = 0.80, P < .01 for MV flow; r = 0.22, P = .55 for TV flow) and between MV flow and TV flow (r = 0.34, P = .34). In patients (mean regurgitation fractions of 13% and 10% for MV flow and TV flow, respectively), correlation between MV flow and TV flow for 3D three-directional VE MR imaging was strong (r = 0.97, P < .01).
Conclusion: Use of 3D three-directional VE MR imaging enables accurate MV and TV flow quantification, even in patients with valve regurgitation.
© RSNA, 2008
]]>Materials and Methods: Informed consent was obtained from all participants for this local ethics committee–approved study. Fifteen patients (11 male patients, four female patients; mean age, 16 years ± 4 [standard deviation]; imaging performed 16.1 years after surgery ± 3.7) and 15 age- and sex-matched control subjects (11 male subjects, four female subjects; mean age, 16 years ± 4) were evaluated. Velocity-encoded MR imaging was used to assess aortic pulse wave velocity (PWV), and a balanced turbo-field-echo sequence was used to assess aortic root distensibility. Standard velocity-encoded and multisection-multiphase imaging sequences were used to assess aortic valve function, systolic LV function, and LV mass. The two-tailed Mann-Whitney U test and Spearman rank correlation coefficient were used for statistical analysis.
Results: Patients treated with the ASO showed aortic root dilatation at three predefined levels (mean difference, 5.7–9.4 mm; P ≤ .007) and reduced aortic elasticity (PWV of aortic arch, 5.1 m/sec ± 1.2 vs 3.9 m/sec ± 0.7, P = .004; aortic root distensibility, [2.2 x 10–3] · mm Hg–1 ± 1.8 vs [4.9 x 10–3] · mm Hg–1 ± 2.9, P < .01) compared with control subjects. Minor degrees of aortic regurgitation (AR) were present (AR fraction, 5% ± 3 in patients vs 1% ± 1 in control subjects; P < .001). Patients had impaired systolic LV function (LV ejection fraction [LVEF], 51% ± 6 vs 58% ± 5 in control subjects; P = .003), in addition to enlarged LV dimensions (end-diastolic volume [EDV], 112 mL/m2 ± 13 vs 95 mL/m2 ± 16, P = .007; end-systolic volume [ESV], 54 mL/m2 ± 11 vs 39 mL/m2 ± 7, P < .001). Degree of AR predicted decreased LVEF (r = 0.41, P = .026) and was correlated with increased LV dimensions (LV EDV: r = 0.48, P = .008; LV ESV: r = 0.67, P < .001).
Conclusion: Aortic root dilatation and reduced elasticity of the proximal aorta are frequently observed in patients who have undergone the ASO, in addition to minor degrees of AR, reduced LV systolic function, and increased LV dimensions.
© RSNA, 2008
]]>Materials and Methods: This retrospective study was approved by the institutional review board, and informed patient consent was waived. The coronary CT angiograms obtained in 371 consecutive patients (268 men, 103 women; mean age, 62.9 years) with 545 stents were reviewed. All patients with stent fractures underwent conventional coronary angiography and/or fluoroscopy as part of their medical care. In phantom studies, magnified radiographs of three types of drug-eluting stents in their fully expanded, maximally bent, and unrolled states were obtained. CT angiography and fluoroscopy of a water phantom that contained two drug-eluting stents—Cypher and Taxus devices—with four fractures each were performed, and two radiologists blinded to the fracture information evaluated the images.
Results: Twenty-four stents with fractures were identified. Eighteen fractured stents (13 Cypher, four Taxus, one S670) in 14 patients were detected with CT angiography; six (33%; two Cypher, four Taxus) of these 18 stents in five (36%) patients were not detected on conventional angiograms at the initial readings. Six fractured stents showed significant (>50%) recurrent in-stent stenosis. Of 58 arteries with overlapping stent placements, eight (14%) had fractures involving 11 stents. In the in vitro studies, 57 stent fractures (31 Cypher, 26 Taxus) were detected with CT angiography and 38 (18 Cypher, 20 Taxus) were detected with fluoroscopy.
Conclusion: Coronary CT angiography depicts stent fractures in patients and phantoms, even those fractures that are not clearly depicted by conventional angiography.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: The study was approved by the local ethics committee; written informed consent was obtained from each participant. Twenty-nine patients with first onset of MI who underwent successful primary percutaneous coronary intervention at least 6 months thereafter were studied. Delayed enhancement magnetic resonance (MR) imaging was performed to define the infarct zone and its viable myocardial ratio (VMR), quantified by the percentage of the nonscarred pixels relative to the total pixels in the infarct zone. First-pass contrast material–enhanced MR imaging was performed to estimate regional perfusion and myocardial perfusion reserve (MPR) in the infarct region. Paired comparisons in perfusion and MPR were tested with nonparametric Wilcoxon matched-pairs test. A difference with P < .05 was considered significant. Correlation was tested with Pearson analysis.
Results: The infarct region showed significant impairment of regional perfusion at rest (mean, 0.966 [mL · min–1]/g ± 0.271 [standard deviation] vs 1.151 [mL · min–1]/g ± 0.282; P = .024) and during stress (mean, 1.789 [mL · min–1]/g ± 0.732 vs 2.753 [mL · min–1]/g ± 0.806; P < .0001) and a reduced MPR (mean, 1.923 ± 0.678 vs 2.486 ± 0.836; P < .0001) as compared with remote myocardium. The estimated perfusion, with stress, of the residual viable myocardium was preserved (2.993 [mL · min–1]/g ± 1.451 vs 2.753 [mL · min–1]/g ± 0.806), and the difference was not significant. Furthermore, stress perfusion (R = 0.385; P = .039) and MPR (R = 0.434; P = .018) in the infarct zone were significantly correlated with VMR, suggesting that preservation of myocardial perfusion in the infarct region reflects the amount of viable myocardium.
Conclusion: Reduced perfusion in the infarct zone is related to the extent of the viable myocardium.
© RSNA, 2008
]]>Materials and Methods: The study was approved by the local bioethics committee. Coronary microembolization was performed by injecting 300 000 microspheres into the distal portion of the left anterior descending artery in 18 anesthetized minipigs to create multifocal areas of myocardial damage. In vivo MR imaging was performed a mean of 6 hours after microembolization by using an inversion-recovery spoiled gradient-echo sequence (repetition time msec/echo time msec, 8/4; inversion time, 240–320 msec; flip angle, 20°; spatial resolution, 1.3 x 1.7 x 5.0 mm3) after injection of 0.2 mmol gadopentetate dimeglumine per kilogram of body weight. High-spatial-resolution imaging of the explanted heart was performed by using the same sequence with a higher spatial resolution (0.5 x 0.5 x 2.0 mm3). Imaging results were verified with histologic examination. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of in vivo and ex vivo images were calculated, and a t test was used to analyze observed differences.
Results: Multifocal myocardial damage was successfully induced in all animals. Areas of LGE with low SNR (mean, 36.3 ± 29.4 [standard deviation]) and CNR (23.7 ± 19.8) were observed in vivo in 12 (67%) of 18 animals, whereas ex vivo imaging revealed spotted to streaky areas of LGE with higher SNR (91.4 ± 27.8, P < .0001) and CNR (72.1 ± 25.4, P < .0001) among normal-appearing myocardium in all cases (100%). Focal myocardial lesions exceeding 5% of myocardium per slice at histologic examination were detected in vivo with a sensitivity of 83%.
Conclusion: Focal myocardial damage exceeding 5% of myocardium within the region of interest seems to be necessary for detection of LGE in vivo in an experimental model of coronary microembolization.
© RSNA, 2008
]]>Materials and Methods: MEDLINE, EMBASE, Web of Science, and Cochrane Database of Systematic Reviews databases, and reference lists of reviews and included papers were searched, without any language restrictions, for relevant articles published before March 2007. Two reviewers independently assessed study eligibility and methodologic quality by using the quality assessment of diagnostic accuracy studies checklist. A pooled random effect was estimated and a fixed coefficient regression model was used to explore the existing heterogeneity.
Results: Twenty-eight studies involving 2905 patients met the inclusion criteria. The pooled estimates of FDG PET for the detection of metastasis in the initial staging of CMM were sensitivity, 83% (95% confidence interval [CI]: 81%, 84%); specificity, 85% (95% CI: 83%, 87%); positive likelihood ratio (LR), 4.56 (95% CI: 3.12, 6.64); negative LR, 0.27 (95% CI: 0.18, 0.40); and diagnostic odds ratio, 19.8 (95% CI: 10.8, 36.4). Results from eight studies suggested that FDG PET was associated with 33% disease management changes (range, 15%–64%).
Conclusion: There is good preliminary evidence that FDG PET is useful for the initial staging of patients with CMM, especially as adjunctive role in AJCC stages III and IV, to help detect deep soft-tissue, lymph node, and visceral metastases. FDG PET–computed tomographic imaging seemed to be more precise than PET alone, as suggested by four eligible studies. Further evaluation by using a well-designed prospective study, with clinical outcome–focused measures and cost effectiveness analysis, is needed to clarify the appropriate role of FDG PET in CMM staging.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: All experiments were approved by the Institutional Animal Care and Use Committee. Phantom studies were performed to quantify microsphere relaxivity and volume susceptibility properties and compare image contrast patterns resulting from aggregate deposition of unlabeled and iron-labeled microspheres. In seven rabbits in which nine VX2 liver tumors were implanted, T2*-weighted gradient-echo (GRE) MR images with negative image contrast (NC), white-marker (WM) GRE images with positive image contrast (PC), and on-resonance water-suppression turbo spin-echo (SE) images with PC were obtained before and after catheter-directed administration of microspheres into the hepatic artery. During each injection, serial GRE acquisitions were performed for real-time visualization of microsphere delivery. Contrast-to-noise ratios (CNRs) were measured between regions of microsphere accumulation and regions of normal liver parenchyma that demonstrated no apparent microsphere accumulation. Pre- and postinjection CNR measurements at identical spatial positions were compared by using paired t test ( = .05).
Results: Conventional microspheres did not produce detectable image contrast in phantoms. Iron-labeled microspheres produced susceptibility-induced dipole patterns with spatial extent of image contrast increasing with increasing microsphere dose. Real-time image series depicted both preferential delivery to tumor tissues and nontargeted delivery to adjacent organs. T2*-weighted GRE, WM GRE, and on-resonance water-suppression turbo SE each permitted in vivo visualization of the microsphere deposition, with postinjection CNR values (mean, 14.29 ± 3.98 [standard deviation], 1.87 ± 0.93, and 19.30 ± 8.72, respectively) significantly greater than corresponding preinjection CNR values (mean, 2.02 ± 4.65, 0.02 ± 0.27, 0.85 ± 2.65, respectively) (P < .05).
Conclusion: Microsphere tracking during radioembolization may permit real-time verification of delivery and detection of extrahepatic shunting.
© RSNA, 2008
]]>Materials and Methods: The handling methods used for all the animals in this study were approved by the institutional committee for animal experiments. Diffusion-tensor MR imaging was performed in four healthy common marmosets, and in two of these animals, manganese-enhanced MR imaging tract tracing was performed by using a 7.0-T MR imaging unit. The visual pathways were quantitatively investigated in terms of the manganese distribution observed on the manganese-enhanced MR images. The images obtained with DTT and manganese-enhanced MR imaging tract tracing were qualitatively compared, and the features of the visual pathway were verified through fusion of the reconstructed images obtained by using these two modalities.
Results: DTT provided information regarding the neuroanatomic features of the marmoset visual pathway and revealed the bilateral branching patterns of the typical primate retinogeniculate pathways, although several incorrectly tracked fibers were noted. The distribution of manganese on the manganese-enhanced MR images revealed bilateral innervation of the retinal projections and depicted the layered internal structure of the lateral geniculate nuclei bilaterally, depending on the ocularity of each layer. These morphologic findings were consistent with those of previous histopathologic studies.
Conclusion: The findings of this preliminary study raise the possibility that DTT is useful for visualizing the neuronal fiber trajectories in primate visual pathways.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: The study was approved by the institutional review board, and informed consent was obtained from all patients. In 40 patients (27 women, 13 men; age range, 33–94 years; mean age, 62.5 years) in whom colorectal cancer (n = 20) and breast cancer (n = 20) had metastasized to the liver, initial tumor volume and thermal-induced necrosis after MR-guided laser-induced thermotherapy were retrospectively analyzed. All patients presented with oligonodular liver metastases and underwent follow-up with contrast-enhanced MR imaging for at least 3 years. No concomitant oncologic therapies were performed.
Results: Volumetric MR imaging evaluation depicted 40 metastases with an initial tumor volume less than 5 mL (x = 1.75), nine metastases with initial volume of 5–20 mL (x = 12.35), and eight metastases with initial volume more than 20 mL (x = 50.57). The mean volume of the thermally damaged area was 498% of the initial volume for colorectal cancer metastases and 604% of the initial volume for breast cancer metastases. The ischemic and necrotic volume for colorectal cancer metastases had decreased by a mean of 48.6% after 3 months, by 63% after 6 months, by 70.2% after 12 months, and by 92.2% after 36 months. For breast cancer metastases at 36 months, the necrotic volume had decreased by 80.61%; the reduction in the volume of the thermally damaged region was statistically significantly lower than that of colorectal cancer metastases.
Conclusion: MR-guided laser-induced thermotherapy induced a high volume of thermal ablation; the greatest reduction in the necrotic volume occurred in the first year, and lower values were seen in the next period. The reduction was statistically significantly higher in colorectal cancer metastases.
© RSNA, 2008
]]>Materials and Methods: Institutional review committee approval and patients' written informed consent were obtained. CT images of the abdomen were obtained by using 2 mL per kilogram of BW of intravenous contrast material (300 mg/mL iodine) injected at 4 mL/sec in 161 consecutive patients (age range, 28–90 years; mean age, 63 years; 95 men, 66 women). CT scans were initiated 45 and 150 seconds after aortic enhancement increased by 50 HU. The patients were divided into low (37–54 kg) and high (55–75 kg) BW groups. The HU/I, where HU is change in CT number and I is iodine dose in grams, and adjusted maximum hepatic enhancement (HU/[I/kg]) were assessed for correlation with BW, body mass index (BMI), and body fat percentage (BFP) by using linear regression.
Results: HU/I correlated (P < .001) inversely with BW in the aorta (r = –0.78) and liver (r = –0.80) and with BMI in the aorta (r = –0.59) and liver (r = –0.61) on portal venous phase images. Regression formula for the low BW group was HU/I = 4.1 – .044 · BW (P < .001) and for the high BW group was HU/I = 2.7 – .017 · BW (P < .001), suggesting that the amount of contrast material required with increased BW is less in the high BW group. Adjusted maximum hepatic enhancement directly correlated with BFP (r = 0.25, P < .01).
Conclusion: Excessive contrast material may inadvertently be given in heavier patients when the dose is determined by patient BW. Contrast material dose may need to be tailored in individual patients by using BFP.
© RSNA, 2008
]]>Materials and Methods: This study received institutional review board approval and was HIPAA compliant. Informed consent was waived. Eight subjects (four men, four women; median age, 40 years; range, 35–57 years), all potential renal donors with no pancreatic pathologic abnormalities, underwent abdominal CT imaging, which resulted in 30 10-mm-thick sections obtained at a single level. Imaging was a direct result of bolus timing employed for standard renal donor protocol; no additional imaging beyond what was clinically warranted was performed. Images were obtained every 3 seconds; scanning was initiated at the onset of contrast material administration. Region-of-interest measurements were obtained for the pancreatic body and the aorta to generate time-enhancement curves (TECs). A one-compartment model was applied by using the aortic and pancreatic TECs as the input and output functions, respectively. Pancreatic volumetric blood flow FV, volume of distribution VD, and blood transit time were determined. Modeled pancreatic TECs were generated and were compared with actual TECs for wellness of fit.
Results: Pancreatic FV values from the single-compartment model ranged from 0.961 to 6.405 min–1 (mean, 3.560 min–1 ± 1.900 [standard deviation]). Volume of distribution VD ranged from 1.491 to 3.080 (mean, 2.383 ± 0.638), while values of ranged from –3.090 to 6.436 seconds (mean, 0.481 second ± 3.000). Modeled pancreatic TECs closely matched true pancreatic TECs for each subject, with R2 values ranging from 0.840 to 0.959.
Conclusion: A simple one-compartment kinetic model can be applied to contrast-enhanced images of normal pancreas to yield accurate pancreatic TECs, which attest to the perfusion parameters obtained. In addition to yielding volumetric blood flow similar to that of other models of tissue perfusion, two additional pancreatic perfusion parameters can be obtained.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: This HIPAA-compliant study was approved by the institutional review board; informed consent was waived. Ninety-five patients (58 female, 37 male; mean age, 51 years; range, 15–91 years) underwent MRCP by using the respiratory-triggered isotropic 3D fast-recovery FSE sequence and endoscopic or percutaneous direct visualization between March 2003 and June 2007. Two independent readers evaluated the MRCP images for strictures, dilatation, and intraductal filling defects. Sensitivity, specificity, and interobserver agreement ( statistics) were determined.
Results: The respective sensitivity and specificity for strictures, dilatation, and intraductal filling defects (all choledocholithiasis) were 86% (40 of 47) and 94% (45 of 48), 98% (57 of 58) and 100% (37 of 37), and 68% (19 of 28) and 97% (65 of 67) for reader 1 and 88% (41 of 47) and 94% (45 of 48), 96% (56 of 58) and 100% (37 of 37), and 75% (21 of 28) and 99% (66 of 67) for reader 2. The sensitivity for stones larger than 3 mm was 94% (15 of 16) for reader 1 and 100% (16 of 16) for reader 2, whereas the sensitivity for stones 3 mm or smaller was 33% (four of 12) for reader 1 and 42% (five of 12) for reader 2. Agreement between readers was good to excellent, with values of 0.76, 0.85, and 0.98 for strictures, dilatation, and choledocholithiasis, respectively.
Conclusion: MRCP by using the respiratory-triggered isotropic 3D fast-recovery FSE sequence with parallel imaging demonstrates excellent diagnostic capabilities for possible biliary disease, although it is limited for stones 3 mm or smaller in size.
© RSNA, 2008
]]>Materials and Methods: The institutional review board approved this retrospective study; informed consent was waived. IVIM DW imaging was tested on three alkane phantoms, on which the signal-intensity decay curves according to b factors were logarithmically plotted. Ten b factors (0, 10, 20, 30, 50, 80, 100, 200, 400, 800 sec/mm2) were used in patients. Patients with documented liver cirrhosis (cirrhotic liver group, n = 12) and patients without chronic liver disease (healthy liver group, n = 25) were included. The mean liver D, D*, and f values were measured and compared with the apparent diffusion coefficient (ADC) computed by using four b values (0, 200, 400, 800 sec/mm2). Liver ADC and D, f, and D* parameters were compared between the cirrhotic liver group and healthy liver group. Means were compared by using the Student t test.
Results: Signal-intensity decay curves were monoexponential on phantoms and biexponential in patients. In vivo, mean ADC values were significantly higher than D in the healthy liver group (ADC = 1.39 x 10–3 mm2/sec ± 0.2 [standard deviation] vs D = 1.10 x 10–3 mm2/sec ± 0.7) and in the cirrhotic liver group (ADC = 1.23 x 10–3 mm2/sec ± 0.4 vs D = 1.19 x 10–3 mm2/sec ± 0.5) (P = .03). ADC and D* were significantly reduced in the cirrhotic liver group compared with those in the healthy liver group (respective P values of .03 and .008).
Conclusion: Restricted diffusion observed in patients with cirrhosis may be related to D* variations, which reflect decreased perfusion, as well as alterations in pure molecular water diffusion in cirrhotic livers.
© RSNA, 2008
]]>Materials and Methods: Eighteen patients (median age, 61 years; range, 44–72 years) gave informed consent for this retrospective Research Ethics Board–approved study. Prior to radical prostatectomy, ADC (b value, 600 sec/mm2) and T2 maps were obtained by using 1.5-T magnetic resonance (MR) imaging. Twenty-eight peripheral zone (PZ) tumors were reviewed by using whole-mount histologic findings, and regions assessed to contain primarily (>60%) normal PZ tissue were delineated. Tumors were categorized as "sparse" if more than 50% of their cross-sectional areas were these primarily normal PZ regions and were considered "dense" otherwise. Normal PZ tissue was outlined separately on the same section. Tumor and normal tissue outlines were transferred to corresponding ADC and T2 maps, and median values were calculated. Values were compared by using multiple regression analysis. Matched-pair tumor-to–normal tissue differences and log2-transformed ratios were assessed by using nonparametric tests.
Results: Thirty-six percent (10 of 28) of tumors were sparse; 64% (18 of 28) were dense. For both overall and intrapatient comparisons, dense tumors had significantly lower ADC and T2 values than normal PZ tissue (P < .05), but no significant differences were observed between sparse tumors and normal tissue. Log2-transformed tumor-to–normal tissue ratios were significantly less than zero for dense tumors for both ADC and T2 (P < .01) measurements but not for sparse tumors. Both matched-pair differences and log2-transformed ratios were significantly different between sparse and dense tumors (P < .01). ADC and T2 values were moderately correlated (Pearson correlation coefficient range, r = 0.770–0.804).
Conclusion: Sparse prostate tumors have similar ADC and T2 values to those of normal PZ tissue. This may limit MR imaging detection and the assessment of tumor volume of some cancers.
© RSNA, 2008
]]>Materials and Methods: This retrospective study was approved by the institutional review board, and the informed consent requirement was waived. The authors analyzed MR images of 50 lesions (36 benign, 14 malignant) in 47 patients. DW MR imaging and dynamic contrast-enhanced MR imaging were performed in all patients. Time-intensity curve (TIC) patterns were categorized as follows: type A, time to peak was more than 120 seconds; type B, time to peak was 120 seconds or less with high washout ratio (≥30%); type C, time to peak was 120 seconds or less with low washout ratio (<30%); and type D, flat. The apparent diffusion coefficient (ADC) values were measured on DW MR images. Sensitivity, specificity, accuracy, and positive and negative predictive values were calculated for type A, B, and D tumors regarded as benign and for type C tumors regarded as malignant. On the basis of DW MR imaging results, ADC threshold values between pleomorphic adenomas and carcinomas and between Warthin tumors and carcinomas were selected. Diagnostic accuracy was compared before and after modification diagnosis referring to the ADC value obtained with the McNemar test. P < .05 was considered to indicate a significant difference.
Results: ADC threshold values were 1.4 x 10–3 mm2/sec between pleomorphic adenomas and carcinomas and 1.0 x 10–3 mm2/sec between Warthin tumors and carcinomas. Accuracy (82% vs 94%) and positive predictive value (67% vs 92%) significantly improved with the addition of ADC values in the evaluation of patients with type B or C tumors.
Conclusion: A persistent or flat TIC pattern on dynamic contrast-enhanced MR images indicates benign disease, but there is added value from including the ADC value in the evaluation of tumors that show a plateau or washout TIC pattern.
© RSNA, 2008
]]>Materials and Methods: Six databases were systematically searched for CEA reports published between 1985 and 2005. All imaging-related studies were selected and grouped according to year, country, and journal of publication, as well as imaging modality and disease being studied. Two readers with formal training in decision analysis and CEA used a seven-point (1, low; 7, high) Likert scale based on reasonableness of assumptions, quality of presentation, and adherence to guidelines to independently evaluate study quality. Quality scores according to year, country, and journal of publication were compared with the unpaired Student t test.
Results: The first radiology-related CEA was published in 1985; 111 radiology-related CEAs were published between 1985 and 2005. The average number of studies increased from 1.6 per year between 1985 and 1995 to 9.4 per year between 1996 and 2005. Eighty-six studies were performed to evaluate diagnostic imaging technologies, and 25 were performed to evaluate interventional imaging technologies. Ultrasonography (35.0%), angiography (31.5%), magnetic resonance imaging (22.5%), and computed tomography (19.8%) were evaluated most frequently. Forty-nine studies received government funds; 42 did not disclose the source of funding. The mean quality score was 4.23 ± 1.12 (standard deviation), without significant improvement over time. Scores in studies performed in the United States were significantly higher than scores in studies that were not performed in the United States (4.45 ± 1.02 vs 3.61 ± 1.17, respectively; P < .01). Scores were also higher in journals with three or more CEA articles published during the study period than in journals with two or fewer CEA articles published during this period (4.54 ± 1.09 vs 3.91 ± 1.06, respectively; P < .01).
Conclusion: CEAs are an important tool with which to analyze the value of diagnostic imaging. However, improvement in the quality of analyses is needed.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: A geometric phantom emulating the attenuation and scatter properties of the adult human thorax was employed to assess eight imaging systems in a total of nine configurations. The noise power spectrum (NPS) was derived from images of the phantom acquired at three exposure levels spanning the operating range of the system. The modulation transfer function (MTF) was measured by using an edge device positioned at the anterior surface of the phantom. Scatter measurements were made by using a beam-stop technique. All measurements, including those of phantom attenuation and estimates of x-ray flux, were used to compute the eDQE.
Results: The MTF results showed notable degradation owing to focal spot blur. Scatter fractions ranged between 11% and 56%, depending on the system. The eDQE(0) results ranged from 1%–17%, indicating a reduction of up to one order of magnitude and different rank ordering and performance among systems, compared with that implied in reported conventional detective quantum efficiency results from the same systems.
Conclusion: The eDQE method was easy to implement, yielded reproducible results, and provided a meaningful reflection of system performance by quantifying image quality in a clinically relevant context. The difference in the magnitude of the measured eDQE and the ideal eDQE of 100% provides a great opportunity for improving the image quality of radiographic and mammographic systems while reducing patient dose.
© RSNA, 2008
]]>Materials and Methods: The study was approved by the institutional review board and complied with HIPAA guidelines. Informed consent was obtained. Twenty adolescent girls, 10 with mild AN (mean age, 15.9 years; range, 13–18 years) and 10 age- and sex-matched normal-weight control subjects (mean age, 15.9 years; range, 12–18 years) underwent flat-panel volume CT of distal radius to determine apparent trabecular bone volume fraction (BV/TV), apparent trabecular number (TbN), apparent trabecular thickness (TbTh), and apparent trabecular separation (TbSp). All subjects underwent DXA of spine, hip, and whole body to determine BMD and body composition. The means and standard deviations (SDs) of structure parameters were calculated for AN and control groups. Groups were compared (Student t test). Linear regression analysis was performed.
Results: AN subjects compared with control subjects, respectively, showed significantly lower mean values for BV/TV (0.37% ± 0.05 [SD] vs 0.46% ± 0.03, P = .0002) and TbTh (0.31 mm ± 0.03 vs 0.39 mm ± 0.03, P < .0001) and higher mean values for TbSp (0.54 mm ± 0.13 vs 0.44 mm ± 0.04, P = .02). TbN was lower in AN subjects than in control subjects, but the difference was not significant (1.17 mm–3 ± 0.15 vs 1.22 mm–3 ± 0.07, P = .43). There was no significant difference in BMD between AN and control subjects. BMD parameters showed positive correlation with BV/TV and TbTh in the control group (r = 0.55–0.84, P = .05–.01) but not in AN patients.
Conclusion: Flat-panel volume CT is effective in evaluation of trabecular structure in adolescent girls with AN and demonstrates that bone structure is abnormal in these patients compared with that in normal-weight control subjects despite normal BMD.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: Institutional review board approval was obtained; informed consent was waived. The study included 57 patients (36 women [mean age, 37 years], 21 men [mean age, 32 years]) who underwent MR arthrography and either surgery or arthroscopy as reference standard. On MR images, location of sublabral contrast material interposition and depth, shape, and extension into the labral substance of contrast material interpositions were described. Abnormal labral signal intensity (areas of high signal intensity), acetabular cartilage lesions, osseous abnormalities, and perilabral cysts were noted. Mann-Whitney U and Fisher exact tests were performed; interobserver agreement was calculated ( statistic and intraclass correlation coefficient).
Results: Surgical procedures revealed that 10 (18%) of 57 patients had recesses and 44 (77%) of 57 had tears. Locations of recesses and tears, respectively, were as follows: seven and none, in the 8-o'clock position; two of each, in the 9-o'clock position; one and 22, in the 10-o'clock position; and none and 20, in the 11-o'clock position. None of the recesses extended into the substance of the labrum or through the full thickness of the labral base; 51% (22 of 43) of tears extended into the substance and 49% (21 of 43) of tears extended along the entire labral base. Shape of sublabral contrast material interposition was linear in five (83%) of six recesses and 21 (49%) of 43 tears. Recesses were not associated with abnormal signal intensity of the labrum, cartilage lesions, osseous abnormalities, or perilabral cysts. Of 43 tears, 32 (74%) were associated with abnormal signal of the labrum; 23 (53%), with cartilage damage; 11 (26%), with osseous abnormalities; and eight (19%), with perilabral cysts.
Conclusion: Recesses occur as normal variants in the anteroinferior part of the acetabulum. Location in the 8-o'clock position, linear shape of contrast material interposition, partial separation of the labrum, and absence of perilabral abnormalities are characteristics of a recess.
© RSNA, 2008
]]>Materials and Methods: This retrospective study was performed with institutional review board committee approval by using a waiver of informed consent. Sixteen immunocompromised patients (nine males, seven females; age range, 2–39 years) underwent magnetic resonance (MR) imaging and cerebrospinal fluid polymerase chain reaction (PCR) testing for HHV6 DNA on the basis of clinical findings suspicious for encephalitis. MR images acquired during acute illness were examined without knowing PCR results.
Results: Nine patients were HHV6 positive. Seven showed signal intensity abnormalities, with prominent involvement of the hippocampus, and six showed additional involvement of the amygdala. Three HHV6-positive patients showed signal intensity abnormality in extrahippocampal divisions of the olfactory cortex and cortical and subcortical structures that maintain prominent connections with the hippocampal formation. Among the seven HHV6-negative patients, six had abnormalities in the hippocampus but only two showed extrahippocampal involvement, which was restricted to the amygdala.
Conclusion: Most patients with HHV6 encephalitis have signal intensity abnormalities in the hippocampal formation and amygdala and, contrary to prior reports, some also have involvement of limbic structures outside of the medial temporal lobe. The presence of MR signal intensity abnormality in the medial temporal lobe should raise the diagnosis of HHV6 encephalitis in immunosuppressed patients, especially when hyperintense lesions are seen in the insular region and inferior frontal lobe.
© RSNA, 2008
]]>Materials and Methods: The local ethics committee approved the study, and informed consent was obtained from all participants. From 12 unrelated families, 25 patients who were NOTCH3 mutation carriers and 13 who were non–mutation carriers were examined clinically and with standardized MR imaging at baseline and after 7 years. The progression of white matter hyperintensities (WMHs), lacunar infarcts, microbleeding, and brain volume loss was measured semiquantitatively. Correlation testing and group comparison testing were performed to identify the risk factors associated with increased progression of CADASIL-related MR abnormalities.
Results: Compared with the non–mutation carriers, the mutation carriers showed significant increases in numbers of lacunar infarct (P < .01), WMH (P < .01), and microbleed (P < .05) lesions but no increased loss of brain volume. The distributions of new WMHs and new lacunar infarcts at follow-up were similar to the distributions of these abnormalities at baseline. High WMH (P < .05), lacunar infarct (P < .01), and microbleed (P < .01) lesion loads at baseline—but not cardiovascular risk factors—were associated with faster progression of these abnormalities.
Conclusion: Patients with CADASIL who have a high MR abnormality lesion load at baseline are at risk for faster progression of MR abnormalities.
© RSNA, 2008
]]>Materials and Methods: Three hundred sixty-one brain and spine MR imaging and MR angiographic examinations that were ordered emergently after hours and given preliminary interpretations by radiology residents were retrospectively reviewed from December 1, 2006 to May 31, 2007 with institutional review board approval. Discrepancies between the interpretations of radiology residents and the final reports of attending neuroradiologists were classified as either false-negative (FN, failure to recognize abnormalities) or false-positive (FP, misinterpreting normal images as abnormal). Discrepancies that could affect patient care or clinical outcome were considered major.
Results: Overall, the agreement rate was 92.8%, the overall discrepancy rate was 7.2%, the major disagreement rate was 4.2%, and the minor disagreement rate was 2.2%. Misinterpretations among 1st-year residents on call were significant (P < .04) when compared with more senior-level residents. There were 23 FN interpretations. The most common misses were acute stroke (n = 3), aneurysm (n = 3), vascular occlusion (n = 3), and disk herniation (n = 2). There were only three FP interpretations (misdiagnoses of syrinx, arachnoiditis, and acute infarct).
Conclusion: There was no adverse clinical outcome as a result of misinterpretations, owing in part to rapid turnaround time for final reporting. Level of residency training has a significant effect on the rate of discrepancy, which may be mitigated by recent changes regarding 1st-year radiology residents' overnight call.
© RSNA, 2008
]]>Materials and Methods: Institutional review board approval was obtained; informed consent was waived for this HIPAA-compliant study. One hundred twenty consecutive patients who underwent 3.0-T three-dimensional high-spatial-resolution contrast material–enhanced MR angiography of the supraaortic arteries with an identical acquisition protocol were assigned to either the high-dose (0.154 mmol per kilogram of body weight), intermediate-dose (0.097 mmol/kg), or low-dose (0.047 mmol/kg) group. Two readers evaluated resulting images for arterial definition, venous contamination, and arterial stenosis. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were evaluated at six consistent sites. Statistical analysis was performed with the Kruskal-Wallis, Wilcoxon signed rank, and analysis of variance tests and the coefficient.
Results: Readers 1 and 2 scored vascular definition as excellent or sufficient for diagnosis in 1311 of 1360 segments and in 1313 of 1360 segments in the high-dose group ( = 0.73), in 1321 of 1354 and in 1319 of 1354 segments in the intermediate-dose group ( = 0.77), and in 1322 of 1350 and in 1320 of 1350 segments in the low-dose group (= 0.66), respectively. Arterial occlusive disease was detected by reader 1 in 52, 27, and 98 segments in the high-, intermediate-, and low-dose groups, respectively. Arterial occlusive disease was detected by reader 2 in 48, 25, and 100 segments in high-, intermediate-, and low-dose groups, respectively. No significant difference existed among the three groups regarding arterial definition scores (reader 1, P = .21; reader 2, P = .25) and venous contamination scores (reader 1, P = .38; reader 2, P = .35). SNRs and CNRs were lower in the low-dose group (P < .01).
Conclusion: At 3.0 T, high-spatial-resolution MR angiography of the supraaortic arteries can be performed with contrast agent doses as low as 0.047 mmol/kg, without compromising image quality, acquisition speed, or spatial resolution.
Supplemental material:
© RSNA, 2008
]]>Materials and Methods: The study protocol was approved by the institutional review board. Signed informed consent was obtained from all subjects. In vitro testing at 3 T was performed with 32 PMs and 45 PM leads that were evaluated for force and torque (by using a floating platform) and radiofrequency (RF)-related heating by using a transmit-receive head coil (maximum specific absorption rate, 3.2 W/kg). Patient examinations at 3 T were performed in 44 patients with a cardiac PM and a strong clinical need; patients underwent a total of 51 MR examinations of the brain by using a transmit-receive head coil to minimize RF exposure of the PM system. An electrocardiograph and pulse oximetry were used for continuous monitoring during MR imaging. The technical and functional PM status was assessed prior to and immediately after MR imaging and at 3 months thereafter. Serum troponin I level was measured before and 12 hours after imaging to detect myocardial thermal injury. PM reprogramming was performed prior to MR imaging depending on the patient's intrinsic heart rate (<60 beats per minute, asynchronous pacing; ≥60 beats per minute, sense-only mode).
Results: For in vitro testing, the maximum translational force was 2150 mN (mean, 374.38 mN ± 392.75 [standard deviation]), and maximum torque was 17.8 x 10–3 N · m (mean, [2.29 ± 4.08] x 10–3 N · m). The maximum temperature increase was 2.98°C (mean, 0.16°C ± 0.45). For patient examinations, all MR examinations (51 of 51) were completed safely. There were no significant (P < .05) changes in lead impedance, pacing capture threshold level, or serum troponin I level.
Conclusion: MR imaging of the brain at 3 T in patients with a cardiac PM can be performed safely when dedicated safety precautions (including the use of a transmit-receive head coil) are taken.
© RSNA, 2008
]]>Materials and Methods: Institutional review board approval was obtained, and the informed consent requirement was waived for this HIPAA-compliant retrospective study. Radiation exposure and fluoroscopy time during VCUG were reviewed in 145 children (75 girls, 70 boys; age range, 3 days to 8 years) who underwent GCPFL or CFL between 2001 and 2002. Children were grouped on the basis of the fluoroscopy unit used and their supine anteroposterior abdominal diameter (group 1, 8.0–8.5-cm diameter; group 2, 10–11-cm diameter; group 3, 12–13-cm diameter). Analysis of variance was used to compare radiation exposure and fluoroscopy time between fluoroscopy units and patient diameter groups. Effective doses were calculated and compared for both fluoroscopes and for estimated RNC dose values.
Results: GCPFL resulted in a significant reduction in total radiation exposure, which was at least eight times lower than that with CFL in all three groups (P < .001 for all). There was no significant difference in fluoroscopy time (P > .50). Effective radiation doses from GCPFL were approximately one order of magnitude lower than those from CFL but one order of magnitude higher than those from RNC.
Conclusion: In children, VCUG can be performed with a GCPFL unit that delivers radiation exposures that are at least eight times lower than those delivered by a conventional CFL unit.
Supplemental material:
© RSNA, 2008
]]>© RSNA, 2008
]]>© RSNA, 2008
]]>© RSNA, 2008
]]>Materials and Methods: The Regional Ethical Review Board approved this study, and all participants gave informed consent. Four thoracic radiologists acted as observers in a jackknife free-response receiver operating characteristic (JAFROC) study conducted in 42 patients with and 47 patients without pulmonary nodules examined with chest tomosynthesis and chest radiography. Multidetector CT served as reference method. The observers marked suspected nodules on the images by using a four-point rating scale for the confidence of presence. The JAFROC figure of merit was used as the measure of detectability. The number of lesion localizations relative to the total number of lesions (lesion localization fraction [LLF]) and the number of nonlesion localizations relative to the total n