Characterization of adrenal adenomas and metastases: correlation between unenhanced computed tomography and chemical shift magnetic resonance imaging

PURPOSE: To evaluate the correlation of absolute attenuation values of unenhanced computed tomography (CT) with signal intensity (SI) quantitative analysis on chemical shift (CS) magnetic resonance (MR) imaging in differentiating adrenal adenomas from metastases. MATERIAL AND METHODS: Forty-one adrenal masses (27 adenomas, 14 metastases) were studied with CS MR imaging and unenhanced CT. MR included T1-weighted breathhold gradient-echo in-phase (IP) and opposed-phase (OP) sequences. The SI index (SI-i) [(SIIP-SIOP/SIIP)] x 100% and chemical-shift ratio (CS-r) relative to the spleen [(SIlesion/ SIspleen)OP/(SIlesion/SIspleen)IP] were calculated for each lesion. CT absolute attenuation values were also determined. RESULTS: The mean attenuation value of metastases was significantly greater than that of adenomas (< 0.0001). On MR, the mean SI-i of adenomas was significantly greater than that of metastases (P < 0.0001) and no overlaps were evident. The CS-r of malignant and benign lesions overlapped considerably, and five adenomas (all with indeterminate Hounsfield Unit values at CT) were misclassified as potentially malignant. CT attenuation values were significantly correlated with both MR quantitative analyses. CONCLUSION: Since CS MR imaging and CT both depict the presence of lipids within adrenal lesions, absolute attenuation values are highly correlated with MR quantitative analysis. SI-i is the most reliable tool for differentiating adrenal adenomas from metastases, showing better accuracy than lesion-to-spleen CS-r, in particular for adenomas with indeterminate absolute attenuation values.     

Differentiation of adrenal adenomas from adrenal metastases in single-phased staging dual-energy CT and radiomics

PURPOSE Differentiation of incidental adrenal lesions remains a challenge in diagnostic imaging, especially on single-phase portal venous computed tomography (CT) in the oncological setting. The aim of the study was to explore the ability of dual-energy CT (DECT)-based iodine quantification and virtual non-contrast (VNC) imaging and advanced radiomic analysis of DECT for differentiation of adrenal adenomas from metastases. METHODS A total of 46 patients with 49 adrenal lesions underwent clinically indicated staging DECT and magnetic resonance imaging. Median values of quantitative parameters such as VNC, fat fraction, and iodine density in DECT images were collected and compared between adenomas and metastases using non-parametric tests. Magnetic resonance imaging, washout CT, and clinical follow-up were used as a reference standard. Diagnostic accuracy was assessed by calculating receiver operating characteristics. A DECT tumor analysis prototype software was used for semiautomatic segmentation of adrenal lesions and extraction of radiomic features. A radiomics prototype was used to analyze the data with multiple logistic regression and random forest classification to determine the area under the curve (AUC).RESULTS The study cohort (60.87% women; mean age: 66.91 ± 12.93 years) consisted of 32 adenomas and 17 metastases. DECT-based VNC imaging (AUC = 0.89) and fat quantification (AUC = 0.86) differentiate between adrenal adenomas and metastases with high diagnostic accuracy (P < .001). Analysis of radiomic features revealed that DECT features such as VNC imaging and fat fraction (AUC = 0.87-0.89; < .001) and radiomic features such as 90th percentile and total energy (AUC = 0.88-0.93; P < .001) differentiate with high diagnostic accuracy between adrenal adenomas and metastases. Random forest classification revealed an AUC of 0.83 for separating adrenal adenomas from metastases. CONCLUSION Virtual non-contrast imaging and fat quantification as well as extraction of radiomic features accurately differentiate between adrenal adenomas and metastases on single-phase oncologic staging DECT.     

Differentiation of malignant from benign adrenal masses: predictive indices on computed tomography

CT findings in 43 adrenal masses were analyzed to see which features correlated most significantly with malignancy. Size, contrast enhancement, and consistency emerged as important discriminators of malignant from benign adrenal masses. These three factors were further analyzed by logistic regression technique to examine the joint influence of computed tomographic (CT) features in prediction of malignancy. As a result of logistic regression analysis, a table of estimated probability of malignancy as a function of tumor size alone and another table of estimated probability as a joint function of size and contrast enhancement were developed. Given a similar patient sample and by using the data given, it would be possible to predict chances of malignancy in an adrenal mass if its CT features are known; for example, a 5-cm adrenal mass without enhancement has a 0.31 probability of malignancy; with enhancement, a 0.68 probability.     

Adrenal masses falsely diagnosed as adenomas on unenhanced and delayed contrast-enhanced computed tomography: Pathological correlation

Objectives To assess the accuracy of CT for the diagnosis of histologically confirmed adrenal adenoma and nonadenoma using CT numbers. Materials and methods Our study included 91 adrenal masses in 83 patients; histopathological diagnoses were 45 adenomas, 31 pheochromocytomas, 6 hyperplasias, 4 metastasis, and 5 miscellaneous lesions. Unenhanced CT in 46 patients and unenhanced and delayed contrast-enhanced (DCE) CT in 37 patients were retrospectively reviewed to examine the correlation between CT findings and those on pathological examination and to obtain diagnostic accuracy. Results Sensitivity, specificity, and accuracy for adenoma were 40% (18/45), 91% (42/46), and 66% (60/91) with unenhanced CT, and 96% (24/25), 61% (11/18), and 81% (35/43) with DCE CT. Adrenal masses falsely diagnosed as adenoma on unenhanced CT included three hyperplasias and one endothelial cyst, and those falsely diagnosed as adenoma on DCE CT were five pheochromocytomas, one oncocytic coritical tumor, and one primary pigmented nodular adrenocortical dysplasia. Twenty-five lipid-poor adenomas were falsely diagnosed as nonadenomas on unenhanced CT and one degenerated adenoma both on unenhanced CT and on DCE CT. Conclusion Diagnosing adenoma merely on CT numbers can lead to misdiagnosis. The lower specificity than expected is due to pheochromocytomas presenting as false positives. An erratum to this article can be found at     

Incidental adrenal lesions: Accuracy of quadriphasic contrast enhanced computed tomography in distinguishing adenomas from nonadenomas

Purpose

To evaluate the accuracy in distinguishing adrenal adenomas from nonadenomas by means of quadriphasic CT exam, including unenhanced (UE), arterial enhanced (AE), portal enhanced (PE) and 5-min delayed enhanced (DE) CT scans.

Methods

This retrospective study had institutional review board approval; the need for informed consent was waived. From September 2007 to September 2009, 104 adrenal masses were evaluated in 87 patients (49 M, 38 F, mean age 58.4 years) undergoing UE, AE (35-s delay), PE (80-s delay) and DE (5-min delay) CT scans. The mean adrenal attenuation during all imaging phases was measured by two readers. The accuracy values of absolute unenhanced attenuation (UE), absolute wash-out (AWO), relative percentage wash-out (RPWO) and percentage enhancement wash-out (PEW) were assessed by using receiver operator curves (ROC) analysis. The overall accuracy of the quadriphasic protocol and other triphasic protocols were evaluated. A value of p ≤ 0.05 was considered significant.

Results

The accuracy in characterizing adrenal lesions was 86.5% (90/104) for UE attenuation (≤10 HU threshold), 90.1% (82/91) for RPWO (≥30% threshold), 85.7% (78/91) for AWO (≥12 HU threshold) and 83.5% (76/91) for PEW (≥30% threshold), respectively. Quadriphasic CT (accuracy 97.1%, 101/104) performed better than triphasic CT including only AE scan (efficiency 90.0%, 94/104; p = 0.011) and triphasic CT including only PE scan (efficiency 96.1%, 100/104; p = 0.025).

Conclusion

Quadriphasic CT protocol including 5-min DE scan may be used to characterize incidentally detected adrenal masses. RPWO represented the best wash-out parameter for characterizing adrenal lesions.     

Differentiation of hemopericardium due to ruptured myocardial infarction or aortic dissection on unenhanced postmortem computed tomography

The aim of the study was to evaluate unenhanced postmortem computed tomography (PMCT) in cases of non-traumatic hemopericardium by establishing the sensitivity, specificity and accuracy of diagnostic criteria for the differentiation between aortic dissection and myocardial wall rupture due to infarction. Twenty six cases were identified as suitable for evaluation, of which ruptured aortic dissection could be identified as the underlying cause of hemopericardium in 50% of the cases, and myocardial wall rupture also in 50% of the cases. All cases underwent a PMCT and 24 of the cases also underwent one or more additional examinations: a subsequent autopsy, or a postmortem magnetic resonance (PMMR), or a PMCT angiography (PMCTA), or combinations of the above. Two radiologists evaluated the PMCT images and classified each case as “aortic dissection”, “myocardial wall rupture” or “undetermined”. Quantification of the pericardial blood was carried out using segmentation techniques. 17 of 26 cases were correctly identified, either as aortic dissections or myocardial ruptures, by both readers. 7 of 13 myocardial wall ruptures were identified by both readers, whereas both readers identified correctly 10 of 13 aortic dissection cases. Taking into account the responses of both readers, specificity was 100% for both causes of hemopericardium and sensitivity as well as accuracy was higher for aortic dissections than myocardial wall ruptures (72.7% and 87.5% vs 53.8% and 75% respectively). Pericardial blood volumes were constantly higher in the aortic dissection group, but a statistical significance of these differences could not be proven, since the small count of cases did not allow for statistical tests. This study showed that diagnostic criteria for the differentiation between ruptured aortic dissection and myocardial wall rupture due to infarction are highly specific and accurate.     

Detection of adrenal melanoma with computed tomography

Primary adrenal melanoma is a rare tumor. We present a case in which the diagnosis was established by computed-tomographic-guided fine-needle aspiration biopsy and subsequent autopsy.     

Detection of adrenal melanoma with computed tomography

Primary adrenal melanoma is a rare tumor. We present a case in which the diagnosis was established by computed-tomographic-guided fineneedle aspiration biopsy and subsequent autopsy.     

肾上腺腺瘤与转移瘤的CT鉴别诊断

目的探讨ＣＴ对肾上腺腺瘤与转移瘤的鉴别诊断价值。方法１０９例共１３０个肾上腺腺瘤和转移瘤，分别观察肿瘤的大小、边界、内部结构及其与同侧肾上腺的关系，测量肿瘤的平扫和增强后ＣＴ值及增强程度，２０例做了像素矩阵图分析，应用接受器工作特性（ＲＯＣ）曲线对结果进行分析。结果肿瘤平扫ＣＴ值的ＲＯＣ曲线下面积是０．９０±０．０５，明显大于增强后ＣＴ值（０．８１±０．０４）、增强程度（０．７１±０．０８）和肿瘤大小（０．８０±０．０４）的ＲＯＣ曲线下面积；以肿瘤直径≤３．０ｃｍ，密度均匀，平扫ＣＴ值≤２０ＨＵ，增强后ＣＴ值≤３５ＨＵ，增强程度≤２０ＨＵ为标准，诊断腺瘤的敏感性分别是８６％、８５％、８８％、７６％、和７４％，特异性分别是６８％、６９％、９０％、７４％和５９％；肿瘤与肾上腺相连（６８％）、平扫ＣＴ值＜０ＨＵ、ＣＴ像素矩阵图上有轻度负ＣＴ值区域只见于腺瘤，边界模糊或（和）侵犯周围结构、呈不规则厚环状强化是转移瘤的特征。全组总体符合率为９３％。结论综合分析肿瘤的各种ＣＴ征象，ＣＴ能鉴别绝大多数肾上腺腺瘤和转移瘤。     

CT象素矩阵图对鉴别肾上腺的腺瘤与恶性肿瘤的价值

目的：进一步探讨CT象素矩阵图对鉴别肾上腺腺瘤与恶性肿瘤的价值。方法：64例经临床、病理证实的肾上腺肿瘤共73个病灶,肿瘤的平扫平均CT值＞0HU。在CT平扫和增强图像上,分别应用CT象素矩阵图观察肿瘤内有无轻度负CT值区域。结果：在平扫和增强CT象素矩阵图上,分别有77％（24／31）和23％（6／26）的肾上腺腺瘤内见轻度负CT值区域,33例共42个恶性肿瘤病灶均未见此征象,特异性为100％。结论：分析CT象素矩阵图能发现病灶内最小CT值,对鉴别肾上腺的腺瘤与恶性肿瘤具有重要的价值。     

Unusual metastases from esophageal carcinoma: computed tomography demonstration

The computed tomography findings of large subcutaneous metastatic implants in the pelvic and inguinal regions from an epidermoid carcinoma of the esophagus are presented. To the best of the authors' knowledge, this appearance has not previously been reported.     

Comparison of unenhanced CT and chemical shift MRI in evaluating lipid-rich adrenal adenomas

OBJECTIVE: Our aim was to evaluate adrenal adenomas in patients who underwent both unenhanced CT and chemical shift MRI to determine if adenomas can be characterized with MRI when the findings of CT are indeterminate. MATERIALS AND METHODS: Between January 1999 and June 2003, 40 patients (42 adrenal masses) underwent unenhanced CT and chemical shift MRI and were retrospectively analyzed. Hounsfield units, adrenal-to-spleen chemical shift ratio, and signal-intensity index were obtained for each adrenal mass. Qualitative analysis for loss of signal in each adrenal mass on the opposed-phase images was also performed by two reviewers and compared with the quantitative analyses. A lipid-rich adenoma was diagnosed if the mass measured equal to or less than 10 H, had an adrenal-to-spleen chemical shift ratio of less than 0.71, and had a signal-intensity index of greater than 16.5% or if the mass fulfilled two of the preceding criteria and had follow-up imaging without change. RESULTS: The sensitivities and specificities for diagnosing a lipid-rich adenoma using the qualitative, adrenal-to-spleen chemical shift ratio, signal-intensity index, and unenhanced CT attenuation analyses were 92% (33/36) and 17% (1/6), 100% (36/36) and 100% (6/6), 100% (36/36) and 67% (4/6), and 78% (28/36) and 83% (5/6), respectively. Twenty-eight (67%) lipid-rich adenomas measured equal to or less than 10 H, had an adrenal-to-spleen chemical shift ratio of less than 0.71, and had a signal-intensity index of greater than 16.5%. Eight masses (19%) measured greater than 10 H but had an adrenal-to-spleen chemical shift ratio of less than 0.71 and a signal-intensity index greater than 16.5% and were unchanged at follow-up. CONCLUSION: Eight (62%) of 13 adrenal adenomas measuring greater than 10 H on unenhanced CT were definitively characterized with chemical shift MRI.     

Prediction of anemia on unenhanced computed tomography of the thorax.

OBJECTIVE: To determine if anemia can be predicted on unenhanced computed tomography (CT) of the thorax. METHODS: Hemoglobin and hematocrit levels were obtained within 24 hours of the unenhanced CT scan of the thorax of 200 patients. Anemia was defined as a hemoglobin level less than 140 g/L for men and less than 120 g/L for women. Regions of interest were placed on the left ventricular cavity, aorta and the interventricular septum if visualized. The attenuation of the interventricular septum and left ventricular cavity were correlated with the presence or absence of anemia. RESULTS: When the interventricular septum was not visualized, for every 1 Hounsfield unit (HU) increase in left ventricular attenuation, hemoglobin increased by 0.435 g/L (SE = 0.253, p < 0.001). Failure to visualize the interventricular septum did not exclude the presence of anemia in either sex. When the interventricular septum was visualized, 100% of males and 89% of females met the criteria for the diagnosis of anemia. The prediction of anemia by visualization of the interventricular septum alone yielded a sensitivity of 75.4% and a specificity of 90.3%, with 80% of patients correctly predicted. The multiple regression analysis model yielded a sensitivity of 94.2% and a specificity of 67.7%, with 86% of patients correctly predicted. CONCLUSION: The diagnosis of anemia should be suggested whenever the interventricular septum is visualized on unenhanced CT.     

Quantitative assessment of the superior sagittal sinus on unenhanced computed tomography

Purpose

To determine the relationship between hemoglobin levels and attenuation measurements of the superior sagittal sinus (SSS) on unenhanced computed tomography (UECT). Secondly, to determine if SSS attenuation values are normally distributed such that measurements below and above certain thresholds are suggestive of pathology, such as anemia or acute venous thrombosis respectively.

Methods

Institutional review board approval was obtained for retrospective review of adult patients having both an UECT head examination and a complete blood count within 24 h of the scan. A cohort of 151 consecutive patients formed the study sample (76 males and 75 females, 17-91 years of age with a mean of 61). The dorsal aspect of the SSS was divided into upper, middle and lower segments. Using freehand and circle region of interest (ROI) techniques, a total of six attenuation measurements were obtained from each patient. Next, statistical analyses were preformed to assess the correlation between Hgb levels and attenuation values, and distribution curves were plotted to assess the normal range of SSS attenuation measurements.

Results

There is a moderate, yet statistically significant (p < 0.001), correlation between Hgb levels and attenuation values in upper, middle and lower segments of the SSS (r = 0.487, 0.497 and 0.533 respectively). Based on the calculated mean, median and mode, the attenuation values are normally distributed. When using the freehand ROI technique, the mean value is 50 Hounsfield Units (HU) with a standard deviation (SD) of 7.5. Attenuation values less than 2 SDs (35 HU) are highly suggestive of anemia (specificity and PPV = 100%).

Conclusion

There is a moderate, yet statistically significant, correlation between Hgb levels and attenuation of the SSS on UECT. Furthermore, attenuation measurements of the SSS are normally distributed with a mean of 50 HU and a SD of 7.5. Therefore, quantitative assessment of the SSS may prove useful in the clinical practice of a radiologist; namely, in the diagnosis of anemia and acute venous thrombosis.     

Density histogram analysis of unenhanced hepatic computed tomography in patients with diffuse liver diseases

OBJECTIVE: The aim of the study was to assess the potential of density histogram analysis of unenhanced hepatic computed tomography (CT) in the diagnosis and differentiation of diffuse liver diseases. METHODS: Twenty-six patients with normal liver parenchyma, 35 patients with diffuse steatosis, 14 patients with acute steatohepatitis, 15 patients with active alcoholic cirrhosis, 23 patients with inactive alcoholic cirrhosis, 15 patients with virus-induced cirrhosis, and 8 patients with hemochromatosis underwent unenhanced hepatic CT. All diffuse liver diseases and the absence of diffuse liver disease were histologically proven. Quantitative analysis of unenhanced liver parenchyma was performed in each patient. RESULTS: The hepatic density histogram showed no significant differences in kurtosis and skewness between the groups (P > 0.05). Except for steatosis, active alcoholic cirrhosis, and hemochromatosis, diffuse liver diseases led to similar densities of liver parenchyma in unenhanced hepatic CT. CONCLUSION: A reliable diagnosis and differentiation of diffuse liver diseases on the basis of density histogram analysis is not possible.     

Computed tomographic histogram analysis in the diagnosis of lipid-poor adenomas: comparison to adrenal washout computed tomography

OBJECTIVE: To evaluate the ability of computed tomographic histogram analysis to diagnose lipid poor adenoma in comparison with adrenal washout computed tomography (CT). MATERIALS AND METHODS: Adrenal CT washout examinations performed during a period from January 2000 to July 2005 were reviewed. Computed tomographic histogram analysis was performed on the unenhanced component of the study, and sensitivity was assessed at thresholds of more than 5% and 10% negative pixels. Liver and spleen were used to represent the control/nonadenoma group. Computed tomographic noise was measured recording standard deviation (SD) of mean CT attenuation in adrenal, liver, and spleen. RESULTS: Twenty-four lipid-poor adenomas included exhibited more than 60% absolute enhancement washout (range, 60%-79%, mean, 69%) and remained stable for a period greater than 6 months. At threshold of more than 5% or 10% negative pixels CT histogram analysis yielded sensitivities of 91.6% and 70.8%, respectively, with 100% specificity. The mean SDs of adrenal, liver, and spleen were 18.2, 16.4 and 15, respectively. These differences in the mean SD were much smaller compared with the differences in the percentage of negative pixels in adrenal, liver, and spleen of 12.75%, 0.75%, and 0.25%, respectively. CONCLUSIONS: Computed tomographic histogram analysis has good potential in the diagnosis of lipid-poor adenoma and can reduce the need to perform adrenal washout CT.     

Ureterolithiasis: classical and atypical findings on unenhanced helical computed tomography

Evaluation of patients with acute flank pain using helical computed tomography (CT) is a well-accepted, rapid, and safe procedure in the emergency setting. Various primary and secondary signs are described in the literature for evaluation of these patients. Our purpose is to demonstrate both the classical findings associated with ureteral calculi on unenhanced helical CT and atypical findings and potential pitfalls. We also provide readers with a systematic approach to interpreting unenhanced helical CT scans performed for acute flank pain. Electronic Publication     

Distinction between adrenal adenomas and metastases using MR imaging

Magnetic resonance imaging was performed on 12 patients with known neoplastic disease and adrenal masses shown by CT. Three patients with metastases had high signal intensity in the adrenals on T2 weighted spin echo scans (SE 2,500/80) and nine patients with nonfunctioning adenomas had low signal intensity on T2 weighted images. The ability to distinguish metastases from nonhyperfunctioning adrenal adenomas may be of use in the pre-operative evaluation in patients with known carcinoma and incidental adrenal masses.