CT-derived textural analysis parameters discriminate high-attenuation renal cysts from solid renal neoplasms.

AIM: To assess the utility of textural features on computed tomography (CT) to differentiate high-attenuation cysts from solid renal neoplasms among indeterminate renal lesions detected incidentally on CT. MATERIALS AND METHODS: Patients were included if they had an indeterminate renal lesion on CT that was subsequently characterised on ultrasound or magnetic resonance imaging (MRI). Up to three lesions per patient were included if they had a size ≥10 mm and density of 20-70 HU on unenhanced CT or any single phase of contrast-enhanced CT. Cases were categorised as benign or most likely benign cysts (Bosniak II and IIF) versus indeterminate (Bosniak III), mixed solid and cystic (Bosniak IV), or solid renal lesions. A random forest model was generated using 95 textural parameters and four clinical parameters for each lesion. RESULTS: Two hundred and thirty-four patients were included who had a total of 278 lesions. Of these, 193 (69%) were benign or most likely benign cysts and 85 (31%) were indeterminate, mixed cystic and solid, or solid renal lesions. The random forest model had an area under the curve of 0.71 (95% confidence interval [CI]: 0.65, 0.78), with a sensitivity and specificity of 81.2% and 38.9%, respectively. CONCLUSION: A multivariate model including textural and clinical parameters had moderate overall performance for discriminating benign or likely benign cysts from indeterminate, mixed solid and cystic, or solid renal lesions. This study serves as a proof of concept and may reduce the need for further follow-up by characterising a significant portion of indeterminate lesions on CT as benign.

Characterization of adnexal lesions using photoacoustic imaging to improve sonographic O-RADS risk assessment.

OBJECTIVE: To assess the impact of photoacoustic imaging (PAI) on the assessment of ovarian/adnexal lesion(s) of different risk categories using the sonographic ovarian-adnexal imaging-reporting-data system (O-RADS) in women undergoing planned oophorectomy. METHOD: This prospective study enrolled women with ovarian/adnexal lesion(s) suggestive of malignancy referred for oophorectomy. Participants underwent clinical ultrasound (US) examination followed by coregistered US and PAI prior to oophorectomy. Each ovarian/adnexal lesion was graded by two radiologists using the US O-RADS scale. PAI was used to compute relative total hemoglobin concentration (rHbT) and blood oxygenation saturation (%sO2 ) colormaps in the region of interest. Lesions were categorized by histopathology into malignant ovarian/adnexal lesion, malignant Fallopian tube only and several benign categories, in order to assess the impact of incorporating PAI in the assessment of risk of malignancy with O-RADS. Malignant and benign histologic groups were compared with respect to rHbT and %sO2 and logistic regression models were developed based on tumor marker CA125 alone, US-based O-RADS alone, PAI-based rHbT with %sO2 , and the combination of CA125, O-RADS, rHbT and %sO2. Areas under the receiver-operating-characteristics curve (AUC) were used to compare the diagnostic performance of the models. RESULTS: There were 93 lesions identified on imaging among 68 women (mean age, 52 (range, 21-79) years). Surgical pathology revealed 14 patients with malignant ovarian/adnexal lesion, two with malignant Fallopian tube only and 52 with benign findings. rHbT was significantly higher in malignant compared with benign lesions. %sO2 was lower in malignant lesions, but the difference was not statistically significant for all benign categories. Feature analysis revealed that rHbT, CA125, O-RADS and %sO2 were the most important predictors of malignancy. Logistic regression models revealed an AUC of 0.789 (95% CI, 0.626-0.953) for CA125 alone, AUC of 0.857 (95% CI, 0.733-0.981) for O-RADS only, AUC of 0.883 (95% CI, 0.760-1) for CA125 and O-RADS and an AUC of 0.900 (95% CI, 0.815-0.985) for rHbT and %sO2 in the prediction of malignancy. A model utilizing all four predictors (CA125, O-RADS, rHbT and %sO2 ) achieved superior performance, with an AUC of 0.970 (95% CI, 0.932-1), sensitivity of 100% and specificity of 82%. CONCLUSIONS: Incorporating the additional information provided by PAI-derived rHbT and %sO2 improves significantly the performance of US-based O-RADS in the diagnosis of adnexal lesions. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

Segmental disorders of the nephron: histopathological and imaging perspective.

Recent advances in molecular genetics and immunocytochemistry have clarified the cell of origin in many renal disorders. Several renal disorders are thought to involve specific segments of the nephron. Renin-secreting tumours arise from juxtaglomerular cells. Clear cell and papillary renal cell carcinoma (RCC) recapitulate the epithelium of the proximal tubules. Oncocytoma and chromophobe RCC differentiate towards Type A and Type B intercalated cells of the cortical collecting duct, respectively. Medullary collecting ducts are the target sites for the development of autosomal recessive polycystic kidney disease, collecting duct carcinoma and medullary carcinoma. Renal papillae are susceptible to unique changes such as necrosis or papillitis. The purpose of our article is threefold: to illustrate the imaging findings of renal disorders that show segmental involvement of the nephron, to describe proximal and distal nephron disorders and to correlate imaging findings of some entities with histopathological features.

Collecting duct carcinoma of the kidney: are imaging findings suggestive of the diagnosis?

OBJECTIVE: Collecting duct carcinoma derives from the renal medulla and has an infiltrative growth pattern at pathologic examination. The purpose of our study was to characterize the imaging features of this aggressive malignancy and determine whether the diagnosis can be reliably suggested from imaging findings. MATERIALS AND METHODS: Radiologic studies from 17 patients with pathologically proven collecting duct carcinoma were analyzed by two reviewers. RESULTS: The tumors varied in size from 1.5 to 19 cm (mean, 7.7 cm). Medullary involvement was present on CT in 16 (94%) of 17 cases, but cortical involvement or an exophytic component was also present in 15 cases (88%) and 10 cases (59%), respectively. The reniform contour of the kidney was preserved in seven cases (41%) and correlated with a smaller tumor size (p<0.01). Tumors showed an infiltrative appearance on CT in 11 cases (65%), but an expansile component was also present in eight of these cases. A cystic component was present on CT in six (35%) of 17 cases. On sonography, the solid tumor component was hyperechoic to normal renal parenchyma in six of seven cases and isoechoic in the other. On MR imaging, all tumors (4/4) were hypointense on T2-weighted imaging. On urography, all lesions (5/5) distorted the intrarenal collecting system. On angiography, all tumors (3/3) were hypovascular. CONCLUSION: Medullary involvement and an infiltrative appearance are common findings on cross-sectional imaging and may suggest the diagnosis of collecting duct carcinoma. In large tumors, however, these features are frequently overshadowed by an exophytic or expansile component that cannot be distinguished from the more common cortical renal cell carcinoma.

Renal cysts: is attenuation artifactually increased on contrast-enhanced CT images?

PURPOSE: To determine if the attenuation values of simple renal cysts are artifactually increased on contrast material-enhanced, clinically acquired spiral computed tomographic (CT) images. MATERIALS AND METHODS: Dual-phase renal spiral CT studies (5-mm collimation; pitch, 1.0) were retrospectively analyzed in 24 consecutive patients who had ultrasonographic (US) documentation of simple renal cysts. Forty-eight cysts were identified. The attenuation values of each cyst were measured on nonenhanced, cortical phase, and nephrographic phase images. The size and the location of each cyst in relation to the renal parenchyma were also recorded. RESULTS: The cysts were 0.6-10.8 cm in diameter (mean, 2.6 cm; SD, 2.0). The mean attenuation change in the cysts from nonenhanced to contrast-enhanced images was statistically significant in a comparison of cortical phase and nephrographic phase images (P: <.01): +1.8 HU (SD, 3.8) for cortical phase and +3. 6 HU (SD, 5.6) for nephrographic phase images. Renal cysts 1.0 cm or smaller showed a higher attenuation increase (mean, +4.0 HU for cortical phase and +11.0 HU for nephrographic phase). None of the renal cysts larger than 1.0 cm demonstrated an increase greater than 10 HU (mean, +1.4 HU for cortical phase and +2.3 HU for nephrographic phase). Intraparenchymal cysts showed higher mean attenuation changes than the exophytic cysts. CONCLUSION: Attenuation values in the renal cysts increased artifactually on contrast-enhanced images, but this pseudoenhancement was not substantial and was less than 10 HU when the cyst was larger than 1. 0 cm in diameter.

Interobserver variability in determining enhancement of renal masses on helical CT.

OBJECTIVE: Because enhancing renal masses are often surgically excised, we assessed interobserver variability in determining enhancement of renal masses. SUBJECTS AND METHODS: Forty-nine complicated cysts or solid lesions were imaged with helical CT. Simple cysts (n = 20) were included, for a total of 69 lesions. Three radiologists obtained three region-of-interest measurements for each lesion in the unenhanced and nephrographic phases. Enhancement was defined as an attenuation increase of 15 H or more, provided that partial volume effects were not the cause. Observer concordance was defined as agreement among all three observers on whether a lesion was enhancing or nonenhancing. Measurement-interpretation agreement was defined as an observer's measuring an attenuation increase of less than 15 H and interpreting the lesion as nonenhancing or measuring an increase of 15 H or greater and interpreting the lesion as enhancing. RESULTS: For the 20 simple cysts, observer concordance was 100% and measurement-interpretation agreement was 100%. For the 49 complicated lesions, observer concordance was 90% and measurement-interpretation agreement was 84% (41/49). Of the five discordant lesions, three were 1.1 or 1.0 cm in size and were interpreted as being of high attenuation on the unenhanced scans, with enhancement, versus a high-attenuation mass with increasing attenuation attributed to volume averaging. The remaining two discordant observations were due to differing interpretations of normally enhancing renal parenchyma versus cyst wall or tumor nodularity. CONCLUSION: For most renal lesions, the observers agreed on whether enhancement was present or absent. Observers were prone to disagree about enhancement for small lesions (1.0-1.5 cm). Occasional disagreements on classification of larger lesions were due to differing interpretations of enhancing normal parenchyma versus abnormal tissue. Discrepancies between an observer's opinion on enhancement and the results of region-of-interest measurements suggest the possibility of disagreement among multiple observers.

Sonography of the female urethra.

OBJECTIVE: The purpose of this study was to explore the role of sonography for women with urethral symptoms and a suspected urethral diverticulum. SUBJECTS AND METHODS: Nineteen women with urethral symptoms underwent voiding cystourethrography (VCUG) and transvaginal, transperineal, and urethral sonography (using a catheter-based transducer). VCUGs and sonograms were evaluated for diverticula, defined on sonography by direct visualization of the neck connecting the periurethral sac with the urethral lumen. The diverticular neck, size, location, and shape were noted. Lesions revealed by sonography as not connected to the urethra were also noted. RESULTS: Of 19 women, 14 had urethral diverticula and one had two diverticula, for a total of 15 diverticula. On sonography the diverticula ranged in diameter from 2 mm to 5 cm. Both sonography and VCUG showed 13 of the 15 diverticula. In addition, sonography revealed two infected periurethral cysts, a periurethral leiomyoma, a diffuse urethritis, and scarring or deformity of one patient's urethra from a prior diverticulectomy. On sonography, eight of the 13 diverticula wrapped around more than 50% of the urethral circumference. The neck was precisely seen (by definition) in 13 of 15 diverticula on sonography and in two of 13 diverticula on VCUG. CONCLUSION: Sonography is useful in this group of women with urethral symptoms and suspected urethral diverticula. It provides information on the extent and location of the diverticular neck, both of which are important in surgical excision. Also, sonography provides information on lesions not connected to the urethra. Sonography may prove useful in a broader group of women with urethral symptomatology.

Renal medullary carcinoma.

PURPOSE: Renal medullary carcinoma is a rare and extremely aggressive neoplasm that almost always develops in young patients with sickle cell trait. To our knowledge all cases to date have been metastatic at surgical resection. Pathological examination reveals an aggressive tumor mainly involving the renal medulla with a varied morphology. The prognosis is dismal. Mean survival from the time of resection is 15 weeks (range 2 to 52). The disease course has not been altered by surgery, radiotherapy or various regimens of chemotherapeutic agents. MATERIALS AND METHODS: We add to the literature our experience treating renal medullary carcinoma in 2 cases and review the existing literature on this disease. RESULTS: Both patients whom we treated died of the disease, as have the other 35 patients described in the literature. CONCLUSIONS: A high index of suspicion may lead to earlier diagnosis and treatment, and survival of patients with renal medullary carcinoma.

CT of cystic renal masses: analysis of diagnostic performance and interobserver variation.

OBJECTIVE: Our objective was to assess the clinical usefulness and interobserver variability of the Bosniak classification scheme for characterizing a series of pathologically proven cystic renal lesions imaged with CT. MATERIALS AND METHODS: Seventy pathologically proven cystic renal masses (38 benign, 32 malignant) in 46 patients were reviewed independently by three radiologists. The cystic masses were categorized by each reviewer according to both the Bosniak classification and the receiver operating characteristic (ROC) analysis. Both the individual results for each reader and the pooled results for all three readers were analyzed. Interobserver agreement and discordance in classifying lesions as Bosniak categories I-II or III-IV were assessed. RESULTS: The distribution of the 70 lesions (based on the average of the three readers) was 22 Bosniak I (0% malignant), eight Bosniak II (13% malignant), 11 Bosniak III (45% malignant), and 29 Bosniak IV (90% malignant). All readers agreed on the Bosniak classification in 59%, or 41 of the 70 lesions (I, 17; II, one: III, four: and IV, 19). Eleven (16%) of the 70 lesions were classified as Bosniak I or II by one reader and as Bosniak III or IV by at least one other reader. The area under the curve for the pooled ROC analysis was calculated to be 0.957. Individual reader values ranged from 0.914 to 0.981. The sensitivities, specificities, and accuracies for the three readers ranged from 94% to 100%, 71% to 92%, and 84% to 93%. Assessment of interobserver variability by kappa analysis yielded scores of .571 and .477 for the Bosniak and ROC analyses, respectively. CONCLUSION: Overall, the Bosniak classification scheme is useful for evaluating renal masses: however, interobserver variation in distinguishing. Bosniak II and Bosniak III lesions may present difficulties in recommending surgical versus conservative management.

Preoperative assessment of ureteropelvic junction obstruction with endoluminal sonography and helical CT.

OBJECTIVE: Our objective was to compare the value of endoluminal sonography with the value of helical CT in the preoperative assessment of crossing vessels in patients with ureteropelvic junction (UPJ) obstruction. SUBJECTS AND METHODS: Thirteen patients with UPJ obstruction underwent contrast-enhanced helical CT with multiplanar reformations and endoluminal sonography. Imaging preceded surgery for UPJ repair. On imaging, vessels were considered significant if greater than or equal to 2 mm in diameter and within 1 cm of the UPJ. RESULTS: Three patients had no crossing vessels revealed by either study. On sonography, another patient had a vessel revealed with a diameter that varied between 1.3 and 2.2 mm; on CT no correlate was detected. The remaining nine patients had vessels revealed by both techniques. On CT, four patients had two vessels revealed and five patients had single vessels revealed for a total of 13 vessels revealed by CT. On sonography, five patients had two vessels revealed and five patients had single vessels revealed. Thus, 15 vessels were revealed by sonography. Both arteries and veins were revealed anterior, posterior, and medial to the UPJ; no lateral vessels were seen. Four patients underwent laparoscopy, during which the absence, presence, and location of vessels were found to correlate with sonography and helical CT. CONCLUSION: Endoluminal sonography and helical CT were similar in revealing crossing vessels in patients with UPJ obstruction.

Primitive neuroectodermal tumor of the kidney--another enigma: a pathologic, immunohistochemical, and molecular diagnostic study.

Primitive neuroectodermal tumor (PNET), the second most common type of sarcoma in the first two decades of life, rarely presents as an organ-based neoplasm. Rather, it is seen typically in the soft tissues of the chest wall and paraspinal region. We report a case of primary PNET of the kidney in a 17-year-old girl who presented with abdominal pain, hematuria, and an abdominal mass. Nodules and sheets of monotonous-appearing primitive round cells and the formation of rosettes focally were the principal microscopic features. The tumor cells were uniformly immunoreactive for vimentin, cytokeratin, neuron-specific enolase, and 013 (CD99). In addition, the characteristic translocation of PNET and Ewing sarcoma, t(11;22)(q24;q12), was detected by polymerase chain reaction (PCR). Eight previous examples of renal PNET have been reported in the literature in the past 2 years, but only three of these cases have had complete immunohistochemical evaluation with the demonstration of 013 positivity. To our knowledge the present case is the only one to date demonstrating the recurrent translocation t(11;22)(q24;q12) by PCR. Assuming that the previous cases in the literature are bona fide examples of PNET, the kidney may be another site of predilection for this usual soft-tissue neoplasm. We are once again confronted with the dilemma about the nature of the progenitor cell.

Helical (spiral) CT angiography for identification of crossing vessels at the ureteropelvic junction.

OBJECTIVE: The purpose of this study was to determine the feasibility of imaging crossing vessels at the ureteropelvic junction (UPJ) with helical (spiral) CT angiography for planning surgical repair of symptomatic UPJ obstruction. SUBJECTS AND METHODS: Twenty-four consecutive patients with symptomatic UPJ obstruction were imaged with dual-phase, contrast-enhanced helical CT (collimation, 3 mm; pitch, 1.3-1.7; reconstruction interval, 2 mm; early phase, 20-42 sec; and delayed phase, 90-112 sec after initiation of IV contract material injection [125 ml of ioversol containing 320 mg of iodine per ml, delivered at 4-5 ml/sec]). All imaging data were viewed interactively on an imaging workstation. Prospective on-line interpretations were correlated with subsequent surgical and clinical findings at laparoscopy (n=3), open surgical repair (n=2), or ureteronephroscopic endopyelotomy (n=11). Vessels at the UPJ that were 2 mm or more in diameter were believed to be significant. Review of the transaxial images was performed to determine qualitatively the relative usefulness of the early versus the delayed phases for distinguishing arteries from veins. Multiplanar reformations also were retrospectively reviewed and compared with direct pyelograms to determine the accuracy with which the location of the UPJ and the proximal ureteral course were depicted with helical CT. RESULTS: Eleven of 24 (46%) patients collectively had 11 anterior and three posterior vessels (> or = 2 mm in diameter) crossing the UPJ on helical CT. Distinction between arteries and veins was significantly better on early-phase than on delayed-phase images (p=.01). Visualization of the UPJ and the proximal ureteral course was good or excellent for 18 (78%) of 23 patients for whom pyelograms were available, regardless of the presence of a ureteral stent (p>.05). Laparoscopy and open surgery findings were in agreement with helical CT angiograms for five of five patients. Uncomplicated endopyelotomy was performed for 11 patients in whom no significant vessels were seen posterior or posterolateral to the UPJ. CONCLUSION: Helical CT angiography can depict vessels crossing the UPJ and is valuable in planning surgical management.

Imaging for ureteropelvic junction obstruction in adults.

The three goals of imaging for ureteropelvic junction (UPJ) obstruction are to determine the presence and degree of renal obstruction, to assess renal function, and to determine the cause of the problem. The diagnostic techniques are intravenous urography, which is unreliable; ultrasonography, with the addition of pulsed Doppler technology; diuretic scintigraphy; and pressure-flow studies (Whitaker test), which is the only direct clinical test of renal outflow resistance. Voiding cystography, retrograde ureteropyelography, and CT have utility in some patients. In determining the cause of obstruction, the patient often is evaluated for crossing vessels using angiography, endoluminal ultrasonography, or spiral CT. It is not possible to formulate a single algorithm for radiographic evaluation of suspected UPJ obstruction, but in the average adult patient, urography and diuretic scintigraphy are sufficient.

Angiomyolipoma and renal cell carcinoma: US differentiation.

PURPOSE: To analyze whether shadowing and other ultrasound (US) features were helpful for distinguishing angiomyolipoma (AML) from renal cell carcinoma (RCC). MATERIALS AND METHODS: US images were reviewed of 49 patients with RCC and 35 patients with AML. Each tumor was evaluated for size, location, echogenicity, homogeneity, shadowing, hypoechoic rim, and intratumoral cysts. When available, computed tomographic (CT) scans of AMLs were analyzed for the amount of fat and soft tissue in each lesion. RESULTS: AMLs tended to be smaller and more frequently echogenic than RCCs, but statistically significant overlap occurred. Shadowing was seen in 12 (33%) AMLs but was not seen in RCCs. Hypoechoic rims and intratumoral cysts were seen only in RCCs (numbers were too small to perform further statistical analysis). CONCLUSION: In hyperechoic renal masses, the presence of shadowing, a hypoechoic rim, and intratumoral cysts are important findings that may help distinguish AML from RCC.

Abdominal aortic aneurysm morphology: CT features in patients with ruptured and nonruptured aneurysms.

OBJECTIVE: We evaluated a variety of internal architectural features in ruptured and nonruptured abdominal aortic aneurysms to determine whether any features are associated more frequently with ruptured abdominal aortic aneurysms. These features may be useful in identifying subtle ruptures when no obvious retroperitoneal hematoma is present and may be helpful in predicting unstable aneurysms at risk for rupture. MATERIALS AND METHODS: The CT scans of 52 patients with ruptured abdominal aortic aneurysms were reviewed and compared with those of 56 patients with asymptomatic nonruptured aneurysms exceeding 4.5 cm in diameter. All aneurysms were evaluated for size, rim calcification, thrombus amount, thrombus calcification, and lumen irregularity. In addition, four different thrombus patterns were identified and evaluated, including homogeneous, diffusely heterogeneous, periluminal halo, and crescent patterns. Statistical comparisons were adjusted for differences in size between the two groups. RESULTS: Age, gender, and aneurysm length were not statistically different between the two groups. A larger diameter was found in the ruptured aneurysm group: 7.4 (anteroposterior) x 7.9 (transverse) cm versus 5.9 x 6.1 cm (p = .00001). More thrombus surrounded the nonruptured aneurysms (p = .014). Thrombus calcification was seen in 25% (14/56) of the control group and in 13% (7/52) of the rupture group (p = .01). Two thrombus patterns, homogeneous and periluminal halo, were encountered with similar frequencies in both groups. The diffusely heterogeneous pattern was seen more in the control group. A crescent of increased attenuation was encountered only in patients with ruptured aneurysms, at an incidence of 21% (11/52) (p = .0005). Thick and thin wall calcifications were seen in both groups, but a focal discontinuity in circumferential calcification was seen only in association with ruptured aneurysms, at an incidence of 8% (4/52) (p = .008). There was no significant difference in the number of patients whose patent lumen was irregular. CONCLUSION: In our series, detection of a high-attenuation crescent or focal gap of otherwise circumferential wall calcification is associated with aneurysm rupture. The homogeneous, diffusely heterogeneous, and periluminal halo patterns are not specifically associated with aortic rupture. There were no significant differences in the amount of wall calcification or frequency of lumenal irregularity between patients with ruptured and those with nonruptured aneurysms.

CT-detected renal arterial calcification: correlation with renal artery stenosis on angiography.

OBJECTIVE: The purpose of this study was to determine whether a relationship exists between CT-detected calcifications in the main renal artery and renal artery stenosis. We also evaluated whether renal artery calcifications are more indicative of stenosis in particular patient subgroups based on age, sex, or hypertensive status. MATERIALS AND METHODS: We retrospectively reviewed the examinations of 70 patients (122 kidneys) undergoing both abdominal CT and angiography for various clinical conditions, most commonly evaluation of a tumor or aneurysm. CT studies were evaluated for degree of calcification at the orifice and in the proximal and distal segments of the main renal artery. The angiograms were evaluated for degree of narrowing. Renal artery stenosis was defined as a reduction in cross-sectional area of 75% or more. A separate analysis was performed on male and female patients above and below 65 years of age, with and without hypertension. RESULTS: Renal arteries with clumps of calcium (> 3 mm in diameter) had a higher percentage of stenosis (7/16, 44%) than did renal arteries with less calcium (16/53, 30%), and renal arteries with no calcium had the fewest stenoses (9/53, 17%; p = .02). Calcifications in the renal artery were not good predictors of stenosis as they were frequently seen in arteries with nonsignificant narrowing and were occasionally seen in arteries with no narrowing. The patient's sex was not a significant factor in this analysis (p = .34). Renal artery calcifications in hypertensive patients showed a stronger association with renal artery stenosis, but the overall ability to predict stenosis remained poor. However, calcification may be significant in persons less than 65 years old, in whom this finding was associated with stenosis all three cases (100%). Similar calcifications in patients 65 years old or more were associated with stenosis in only 4 (31%) of 13 cases (p < .01). CONCLUSION: CT-detected calcifications in renal arteries are associated with stenosis, but their predictive value in the general population or in hypertensive patients is poor. These calcifications in a patient less than 65 years of age are a better indicator of stenosis of the renal artery.

Radiologic features of well-differentiated osteosarcoma.

Eight cases of well-differentiated, intramedullary osteosarcoma were studied. In three cases, the lesions previously had been diagnosed, pathologically and radiologically, as benign. The distal femur was the most frequent site. Radiographically, the tumor margins were poorly defined in five cases and were well defined in three. In seven cases, the bone response pattern was both lytic and blastic; in one case, it was purely lytic. Tumor matrix patterns were mostly homogeneous and cloudlike. Four of the lesions were expansile. All cases had cortical thinning and discontinuity. Extraosseous involvement of the soft tissues was present in six cases, and active periosteal response was present in two. High-grade malignancies subsequently developed in two patients--one was a local recurrence and the other was a distant metastasis. A broad spectrum of radiologic features was observed in well-differentiated intramedullary osteosarcomas, but all cases had at least two of three radiographic findings (cortical discontinuity, poorly marginated soft-tissue extension, or cloudlike tumor matrix pattern) that strongly suggested malignancy.