Giant cell tumours of the mobile spine: characteristic imaging features and differential diagnosis.

PURPOSE: The aim of this study was to investigate the characteristic imaging features of giant cell tumours (GCTs) of the mobile spine. MATERIALS AND METHODS: Thirty pathologically proven GCTs of the mobile spine were reviewed. X-ray (n = 18), computed tomography (CT) (n = 24) and magnetic resonance (MR) (n = 21) images were retrospectively evaluated. RESULTS: Five tumours were located in the cervical spine, 15 tumours were located in the thoracic spine and 10 tumours in the lumbar spine. The characteristic X-ray findings included an osteolytic and expansile lesion with a "soap bubble" or purely lytic appearance. Cortical destruction was commonly seen. Margin sclerosis was seen in two lesions. No mineralised tumour matrix or periosteal reaction appeared. The CT findings were similar but outlined the cortical alterations in a more accurate way. The characteristic MR findings included a well-defined and expansile mass with heterogeneous low-to-iso signal intensity on T2-weighted images. Cystic areas were commonly seen in 17 cases. Five cases presented fluid-fluid levels, suggesting the development of aneurysmal bone cyst. The solid portions of the tumours were enhanced with a very heterogeneous signal pattern reflecting high blood supply after contrast-enhanced scan. Tumour involvement in the epidural space occurred in 12 cases, causing spinal cord and/or nerve root compression. Involvement of intervertebral discs and/or adjacent vertebrae appeared in two cases. CONCLUSIONS: Although rare, GCT can occur in the mobile spine as a kind of benign but locally aggressive tumour. Radiologists should be familiar with its characteristic imaging features in order to make a correct diagnosis and to help preoperative evaluations.

In-depth analysis of local recurrence of giant cell tumour of bone with soft tissue extension after intralesional curettage.

PURPOSE: The aim of this study was to assess the local recurrence rate of giant cell tumour of bone (GCTB) with soft tissue extension, to identify characteristics of the soft tissue extension that can best indicate recurrence of GCTB after intralesional curettage. MATERIALS AND METHODS: A total of 48 cases of GCTB with soft tissue extension after intralesional curettage were recruited. Patients were divided into two groups based on various objective features of soft tissue extension including size, number, margins, involvement of adjacent tissues, signal intensity, static enhancement and Jaffe grade. The local recurrence rate was compared using the Chi-square test and Chi-square value correction for continuity. Risk factors were assessed by multivariate logistic regression analysis. RESULTS: The local recurrence rate was significantly different according to soft tissue extension size, number and margins (p < 0.05). There was no significant difference in the groups of adjacent tissue involvement and Jaffe grade (p > 0.05). Size, number and margins of the soft tissue extension were independent risk factors of local recurrence of GCTB after intralesional curettage (p < 0.05). CONCLUSIONS: The local recurrence rate of GCTB with soft tissue extension after intralesional curettage is higher if the soft tissue extension is large, multiple and lacking bone envelope integrity. For cases with the above-mentioned features, we suggest that the higher recurrence rate can be taken into full consideration when choosing appropriate surgical procedures.

Recurrence in giant cell tumour of bone: imaging features and risk factors.

PURPOSE: This study was done to investigate X-ray, computed tomography (CT) and magnetic resonance (MR) imaging features of recurrence in giant cell tumour of bone (GCTB) and to evaluate risk factors. MATERIALS AND METHODS: Medical records and imaging data were reviewed for 55 cases of recurrent GCTB. All images were reviewed retrospectively and independently by two radiologists experienced in skeletal musculature. The common radiological findings; factors related to tumour recurrence such as gender, age, location; pathological fracture, Campanacci grading and surgical procedure were analysed by nonparametric test (Mann-Whitney U test for two independent samples test and Kruskal-Wallis H test for multiple independent samples test). p values <0.05 were considered to indicate a statistically significant difference. RESULTS: The imaging features of recurrent GCTB were as follows: osteolytic destruction or bone resorption of graft bone or around the polymethylmethacrylate (PMMA), soft tissue mass formation and expansile change. Tumour parenchyma showed markedly heterogeneous enhancement, except for necrotic cystic cavities, on contrast-enhanced MR images. Wide resection had a smaller (p=0.031) risk of local recurrence than did intralesional curettage. There was no statistical significance in gender, age, location, pathological fracture and Campanacci staging (p>0.05). CONCLUSIONS: The risk of recurrence in GCTB was influenced by the type of surgery and adjuvants. Bone transformaresorption, soft tissue mass formation and aggravated expansile change are reliable signs of recurrence on imaging.

Feasibility of diagnosing and staging liver fibrosis with diffusion weighted imaging.

OBJECTIVE: To assess the clinical feasibility of diagnosing and staging liver fibrosis by apparent diffusion coefficient (ADC). METHODS: Totally, 43 patients (mean age 29.3 years) with chronic hepatitis by liver biopsy and 7 healthy controls (mean age 39.9 years) underwent liver diffusion weighted imaging (DWI) with four b values: 0, 200, 500, and 1000 s/mm2 respectively. The liver fibrosis was staged according to Ishak fibrosis stage. The ADC value of liver fibrosis patients and healthy controls was compared. The correlation of ADC value and liver fibrosis staging was analyzed. RESULT: The histological staging showed 8 stage 1 patients, 10 stage 2 patients, 6 stage 3 patients, 9 stage 4 patients, 8 stage 5 patients and 2 stage 6 patients. The mean ADC value of liver fibrosis patients was significantly lower than that of healthy controls except for stage 1 group (P < 0.05). There was a negative correlation between liver fibrosis staging and ADC value (r = -0.697 with b=500 s/mm2, P < 0.01). Receiver operating characteristic (ROC) curve of ADC value of advanced liver fibrosis (Ishak stage F3 and higher) showed that area under curve = 0.913, 0.825, and 0.794 with b = 500, 1000, and 200 s/mm2, respectively (95% confidence interval: 83.6%-99.0%, 70.7%-94.3%, 66.5%-92.4%; P < 0.05). When b value was 500 s/mm2, the sensitivity (84%) and specificity (80%) of DWI for diagnosis of advanced liver fibrosis were the highest. CONCLUSION: DWI is proved to be a useful clinical tool in the quantitative evaluation of liver fibrosis and in the prediction of the process of liver fibrosis with the recommendable b value (500 s/mm2).

Triple-phase dynamic MRI: a new clue to predict malignant transformation of giant cell tumor of bone.

OBJECTIVE: Our purpose was, through the comparison of the characteristics of time-intensity curve on triple-phase dynamic contrast-enhanced MRI among groups of giant cell tumor of bone (GCTB), recurrent benign giant cell tumor of bone (RBGCTB), and secondary malignant giant cell tumor of bone (SMGCTB), to find clues to predict the malignant transformation of GCTB. SUBJECTS AND METHODS: 21 patients diagnosed as GCTB were included in this study. All cases took recurrence after intralesional curettage. 9 cases were confirmed as SMGCTB and 12 cases were confirmed as RBGCTB. Cases were divided into four groups: group A, GCTB (n=9); group B, SMGCTB (n=9); group C, GCTB (n=12); group D, RBGCTB (n=12). Enhancement index(EI) of lesions on DCEMRI was calculated using formula: EI(t)=[S(t)-S(0)]/S(0), where S(0) was signal intensity of lesion on non-contrast-enhanced T1-weighted images and S(t) was signal intensity of lesion on DCEMRI (t=30, 60, 180s). Enhancement index of each group in each phase was compared using One-Way ANOVA analysis. Slope values of time-intensity curve were compared by the same way. RESULTS: Time-intensity curve of SMGCTB was characterized by a steep upward slope followed by an early and rapid washout phase. Time-intensity curve of GCTB and RBGCTB was characterized by a steep slope followed by a relatively slow washout phase. No significant difference in enhancement index was found in the first phase (p>0.05). There was significant difference in the second and the third phase (p<0.05). Enhancement index of group B (SMGCTB) was smaller. There was no difference in rising slope value (p>0.05). CONCLUSIONS: Dynamic contrast-enhanced MRI appears a helpful method to find new clues to predict malignant transformation of GCTB.

Differentiation of primary chordoma, giant cell tumor and schwannoma of the sacrum by CT and MRI.

OBJECTIVE: To evaluate criteria to differentiate sacral chordoma (SC), sacral giant cell tumor (SGCT) and giant sacral schwannoma (GSS) with CT and MRI. MATERIALS AND METHODS: CT and MR images of 22 SCs, 19 SGCTs and 8 GSSs were reviewed. The clinical and imaging features of each tumor were analyzed. RESULTS: The mean ages of SC, SGCT and GSS were 55.1 ± 10.7, 34.3 ± 10.7 and 42.4 ± 15.7 years old. SCs (77.3%) were predominantly located in the midline of lower sacrum, while most SGCTs (73.7%) and GSSs (87.5%) were eccentrically located in upper sacrum. There were significant differences in age, location, eccentricity, morphology of bone residues, intratumoral bleeding and septations. Multiple small cysts were mainly observed in SGCTs (73.7%) with large central cysts in GSSs (87.5%). SGCTs expanded mainly inside sacrum while SCs and GSSs often extended into pelvic cavity (P = 0.0022). Involvement of sacroiliac joints and muscles were also different. Ascending extension within sacral canal was only displayed in SCs. The preservation of intervertebral discs showed difference between large and small tumors (P = 0.0002), regardless of tumor type (P = 0.095). No significant difference was displayed in gender (P = 0.234) or tumor size (P = 0.0832) among three groups. CONCLUSION: Age, epicenter of the lesion (midline vs. eccentric and upper vs. lower sacral vertebra), bone residues, cysts, bleeding, septation, expanding pattern, muscles and sacroiliac joint involvement can be criteria for diagnosis. Fluid-fluid level is specific for SGCTs and ascending extension within the sacral canal for SCs. The preservation of intervertebral discs is related to tumor size rather than tumor type.

Noninvasive assessment of response to neoadjuvant chemotherapy in osteosarcoma of long bones with diffusion-weighted imaging: an initial in vivo study.

OBJECTIVES: The purpose of our study is to investigate whether diffusion-weighted imaging (DWI) is useful for monitoring the therapeutic response after neoadjuvant chemotherapy in osteosarcoma of long bones. MATERIALS AND METHODS: Conventional magnetic resonance imaging (MRI) and DWI were obtained from 35 patients with histologically proven osteosarcomas. MR examinations were performed in all patients before and after 4 courses of preoperative neoadjuvant chemotherapy. Apparent diffusion coefficients (ADC) were measured. The degree of tumor necrosis was assessed macroscopically and histologically by two experienced pathologists after operation. Student's t test was performed for testing changes in ADC value. Pearson's correlation coefficient was used to estimate the correlation between necrosis rate and post- neoadjuvant chemotherapy ADC values. P<0.05 was considered to denote a significant difference. RESULTS: The difference of the whole osteosarcoma between pre- neoadjuvant chemotherapy ADC value (1.24±0.17×10(-3) mm(2)/s) and post- (1.93±0.39×10(-3) mm(2)/s) was significant difference (P<0.01). Regarding in patients with good response, the post- neoadjuvant chemotherapy values were significantly higher than the pre- neoadjuvant chemotherapy values (P<0.01). The post- neoadjuvant chemotherapy ADC value in patients with good response was higher than that of poor response (t = 8.995, P<0.01). The differences in post- neoadjuvant chemotherapy ADC between viable (1.03±0.17×10(-3) mm(2)/s) and necrotic (2.38±0.25×10(-3) mm(2)/s) tumor was highly significant (t = 23.905, P<0.01). A positive correlation between necrosis rates and the whole tumor ADC values (r = 0.769, P<0.01) was noted, but necrosis rates were not correlated with the ADC values of necrotic (r = -0.191, P = 0.272) and viable tumor areas (r = 0.292, P = 0.089). CONCLUSIONS: DWI can identify residual viable tumor tissues and tumor necrosis induced by neoadjuvant chemotherapy in osteosarcoma. The ADC value can directly reflect the degree of tumor necrosis, and it is useful to evaluate the preoperative neoadjuvant chemotherapy response in patients with osteosarcoma.

Hepatitis B virus-induced liver fibrosis and cirrhosis: the value of liver and spleen volumetry with multi-detector spiral computed tomography.

OBJECTIVE: To study the correlation of liver and spleen volume with the degree of liver fibrosis and cirrhosis induced by hepatitis B virus infection. METHODS: 128 participants who had undergone liver and spleen volumetry were enrolled. The control group consisted of 41 participants who were potential living liver donors. The liver fibrosis group consisted of 63 histologically proved liver fibrosis patients who were further divided into two subgroups: 44 patients with slight liver fibrosis, and 19 patients with advanced liver fibrosis. The liver cirrhosis group consisted of 24 patients. The following parameters were determined by multi-detector spiral computed tomography (MSCT) examination: total liver volume (TLV), right liver lobe volume (RV), left lateral liver segment volume (LLV), left medial liver segment volume (LMV), caudate lobe volume (CV), and spleen volume (SV). The ratios of CV to TLV (C/T), RV to TLV (R/T), LLV to TLV (LL/T), LMV to TLV (LM/T), and SV to TLV (S/T) were calculated. RESULTS: TLV, RV, LMV tended to decrease and SV, C/T, S/T tended to increase gradually with the increased degree of fibrosis. C/T >or= 3.34% and S/T >or= 47.36% were identified as the cut-off values of fibrosis >or=F3 (advanced liver fibrosis) and cirrhosis, respectively. Their sensitivities were 68.4% and 87.5% and their specificities were 59.1% and 89.5%, respectively. CONCLUSION: Variations in liver and spleen volume correlated with the degree of liver fibrosis and cirrhosis and could be used in the non-invasive follow-up of the development of liver fibrosis.