Incidental diagnosis of tumor thrombosis on FDG PET/CT imaging

OBJECTIVE

Clinical data are presented on patients with tumor thrombosis (TT) incidentally detected on FDG PET/CT imaging, as well as determining its prevalence and metabolic characteristics.

MATERIALS AND METHODS

Out of 12,500 consecutive PET/CT examinations of patients with malignancy, the PET/CT images of 15 patients with TT as an incidental finding were retrospectively investigated. A visual and semiquantitative analyses was performed on the PET/CT scans. An evaluation was made of the pattern of FDG uptake in the involved vessel as linear or focal via visual analyses. For the semiquantitative analyses, the metabolic activity was measured using SUVmax by drawing the region of interest at the site of the thrombosis and tumor (if any).

RESULTS

The prevalence of occult TT was 0.12%. A total of 15 patients had various malignancies including renal (1 patient), liver (4), pancreas (2), stomach (1), colon (1), non-Hodgkin lymphoma (1), leiomyosarcoma (1), endometrial (1), ovarian (1), malign melanoma (1) and parotid (1). Nineteen vessels with TT were identified in 15 patients; three patients had more than one vessel. Various vessels were affected; the most common was the inferior vena cava (n=7) followed by the portal (n=5), renal (n=3), splenic (n=1), jugular (n=1), common iliac (n=1) and ovarian vein (n=1). The FDG uptake pattern was linear in 12 and focal in 3 patients. The mean SUVmax values in the TT and primary tumors were 8.40±4.56 and 13.77±6.80, respectively.

CONCLUSION

Occult TT from various malignancies and locations was found incidentally in 0.12% of patients. Interesting cases with malign melanoma and parotid carcinoma and with TT in ovarian vein were first described by FDG PET/CT. Based on the linear FDG uptake pattern and high SUVmax value, PET/CT may accurately detect occult TT, help with the assessment of treatment response, contribute to correct tumor staging, and provide additional information on the survival rates of oncology patients.

Silent coronary artery disease in patients with type 2 diabetes mellitus

The purpose of this study was to estimate the prevalence and risk factors of silent CAD in asymptomatic type 2 diabetic patients aged over 40 years. A total of 172 asymptomatic type 2 diabetic patients, mean age 54.42 years, with normal resting electrocardiogram were included in the study. Technetium-99m (Tc-99m) tetrofosmin cardiac single photon emission computed tomography myocardial scintigraphy with exercise testing or dipyridamole injection was performed on all patients. If this test was positive, coronary angiography was carried out and was considered to be positive with a stenosis of > or =70%. Abnormal perfusion pattern was found in 14 patients (8.14%). Significant coronary artery stenosis was found in 13 subjects (7.56%), confirming a high positive predictive value (92.86%) of this diagnostic procedure. A significant correlation was observed between silent CAD and male sex, retinopathy, hypertension, post-prandial blood glucose level, and low HDL-cholesterol level. Sex (OR=4.026; 95% CI, 1.187-13.659), hypertension (OR=5.564; 95% CI, 1.446-21.400) and retinopathy (OR=3.766; 95% CI, 1.096-12.948) were risk factors for CAD. Overall, 14.06% of asymptomatic male patients with type 2 diabetes mellitus presented silent CAD with significant angiographically documented coronary stenosis. This finding, along with the high positive predictive value of a noninvasive technique, indicates that routine screening for silent CAD would be useful in this patient subgroup especially when they have retinopathy or hypertension.

99mTc-tetrofosmin scintigraphy in musculoskeletal tumours: the relationship between P-glycoprotein expression and tetrofosmin uptake in malignant lesions

The aims of this study were to assess the role of (99m)Tc-tetrofosmin scintigraphy in the diagnosis of malignant vs. benign musculoskeletal tumours and to determine the relationship between P-glycoprotein expression and tetrofosmin uptake in malignant lesions. Forty-six patients (32 malignant, 14 benign) with various musculoskeletal lesions were studied. Each patient underwent (99m)Tc-methylene diphosphonate three-phase bone scanning initially. At least 2 days later, dynamic and static (99m)Tc-tetrofosmin scans were obtained. The tetrofosmin scans were evaluated by visual and quantitative analysis. The count ratio of the lesion to the contralateral normal area (uptake ratio, UR) was calculated from the region of interest drawn on the tetrofosmin scan. The lesions were then resected by open biopsy to obtain a histopathological diagnosis. P-glycoprotein levels were determined immunohistochemically in 22 of 32 malignant lesions. A significant difference between the mean UR values of benign and malignant lesions was found (1.36 +/- 0.47 vs. 3.35 +/- 2.08, P = 0.000). Visual analysis showed an accuracy of 85%, and the accuracy of the quantitative analysis was 87% with the threshold level of UR as 1.76. When perfusion findings were added to the evaluation criteria, the accuracies of visual and quantitative analysis were increased to 87% and 89%, respectively. The relationship between the levels of P-glycoprotein and the UR values of tetrofosmin was not statistically significant (r = -0.235, P = 0.2). In addition, the mean UR value of the patients with P-glycoprotein expression was not statistically different from that of the patients without P-glycoprotein expression (3.01 +/- 1.48 vs. 4.27 +/- 2.90, P = 0.297). In conclusion, visually significant tetrofosmin uptake and increased perfusion in a musculoskeletal lesion strongly suggest that the lesion is malignant (positive predictive value, 96%). P-glycoprotein expression was not found to be a major factor interfering with 30 min tetrofosmin uptake in a malignant musculoskeletal lesion. However, the relatively high false-negative rate among negative results (28%) limits the value of (99m)Tc-tetrofosmin scintigraphy as a single diagnostic tool in differentiating between benign and malignant musculoskeletal tumours.

The efficacy of technetium-99m ciprofloxacin (Infecton) imaging in suspected orthopaedic infection: a comparison with sequential bone/gallium imaging

Technetium-99m ciprofloxacin (Infecton) has recently become established as a new radiopharmaceutical for the imaging of infection. This study was performed to determine the value of Infecton imaging in demonstrating orthopaedic infection and to compare the results with bone/gallium imaging. Twenty-two patients (12 female, 10 male; mean age 51.7+/-16.8 years) with suspected orthopaedic infective conditions were included in the study. The patients underwent three scintigraphic studies in the following sequence: 740 MBq 99mTc-methylene diphosphonate (MDP) three-phase bone scintigraphy; at least 2 days later, 370 MBq Infecton scan at 1-4 h; and finally, 185 MBq gallium-67 scintigraphy. 67Ga imaging could not be performed on four patients. All images were blindly interpreted by two independent observers. The final diagnosis was made by consensus when the readings were different. Interpretation of the early and late Infecton images was made separately, with visual findings being classified according to a four-grade scale (0, +, ++, +++). Images graded 0 and +, and also those regions which showed a decrease in uptake grade on late images as compared with early images, were classified as negative for infection; grades ++ and +++ were classified as positive. Bone/gallium images were considered positive when the images were spatially incongruent or when gallium uptake was more intense than that of 99mTc-MDP. The diagnosis was confirmed by intraoperative microbiological or histological findings, or by the presence of gross purulence. The sensitivity of Infecton imaging was found to be 85%, the specificity 92% and the accuracy 88%, as compared to figures of 78%, 100% and 90%, respectively, for bone/gallium imaging. Although the two modalities showed a similar clinical yield, the easy availability of Infecton and the short investigation time make Infecton imaging the better option for the detection of orthopaedic infection.

Prospective validation of single plasma sample 99mTc-ethylenedicysteine clearance in adults

99mTc-L,L-ethylene, L, dicysteine (EC) clearance shows strong correlation with orthoiodohippurate clearance, and it is possible to estimate effective renal plasma flow from 99mTc-EC clearance. In routine clinical studies, it is practical to use the one or two plasma sample method instead of multiple plasma samples for clearance determination. A single-sample technique was developed for 99mTc-EC, and a regression formula was generated. A prospective study tested the validity of this regression formula.

METHODS

The study population was composed of 26 patients with a wide range of renal function. Multiple plasma sample 99mTc-EC clearances were calculated from all patients using the open two-compartment model. Single plasma sample clearances were also determined from the 54-min plasma sample using the regression formula published previously.

RESULTS

The multiple-sample plasma clearance of 99mTc-EC ranged from 46 to 668 mL/min with a mean of 300.76 +/- 150.73 mL/min. The clearances obtained from the 54-min plasma sample ranged from 49 to 699 mL/min, with a mean of 297.39 +/- 152.23 mL/min. There was an excellent correlation between the clearances obtained by the two techniques (r = 0.99, slope = 0.9911). The standard error of estimation was found to be 25.9 mL/min.

CONCLUSION

This study suggests that 99mTc-EC clearance can be estimated from 54-min plasma samples with an acceptable error of estimation for most routine clinical studies.

Simplified technetium-99m-EC clearance in adults from a single plasma sample

Technetium-99m-EC has recently been introduced as an alternative renal tubular agent to 131I-ortho iodohippurate (OIH). It has been shown that 99mTc-EC clearance shows strong correlation with OIH clearance and it is possible to estimate effective renal plasma flow. In routine clinical studies, it is practical to use one or two plasma sample methods instead of multiple plasma samples for clearance determination. An attempt was made to determine 99mTc-EC clearance with one sample method.

METHODS

Data from 72 subjects were collected. To increase the range of renal function, two anuric hemodialysis patients were also included. Clearances were determined by the open two-compartment model.

RESULTS

The clearance range was 12 ml/min to 660 ml/min with a mean of 275 +/- 117 ml/min. Analysis of correlation was made by Tauxe's method. The least standard error of estimation (s.e.e. = 32.71 ml/min) and the best correlation (r = 0.97) between the theoretical volume distribution and the clearance estimations were obtained from the 54-min plasma sample.

CONCLUSION

This study suggests that EC clearance could be determined by a simplified single-sample method with an acceptable s.e.e.