|
|
||||||||
Nuclear Medicine |
1 From the Departments of Nuclear Medicine (T.H., J.T., S.K., H.B., N.T.), Neurology (M.Y.), First Surgery (M.K.), Radiology (O.M.), and Pathology (A.N.), Kanazawa University School of Medicine, Takaramachi 13-1, Kanazawa, Ishikawa, Japan. Received June 5, 2000; revision requested July 28; final revision received April 3, 2001; accepted April 9. Address correspondence to T.H. (e-mail: higuchi@med.kanazawa-u.ac.jp).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: 201Tl SPECT and computed tomography (CT) were performed preoperatively in 46 patients with MG who had undergone thymectomy. SPECT was conducted 15 (early image) and 180 (delayed image) minutes after 201Tl injection. Results were visually assessed, and 201Tl uptake ratios (thymic lesion count density/lung count density) were measured for quantitative analysis. Uptake was analyzed among the normal thymus, LFH, and thymoma patient groups.
RESULTS: Histopathologic results indicated a normal thymus, LFH, and thymoma in 19, 16, and 11 patients, respectively. Mean uptake ratios in the normal thymus, LFH, and thymoma were 0.96 (95% CI: 0.90, 1.03), 1.14 (95% CI: 1.04, 1.25), and 1.87 (95% CI: 1.56, 2.25), respectively, on early images and 1.09 (95% CI: 1.00, 1.18), 1.65 (95% CI: 1.48, 1.85), and 2.03 (95% CI: 1.65, 2.50), respectively, on delayed images. Thymoma showed more intense 201Tl accumulation than did the normal thymus (P < .001) and LFH (P < .001) on early images. Both thymoma (P < .001) and LFH (P < .001) displayed more intense uptake than did the normal thymus on delayed images.
CONCLUSION: 201Tl SPECT can enable differentiation between normal thymus, LFH, and thymoma in patients with MG.
Index terms: Mediastinum, radionuclide studies, 675.12162 Myasthenia gravis Thallium, radioactive Thymus, CT, 675.1211 Thymus, neoplasms, 675.3154, 675.3155, 675.3169 Thymus, radionuclide studies, 675.12162
| INTRODUCTION |
|---|
|
|
|---|
Thymectomy has been attempted in a large number of patients with MG. The procedure is recognized as an effective treatment for sustaining improvement of symptoms and preventing invasiveness and metastatic spread of thymoma (4). Currently, thymectomy is used when medical treatment of patients is not successful or when thymoma is suspected. The factors that determine the response to thymectomy in patients with MG have been investigated in several reports (58). The authors concluded that the pathologic findings of the thymus might be important predictors of clinical outcome. Namely, LFH has been associated with the highest rate of clinical remission following thymectomy. Therefore, the preoperative evaluation of thymic status is important in planning the therapeutic strategy.
Moreover, minimally invasive surgical approaches, including transcervical incision and video-assisted thoracoscopic surgery, have been under investigation (9,10). These methods are not recommended in cases of thymoma because of the risk of incomplete resection that is attributable to technical difficulties and a limited field of vision (6,9). Currently, preoperative evaluation of the thymus associated with MG is performed with computed tomography (CT). However, CT has demonstrated insufficient resultsfor example, LFH does not always show an enlarged thymus, which may be overlooked at CT (1115).
Thallium 201 (201Tl) uptake is considered to reflect various factors, including cellular metabolic activity, regional blood flow, and the number of viable cells in the lesion (1619). Therefore, we hypothesized that, as a functional imaging technique, 201Tl single photon emission CT (SPECT) may be effective in the evaluation of thymic lesions associated with MG. The purpose of our study was to assess the usefulness of 201Tl SPECT in the evaluation of thymic lesions associated with MG, including LFH and thymoma.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Early and delayed 201Tl SPECT was performed 15 and 180 minutes, respectively, after 201Tl chloride (111 MBq) injection. A triple-detector gamma camera system (9300A; Toshiba Medical, Tokyo, Japan) equipped with low-energy, high-spatial-resolution, parallel-hole collimators was used. Data were acquired from 60 projections (40 seconds per projection) over 360° by using 64 x 64 matrices. Filtered-back projection with ramp filters and Butterworth filters with an order of 8 and cutoff frequencies of 0.31 cycles per centimeter was performed for reconstruction. All patients were instructed to relax while in the chair for at least 10 minutes before 201Tl injection; moreover, exercise was prohibited until after delayed imaging. The patients arms were positioned upward during image acquisition.
201Tl uptake in the thymic lesion was evaluated on the early and delayed 201Tl images both visually and quantitatively with knowledge that the patients had MG. At visual analysis, two observers (T.H., J.T.) independently evaluated the degree of radionuclide uptake in the thymic lesion by using the following five-point uptake scoring system, with no information on the CT findings: 0, uptake equal to background activity within the lung; 1, slight increase in uptake; 2, moderate increase in uptake; 3, strong increase in uptake; and 4, uptake greater than heart uptake. When uptake scores differed between the observers, a third observer (S.K.) was consulted and differences were resolved with consensus.
At quantitative analysis, two authors (T.H., J.T.) in consensus manually drew regions of interest (ROIs) on the entire area of uptake in the thymic lesion or thymus on the transverse sections that demonstrated the highest 201Tl uptake (Fig 1). To set ROIs, the CT images were referenced to determine the location and size of the thymic lesions. The background ROI was set on the entire right lung depicted on a 201Tl section identical to the SPECT section that showed the thymic lesion (Fig 1). The count densities of the ROIs were measured. The uptake ratio was then calculated by dividing the count density of the thymic lesion by the count density of the background.
|
Histopathologic examination of specimens obtained at surgery was conducted by one of the authors (A.N.), who was blinded to 201Tl SPECT and CT results. Samples were defined as thymoma, LFH, or normal thymus.
Differences in age, 201Tl uptake ratio, change in 201Tl uptake ratio between early and delayed imaging, and 201Tl uptake score were statistically analyzed among the three histopathologic groups: normal thymus, LFH, and thymoma. The association between age and 201Tl uptake was assessed by using the Pearson correlation test.
For multiple-group comparisons of age and 201Tl uptake ratio, homogeneity of variance was assessed by using the Bartlett test. In addition, parametric comparisons were performed with one-way analysis of variance. The significance of individual differences was evaluated by using the Scheffé F test in cases in which one-way analysis of variance results were significantly different. Changes in 201Tl uptake ratio on early and delayed images were assessed by performing repeated-measures analysis of the histopathologic findings. Moreover, changes in the 201Tl uptake ratio in the individual histopathologic groups were assessed by using the paired t test. Comparisons of 201Tl uptake scores were evaluated by using the Kruskal-Wallis H test, followed by the Dunn test for multiple comparisons, because the 201Tl uptake score was assumed to be rank ordered. P values less than .05 were considered to indicate statistically significant differences.
Normally distributed data were expressed as means plus or minus the SD. Data that were not normally distributed were logarithmically transformed to improve normality before statistical analysis was performed. Data were expressed as the mean with a 95% CI, which was calculated in logarithmic scale and back transformed to the original units.
| RESULTS |
|---|
|
|
|---|
The results of visual analysis of the 201Tl SPECT images are summarized in Table 1. Meaningful differences in 201Tl uptake scores were observed on early images (
2 = 34, df = 3, P < .001). The patients with LFH (Fig 2) and a normal thymus did not show significant differences at SPECT; however, those with thymoma (Fig 3) had elevated scores compared with those of the other two groups (P < .01). There were also significant differences in 201Tl uptake scores at delayed SPECT (
2 = 30.5, df = 3, P < .001). The uptake scores of the patients with LFH and thymoma were not markedly different; however, compared with the other two groups, the normal thymus group had lower scores (P < .01).
|
|
|
|
|
|
|
There were significant differences in changes in 201Tl uptake ratio among the groups (Scheffé F = 10.4, df = 2, P < .001). Increases in uptake ratio from early to delayed image were significant in the LFH (P < .001) and normal thymus groups (P = .025) (Fig 4). No meaningful changes (P = .508) were observed in the thymoma group.
|
|
At visual analysis, the lesions with a score of 0 for both early and delayed images were considered normal. Lesions with an early image score of 0 and a delayed image score higher than 1 were considered to be LFH. Lesions with an early image score higher than 1 were considered to be thymoma. The numbers of patients with a diagnosis of LFH and thymoma according to these image criteria are presented in Table 2.
With 201Tl SPECT, accurate diagnoses were obtained in 40 (87%) and 36 (78%) of 46 patients by using quantitative analysis and visual uptake scores, respectively. In contrast, accurate diagnoses were obtained with CT in 32 (70%) of 46 patients. The differing proportions of accurate diagnoses achieved by using 201Tl uptake ratio and CT were evaluated by performing the Fisher exact test. 201Tl uptake ratio was more accurate than CT in enabling the detection of LFH (13 vs five of 16 patients; P = .006). Moreover, no significant difference in thymoma detection was observed between the two modalities (nine vs 11 of 11 patients; P = .23).
| DISCUSSION |
|---|
|
|
|---|
According to the results of this study, thymoma should be considered when increased 201Tl uptake is detected on both early and delayed images, whereas LFH should be suspected when more intense uptake is observed on the delayed image relative to that on the early image. Detection of LFH with 201Tl SPECT was superior to that with CT in this study. In addition, our study results were superior to those obtained in previous investigations (1115) involving CT. LFH diagnosis is based on microscopic findings of thymic germinal centers; furthermore, LFH associated with MG occasionally does not manifest actual macroscopic enlargement (13,15). This observation may account for the poor detection of LFH with CT in our study.
Visual analysis of 201Tl SPECT findings revealed that CT had similar sensitivity in the detection of thymoma. However, CT may have some advantages over 201Tl SPECT in the detection of small tumors or in the assessment of tumor shape or precise tumor location owing to its higher spatial resolution. In contrast, the limitation of CT in depicting thymoma is due to difficulty in differentiating thymoma from LFH on the basis of lesion shape. Moreover, LFH may appear as a focal mass at CT and consequently be incorrectly diagnosed as thymoma (12). Two such cases of LFH occurred in this study.
The mechanism of 201Tl uptake is not clearly understood. Factors that possibly influence 201Tl uptake include blood flow, cell viability, number of cells in the lesion, tumor type, Na+-K+ ATPase, calcium ion channel, and vascular permeability (1618, 23,24). Thymoma has higher vascularity and greater blood flow; therefore, rapid delivery of 201Tl is promoted and may result in greater 201Tl uptake by thymoma lesions on early images. In LFH, 201Tl accumulates mainly on the delayed image. LFH vascularity is usually lower than that of thymoma. LFH, however, exhibits lymphoid follicles that consist of packed small cell nests displaying higher cellular activity. This occurrence may be a reason for the 201Tl accumulation mainly on the delayed image in LFH.
In the current study, neither thin-section CT nor contrast material administration was performed. However, in a previous investigation (11), contrast enhancement did not provide additional information in the differential diagnosis of thymic lesions in patients with MG. The advantage of thin-section CT is that it enables the observation of lesion details; however, its effectiveness to enable the diagnosis of LFH in MG is unknown.
Magnetic resonance (MR) imaging was not performed in our study patients. To our knowledge, there are few reports regarding the MR imaging evaluation of thymic hyperplasia in patients with MG. However, Batra et al (13), in a series of 16 patients with MG, examined the correlation between CT and MR imaging of thymic lesions associated with MG and concluded that MR examination of the thymus added no clinically important information to the CT results.
Uptake of indium 111 (111In) octreotide, a somatostatin receptor imaging agent, was reportedly substantial in thymoma but not in LFH; this finding indicated the capability of this agent to enable differentiation of thymoma from LFH (25). However, 111In octreotide cannot enable the differentiation of LFH from a normal thymus, because it does not accumulate in either LFH or the normal thymus.
In several reports (2628), thymic lesions have been evaluated with 2-[fluorine-18]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET). Conclusions indicate that FDG may be helpful in evaluating the malignant nature of mediastinal tumors. In addition, FDG has an advantage over 201Tl with respect to quantification of uptake (29). However, FDG PET is not available in most medical centers. Moreover, to our knowledge, the evaluation of thymic lesions associated with MG by using FDG PET has not yet been investigated.
Visual and quantitative evaluations of 201Tl accumulation were performed in the current study. Although visual analysis for the detection of LFH was superior to that of CT, quantification of 201Tl uptake allowed more accurate diagnoses. Therefore, we recommend calculating the 201Tl uptake ratio in each case, in addition to performing visual evaluation for assessment of the thymus in MG. The entire lung on a 201Tl SPECT section was used as a reference ROI in the quantitative evaluation, because the ROI setting on the entire lung was readily obtained and reproducible on the identical SPECT section of the thymic lesion. In the current investigation, no patients presented with symptoms of pulmonary disease or heart failure, which may cause 201Tl uptake in the lung (30) and thereby result in an underestimation of the uptake ratio. Determining the reference ROI for calculation of the uptake ratio requires particular caution in such cases.
In summary, accumulation of 201Tl on early and delayed SPECT images reflects the histopathologic findings of thymic lesions associated with MG. In our patient population, the normal thymus showed no increase in 201Tl uptake on either early and delayed images, whereas LFH exhibited moderate 201Tl uptake on only delayed images. Moreover, thymoma displayed significant (moderate to strong) 201Tl uptake on early and delayed images. 201Tl SPECT had greater sensitivity for the diagnosis of LFH than did CT. The results of this study may justify routine use of this scintigraphic method as a preoperative examination in patients with MG. Further prospective evaluation of this method with large patient populations is warranted.
| FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, N.T.; study concepts and design, J.T., T.H.; literature research, T.H.; clinical studies, A.N., M.Y., M.K.; data acquisition, T.H., N.T.; data analysis/interpretation, T.H., J.T., S.K., O.M.; statistical analysis, T.H.; manuscript preparation, T.H.; manuscript definition of intellectual content, J.T., S.K., T.H.; manuscript editing, T.H., J.T., S.K.; manuscript revision/review, J.T., H.B., M.Y.; manuscript final version approval, all authors.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Inaoka, K. Takahashi, M. Mineta, T. Yamada, N. Shuke, A. Okizaki, K. Nagasawa, H. Sugimori, and T. Aburano Thymic Hyperplasia and Thymus Gland Tumors: Differentiation with Chemical Shift MR Imaging Radiology, June 1, 2007; 243(3): 869 - 876. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Iwata, K. Inoue, S. Mizuguchi, R. Morita, T. Tsukioka, and S. Suehiro Thymic Small Cell Carcinoma Associated With Pulmonary Squamous Cell Carcinoma Ann. Thorac. Surg., December 1, 2006; 82(6): 2266 - 2268. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Santana, A. Givica, and C. Camacho Best Cases from the AFIP: Thymoma RadioGraphics, October 1, 2002; 22(90001): S95 - 102. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |