| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Nuclear Medicine, Seton Medical Center, 1900 Sullivan Avenue, Daly City, CA 94015. e-mail: stephengerard@dochs.org
Editor:
In the article by Dr Dam and colleagues in the August 2004 issue of Radiology (1), the authors claimed a lack of correlation between visually observed stunning with iodine 131 (131I) therapeutic scanning and suboptimal therapeutic outcome. A number of serious methodologic concerns compromise this investigation.
Qualitative scan comparisons used in this study may be an unreliable index of decreased radioiodine uptake and stunning. First, differences in the postdose imaging times between the diagnostic and therapeutic scans (2 days vs 7 days, respectively) may confound their comparison. Second, it is difficult to standardize image intensity over a potentially large 20–40-fold dose range without any quantitative measurements. For example, the authors claim to show stunning in figure 1. However, there is no description of any methodology as to how these images were normalized to each other. Clearly, target-to-background ratio is lower in the posttreatment scan, but this could be primarily a result of the longer postdose imaging interval, as well as the 9-day-longer stimulation by elevated thyroid-stimulating hormone, or TSH, increasing the rate of iodine turnover (2). This concern is even more evident when comparing the images in figure 2a and 2b. How were these images graded with regard to stunning?
Another major technical criticism is that the authors used medium-energy collimation for 131I imaging, which compromises scan resolution and confounding comparisons. High-energy collimation should always be used for imaging this high-energy gamma emitter (3).
The authors have defined therapeutic efficacy solely in terms of absent thyroid tissue in follow-up diagnostic radioiodine scans. However, thyroglobulin levels after TSH stimulation are used routinely in conjunction with radioiodine scanning to augment sensitivity (4).
The authors incorrectly claim in the discussion that there is no conclusive evidence that stunning results in unsuccessful ablation. Lees et al (5) reported a statistically significant increase in both the number of therapeutic 131I administrations and the average cumulative 131I dose required to achieve effective ablation when 185-MBq (5-mCi) 131I diagnostic scans preceded therapeutic 131I administration. Similar significant differences in the required total cumulative 131I dose and the number of total treatments were reported by Chmielowiec et al (6). Muratet et al (7), whose study is reference 8 in the article of Dr Dam and colleagues, also reported a statistically significant 25% reduction in therapeutic efficacy with a prior 111-MBq (3-mCi) 131I diagnostic dose versus that with a 37-MBq (1-mCi) dose.
The consensus of evidence in favor of stunning (5–14) compels serious consideration of its avoidance by reducing the diagnostic 131I dose to a minimum—that is, up to 74 MBq (2.0 mCi) (15)—or by using 123I for diagnostic scanning (16,17).
REFERENCES
Division of Nuclear Medicine, Department of Medicine, Christiana Care, 4755 Ogletown-Stanton Road, Newark, DE 19718. e-mail: hdam@christianacare.org
We thank Dr Gerard for his interest in and his comments regarding our article published in August 2004 (1).
We believe that there are many variables in any comparison of diagnostic and postablation scans. These variables include the interval between diagnostic dose and scanning, interval between ablation dose and scanning, and differences in diagnostic dose and ablation dose, as Dr Gerard has noted. Our study was retrospective in nature, however, and such variables could not be controlled. One of his criticisms of our article was the difference in imaging time between the diagnostic and postablation scans (2 days vs 7 days). However, we chose to perform the postablation scan at 7 days because of improved target-to-background ratio compared with a 2–3-day postablation scan (2).
The images in figure 2a and 2b were assigned grade 2 (no change between the scans) on our semiquantitative scale. Our study population consisted of a 9-year retrospective review from 1994 to 2002. Since images obtained in most of the scans (1994–1999) were analog, we could not quantitatively compare the scans. Nevertheless, we do not agree that qualitative data analysis in our series is as unreliable as Dr Gerard asserts. We were not describing subtle differences between diagnostic and postablation scans for our stunned population. As stated in our materials and methods section, we defined stunning as grade 0 (a definite decrease between the two scans). Therefore, purely quantitative analysis is not needed, and qualitative analysis is more than sufficient. Furthermore, we have seen cases where thyroid tissue activity seen in the diagnostic scan is not even evident (ie, absence of activity) in the postablation scan, regardless of changing intensity on digital display. In such cases, normalization would have been of little benefit.
Although we defined treatment success as an absence of visualized thyroid tissue on follow-up diagnostic scans, we agree with Dr Gerard that an elevated serum thyroglobulin level is useful for detection of recurrent disease. Our study spanned back to 1994, before thyroglobulin levels were used routinely in clinical practice. As a result, many of our earlier patients did not have serum thyroglobulin levels checked. The exclusion of these patients would have compromised the statistical power of our study. In addition, our referring endocrinologists were satisfied that there was no recurrent disease with eventual normal follow-up diagnostic 131I scans performed during the earlier years of our study.
We disagree with Dr Gerard in regards to the effect of stunning on ablation success rates. The data from the study by Lees et al (3) directly conflict with results of a similar study by Morris et al (4), who reported no statistically significant difference in initial treatment ablation rates between a patient who underwent 111–185-MBq 131I diagnostic scanning and those who did not undergo 131I diagnostic scanning prior to therapy. Although Lees et al stated that this discordant result between the two studies may be due to the variability in initial 131I therapy doses in the study by Morris et al, they failed to provide the data in their own report.
The study of Muratet et al (5), which showed a significant (P < .005) decrease in therapeutic efficacy between prior 111-MBq (3-mCi) 131I versus 37-MBq (1-mCi) diagnostic dose, implied that the entire sample population underwent stunning. As noted in our discussion section, we agree that all patients seemed to have a baseline level of stunning from the diagnostic 131I doses. However, we believe that there may be an additional subset of patients who undergo "visual" stunning. Clearly, the existence of stunning and its effect on therapeutic efficacy remains controversial.
On the basis of the results of our report, which included patients with thyroid carcinoma metastases and had relatively long-term follow-up (mean, 30.0 months), we agree with other investigators (4,6,7), who found no difference in treatment outcomes as a result of stunning. At this juncture, some investigators (8,9) have advocated the use of 123I for diagnostic imaging of thyroid carcinoma. We agree that the use of 123I for diagnostic imaging does have merit, considering that 123I is a pure gamma emitter without any therapeutic effect. However, there are cost constraints with the use of 123I. Furthermore, the sensitivity of 123I is inferior to 131I for the detection of thyroid carcinoma metastases (10). Further research will help clarify these matters.
Ideally, a high-energy collimator should be used for imaging a high-energy gamma emitter. However, other investigators have used also used medium-energy collimators for imaging 131I (3,10). Since a high-energy collimator could only be used with 131I in our practice, we chose to purchase a high-quality medium-energy collimator, since it is more versatile in a clinical setting.
Since publishing our article, we have identified three additional patients who met our criteria for stunning and have had negative follow-up diagnostic scans and negative serum thyroglobulin levels. We remain convinced that the "visual" stunning phenomenon does not result in decreased therapeutic efficacy.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  K. P. Donahue, N. P. Shah, S. L. Lee, and M. E. Oates Initial Staging of Differentiated Thyroid Carcinoma: Continued Utility of Posttherapy 131I Whole-Body Scintigraphy Radiology, March 1, 2008; 246(3): 887 - 894. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
