Estimation of Chest Wall Attenuation of ¹²³I Emissions in Substernal Goiter: A Phantom Study

DISCUSSION

Substernal goiter is the process wherein all or a portion of the thyroid gland extends into the mediastinum. It has been estimated to occur in 6% of individuals being evaluated for thyroidectomy (3,4). By including asymptomatic patients, the overall prevalence would be lower. Substernal goiter is often seen in elderly patients and is more common in women than men (6). Although substernal goiters are more often hypofunctioning, patients with hyperthyroidism may have a substernal component to their thyroid gland.

Measurement of thyroid uptake is a commonly performed procedure in which the fraction of ingested iodine within the thyroid gland is calculated after a specified period (usually 24 h and sometimes with the addition of a 4- to 6-h measurement). This test plays a role in characterizing a patient with hyperthyroidism. The measurement is often used to calculate the dose of ¹³¹I for treating hyperthyroidism. There are several different ways that this calculation can be done.

We found that our neck phantom attenuated the capsule emissions by 18%—presumably, the approximate attenuation of ¹²³I emissions for an average-sized neck. For a cadaver chest wall containing the sternum, the attenuation increased to 35%, essentially doubling.

The implication is that the thyroid uptake measurement is likely to be incorrect if a portion of the thyroid is substernal—a possibility that may be of consequence if the measurement is used to calculate ¹³¹I doses for treating hyperthyroidism. Uptake measurements may be underestimated, leading to an increase in the calculated therapeutic dose. Administering doses of ¹³¹I higher than necessary may lead to an increase in both short-term side effects and long-term unwanted consequences. It is even possible that, with uptake underestimation, the therapy will not be performed. This has a lower significance for the treatment of thyroid cancer with ¹³¹I: most treated patients will have undergone thyroidectomy, and uptake measurements are used less frequently in this population.

How would this translate into therapy dose calculations? Presume that the thyroid has an uptake of 50% and the goal is to administer 370 MBq (10 mCi) for treatment of hyperthyroidism. The administered dose would be 740 MBq (20 mCi). In the situation of a normally positioned, cervical thyroid gland, no additional calculation is needed. The phantom simulates neck attenuation. In the second scenario, the thyroid gland is completely substernal. Uptake would be [(1.00 − 0.35)/(1.00 − 0.18)], or 79% of what it would have been if the gland were normally positioned. The thyroid uptake would be measured as 0.50 × 0.79, or 0.40. The administered dose would be 370 MBq (10 mCi)/0.40, or 925 MBq (25 mCi), rather than 740 MBq (20 mCi). If the thyroid uptake is lower, the effect is magnified. In the case of an uptake of 30%, the administered dose would rise from 370 MBq (10 mCi)/0.3, or 1,221 MBq (33 mCi), to 370 MBq (10 mCi)/(0.3 × 0.79), or 1,554 MBq (42 mCi). This is a worst-case scenario. Most substernal glands are not completely behind the sternum, and measurement error would be less extreme.

Park et al., in an article (7) describing their study evaluating the ability of thyroid scintigraphy to detect intrathoracic glands, briefly mentioned that they measured 21% attenuation of ¹²³I photons through a fresh adult cadaver sternum. Details regarding how this was calculated were not provided.

There are several ways in which this analysis will differ from real-life scenarios. First, the thyroid gland is rarely located completely behind the sternum. The cadaver chest wall is an approximation of the chest wall of a living human. Its constituent tissue will not have identical attenuation coefficients. Finally, we used 1 cadaver specimen; chest walls in cadavers, as in living patients, come in many shapes and sizes. There is no phantom that can function as an exact representation of a patient. Our goal was to provide a rough idea of how severe the attenuation may be.

Another separate, though related, problem is that of improper aiming of the thyroid probe. The probe is a collimated device, and moving the target away from its center point will lower the count. It would be interesting to investigate the effect of malpositioning of a thyroid probe.

It is likely that many patients who undergo ¹³¹I therapy will not have had their anatomy defined, and it will be unknown whether there is a substernal component of their thyroid gland. There are patients, however, in whom this is known. Patients may have had thyroid scintigraphy with a sternal marker performed before therapy. Some patients will have had a prior CT scan of their neck or chest. In this scenario, consideration might be given to lowering the treatment dose from what would have been administered otherwise. There are also CT-based methods to calculate the actual attenuation coefficient of the interceding chest wall using the thickness of the tissues, density calculation with Hounsfield units, and the mass attenuation coefficients of the tissues. The chest wall may include soft tissue, cortical bone, cancellous bone, marrow, and cartilage. It is highly doubtful that there would be benefit from this demanding calculation. We do not recommend actually performing a CT scan on hyperthyroid patients solely for this purpose. Our aim was to provide an understanding of the potential magnitude of the attenuation.