Lacrimal gland and nasolacrimal duct uptake on ¹²⁴I PET/CT imaging.

Background: While it is well known that radioiodine is visualized in the lacrimal system correlating to known locations sodium-iodide symporter, the characteristics of that uptake in the lacrimal glands have not been previously described in any detail. The objective of this study was to evaluate the frequency, intensity, and various patterns of iodine uptake in the lacrimal glands (LG) and nasolacrimal ducts (NLD) on ¹²⁴I positron emission tomography (PET) imaging in patients with or without prior ¹³¹I therapy for differentiated thyroid cancer (DTC).

Methods: The performance of ¹²⁴I PET imaging has been previously described in detail (Van Nostrand et al. 2010 Thyroid). In brief, DTC patients were prepared for imaging with low-iodine diet and adequate TSH stimulation. ¹²⁴I PET and low-dose localization CT were performed at 2h, 24h, 48h, 72h, and 96h after oral administration of 31.5 or 62.9 MBq (0.85 or 1.7 mCi) of ¹²⁴I. Images were obtained with GE Advance Nxi (5 min/bed position) and Philips Gemini TF 64 (4 min/bed position, CT attenuation scans, 120 kV, 30 mAs). Each scan was independently reviewed by two physicians, and discordant interpretations were resolved by consensus. ¹²⁴I uptake in the left (L) and right (R) lacrimal gland (LG) and nasal lacrimal sac/duct (NLD) was tabulated for frequency, location, pattern, and intensity. Intensity Grade was visually assessed as 0= no uptake; 1= equivocal uptake; 2= definite but mild uptake; 3= moderate uptake; 4=intense uptake. The visual patterns were characterized as: circular, oblong, linear, circular on top and lower linear, dumbbell, and irregular.

Results: A total of 173 ¹²⁴I PET/CT scans (i.e., 35 sets of five time points) were reviewed for 31 DTC patients. However, two scans were not available. In these 35 sets of ¹²⁴I scans, the following were visualized: 33 L-NLD, 31 R-NLD, 4 L-LG and 4 R-LG. L-NLD was most frequently seen at 2h & 24h; R-LG, L-NLD, and R-NLD were all most frequently visualized at 2h. Visual intensity of L-NLD and R-NLD were greatest at 2h & 24h with more cases of moderate uptake (grade 3) seen at 2h. Lacrimal glands were difficult to visualize due to the low uptake intensity compared to nasolacrimal ducts; L-LG and R-LG were only graded as equivocal or mild uptake at either 2h or 24h. Most of the NLD uptake pattern was circular (30-35%), oblong (27-29%), or linear (26-27%). On comparison between THW and rhTSH-stimulated scans, there was no statistical difference in the frequency of ¹²⁴I uptake in the L-NLD (14/14 vs 20/21), R-NLD (13/14 vs 18/21), L-LG (1/14 vs 3/21) and R-LG (1/14 vs 3/21). Also, there were no difference in the highest intensity (Grade 4) of ¹²⁴I uptake in the L-NLD (5/14 vs 7/21) and R-NLD (2/14 vs 8/21) on THW versus rhTSH-stimulated scans. However, patients who underwent THW seemed to have more frequent and more intense uptake at 24h, while patients injected with rhTSH seemed to have more frequent and more intense uptake at 2h. In comparing ¹²⁴I scans between patients with or without prior ¹³¹I therapy, the numbers of patients who did not have prior ¹³¹I therapy were too low to reach a statistical significant.

Conclusions: In the nasolacrimal duct, ¹²⁴I activity is frequently seen with variable intensity and most frequently seen at 2h and 24 hours. In the lacrimal gland, ¹²⁴I activity is infrequently seen, has low intensity, and is best seen at 2h and 24 hours. The patterns of activity of the lacrimal sac and duct are more variable than the patterns seen with the lacrimal glands and may range from circular, oblong, or linear. The pattern observed may depend on the relationship of the cross-sectional image to the sac and/or duct being sliced. There is no significant difference between THW and rhTSH-stimulated scans in visualizing NLD and LG, apart from the optimal time for visualization.