TO THE EDITOR:
In their paper on salivary gland damage after high-dose radioiodine ablation therapy, Nakada et al. discuss an interesting and practical issue (1). The authors propose a radiation protection paradigm based on a concept of the dynamics of salivary gland function. The authors claim to have compared 2 radiation protection regimens in similar patients groups: a group that started sucking lemon candy straight after radioiodine treatment (group A), versus a group that started 24 h later (group B). Based on this comparison, the authors recommended that patients start sucking lemon candy no earlier than 24 h after radioiodine treatment.
Some aspects of this study are puzzling.
First, the authors state that “On encountering unexpectedly higher salivary gland side effects in group A, the patients in group B tended to be treated more intensively with steroids or nonsteroidal antiinflammatory drugs for sialoadenitis and with a drug containing zinc acetate or vitamin B12 for taste dysfunction” (1). Might these differences in medical treatment (which were not specified quantitatively) have contributed to the observed reduction of radiation damage in group B?
Second, as cited by the authors, the reported incidence of salivary gland injury varies considerably depending on the diagnostic criteria (2–5). The incidence of sialoadenitis, dry mouth, and loss of taste was investigated. Which criteria were used to define these primary endpoints?
Third, which criteria were used to consider including sialoscintigraphy in the assessment of salivary function? What was the total number of these procedures in each group, and what was the outcome? We believe that these data are relevant because of the reported disagreement between subjective symptoms and sialoscintigraphy.
Fourth, the authors’ concept of salivary gland function has not been studied physically. It is uncertain whether at any time point there is indeed an imbalance of the salivary blood flow and the counteracting saliva flow. Whether delayed stimulation of the salivary glands has a radioprotective effect therefore remains a matter of debate. If delayed stimulation is indeed beneficial, why should the optimal starting point of sucking lemon candy be 24 h?
If this extra information could be supplied, the clinical significance of this study would be greatly enhanced.
REFERENCES
REPLY:
Our concern regarding the sucking of lemon candy early after radioiodine therapy was radioactivity in the blood and saliva (1). Radioiodine is constantly transported into the salivary glands until excretion is terminated. Because the thyroid gland has been removed from the patients, the radioactivity of 131I in their blood is higher than that in healthy subjects. Additionally, clearance of 131I from the body is considerably slowed in hypothyroidism. My colleagues and I determined radioactivity in the saliva and blood 24 h after 3.7 GBq of 131I therapy had been administered to 4 patients. The saliva-to-blood ratios of the radioactivity ranged from 33 to 51 (mean, 41). Our finding seems compatible with a previous report that the concentration of 131I in the salivary gland is 30–40 times higher than that in the blood (2). Our specific answers to the 4 questions of Drs. Lam and van Isselt are as follows.
First, parasympathetic and sympathetic refluxes mediate blood flow and salivary secretion in salivary glands. Therefore, enhancement of salivary secretion is not always proportionate to the increase in salivary blood flow caused by gustatory stimulation (3). A Doppler ultrasound study showed no significant correlations between salivary secretion and the maximum velocity, minimum velocity, and pulsatility index of the facial artery in the submandibular gland after sucking of a lemon slice (4).
Second, we used mainly methyl prednisolone sodium succinate (125–250 mg/d) as the steroid and ibuprofen (600 mg/d) as the nonsteroidal antiinflammatory drug (NSAID) in treating acute sialoadenitis. Vitamin B12 (mecobalamin, 1,500 μg/d) and zinc (polaprezinc, 225 mg/d) were used to treat taste dysfunction. A combination therapy of steroids, vitamin B12, and zinc was given to a patient in whom both sialoadenitis and taste dysfunction were present. Because none of the drugs was used for prophylactic treatment, they must not have affected the incidence of acute side effects. Of 84 patients with acute sialoadenitis or taste dysfunction in group A, 44 (52%) were treated with steroids or NSAIDs, vitamin B12 and zinc, or combination therapy. Similarly, 50 (81%) of 62 patients with sialoadenitis or taste dysfunction in group B underwent either of the symptomatic treatments. We eventually found 14 patients with xerostomia in group A and 7 in group B. Of these patients, 4 in group A and 2 in group B were not receiving any medications because they did not experience acute side effects. Of the remaining patients, 6 (60%) in group A and 5 (100%) in group B were given either of the medications. Therefore, whether there was a relationship between the incidence of late xerostomia and the incidence of symptomatic treatment is uncertain.
Third, acute side effects were monitored by regular visits from nuclear medicine physicians to patients during hospitalization. At discharge, the patients were instructed to contact our staff anytime they suspected themselves of having sialoadenitis or taste dysfunction and to visit our outpatient clinic. The diagnostic criteria for xerostomia consisted of a visual analog scale and salivary scintigraphy (1). When a marking in the severe-dry-mouth zone on the visual analog scale was associated with a nonfunctioning pattern in all 4 major salivary glands on the time–activity curves, xerostomia was confirmed. All patients underwent salivary scintigraphy at least once within 13 mo after radioiodine therapy. In patients who experienced acute side effects or who were suspected of having xerostomia, scintigraphy was repeated at 3- to 6-mo intervals. Forty-three (41%) of 105 patients in group A and 65 (52%) of 125 in group B underwent scintigraphy twice, and 28 (27%) in group A and 40 (32%) in group B were evaluated more than 3 times within 2 y after radioiodine therapy. The salivary function of the patients was monitored for at least 24 mo after 131I therapy. We still monitor the salivary function of the patients enrolled in the study. We have not encountered additional patients who met our criteria for xerostomia after 25 mo. Thus, 24 mo after radioiodine therapy seems an appropriate endpoint for monitoring salivary function.
Fourth, delaying the sucking of lemon candy for 24 h was based on a hypothesis that the majority of 131I administered should be excreted into the urine and that lemon candy–induced enhancement of blood flow may not enhance irradiation in the salivary gland. To determine optimal timing, a study with a variable start time seems essential (e.g., 6, 12, 24, and 48 h).
Recent studies have proposed novel perspectives on the mechanisms of irradiation damage to the salivary gland. It has been suggested that water secretion is selectively hampered during the first day after a single-dose irradiation without loss of the acinar cells, because of selective radiation damage to the plasma membrane of the secretory cells, disturbing muscarinic receptor–stimulated watery secretion (5). Also, it has been suggested that the sodium iodide symporter is detected mainly in the basolateral membrane of ductal cells and that radioiodine is transported mainly by ductal cells, not by acinar cells (6,7). Continuous stimulation of salivation may increase the radiation exposure of ductal cells, which are more sensitive to irradiated damage than are acinar cells, even if the residual time of 131I in the salivary gland is shortened. We consider that radioprotection may be better achieved by suppressing radioiodine uptake in the salivary glands (8) rather than by stimulating salivation alone.