Radioactive iodine (RAI) has been the core therapy in treating multiple thyroid diseases. It was first used to treat Grave’s disease in 1942 [
Physiologic uptake after RAI is expected in the liver, stomach, oropharynx, nasopharynx, esophagus, and salivary glands [
Risk factors to develop sialadenitis are linked to the radioactive iodine uptake and include age, gender, RAI dose, and preexisting salivary gland disease [
This study aimed to determine the prevalence and risk factors for early (i.e., within 48 hours) sialadenitis in patients receiving RAI in Lebanon, a middle-income country of EMRO.
The study was conducted at the American University of Beirut Medical Center (AUBMC), which is a major referral center to the Lebanese population, as well as citizens of other neighboring countries from the EMRO region, especially Syria and Iraq. Twelve centers provide RAI in Lebanon, out of which seven are active. The load seen at AUBMC constitutes about 18–24% of the country load, which allows for a good reflection on the population studied.
All consecutive patients between 18 and 79 years, admitted to AUBMC to receive RAI for DTC, were included. Patients receiving RAI for hyperthyroidism (toxic nodule, toxic multinodular goiter, and Grave’s disease) and those with salivary gland pathology (stone, tumor, and surgery) were excluded.
This was a retrospective study conducted through medical charts’ revision of patients admitted to receive RAI between January 1, 2012, and December 31, 2015. These patients were admitted for 48 hours in a special isolation room. The following variables were recorded: date of RAI, date of birth, age
Two of the authors conducted data extraction. To ensure homogeneity in data collection and in defining sialadenitis, the other investigator blindly filled ten charts for each investigator, and the concordance rate in detecting sialadenitis and extracting the information in the chart was almost 100%.
In the majority of cases, the authors relied on notes and assessments by endocrinology fellows on call for the diagnosis of sialadenitis, detailing the clinical assessment of patients with symptoms of sialadenitis, including neck swelling and pain. In the absence of a clear assessment note, sialadenitis was deemed present if there were any records of neck swelling, neck pain, dry mouth, or difficulty swallowing within 48 hours of RAI administration. In addition, whether subjects required treatment with nonsteroidal anti-inflammatory drugs or glucocorticoids was noted.
Continuous variables were reported as mean ± (SD), while categorical variables were reported as number (percent). Univariate analysis was performed by independent
Multivariate analysis of potentially significant predictors of sialadenitis was also performed. Predictors included gender, age, lymph node involvement, RAI dose, days off LT4, and WBS result. When confounders were suspected, the analysis was undertaken again with isolation of one of these variables.
AUBMC Institutional Review Board approved the study.
Based on the retrospective review of medical records, the total number of patients admitted to receive RAI from January 2012 to December 2015 was 230. Out of these, 56 patients received iodine for benign thyroid pathology and were therefore excluded. The remaining 174 patients had all undergone total thyroidectomy followed by RAI for histologically confirmed DTC.
Patients admitted for RAI for DTC at our institution had a mean age of 42.8 ± 14.4 years and were predominantly women (71.3%). The patients were mostly Lebanese, but around one-third were from other nationalities, mainly Iraqi and Syrian. The majority had papillary thyroid cancer (93.1%), while follicular cancer accounted for 5.7% only. The DTC was recurrent in 14.9% and 64.5% of the patients had lymph node involvement. Most patients received 100 millicuries (mCi) (63.2%), followed by 50 mCi (23%), while 13.8% received 150 mCi (Table
Characteristics of subjects with and without sialadenitis who received radioactive iodine for differentiated thyroid cancer.
Total | Sialadenitis | |||
---|---|---|---|---|
No | Yes | |||
Gender | 124 (71.3) | 96 (69.1) | 28 (80.0) | 0.20 |
Age, mean (±SD) | 42.8 ± 14.4 | 43.0 ± 14.8 | 41.8 ± 13.1 | 0.65 |
Nationality | ||||
Lebanese | 118 (67.8) | 100 (71.9) | 18 (51.4) | 0.02 |
Non-Lebanese | 56 (32.2) | 39 (28.1) | 17 (48.6) | |
BMI, mean (±SD) | 28.6 ± 6.3 | 28.4 ± 6.4 | 29.2 ± 5.7 | 0.49 |
SBP, mean (±SD) | 125.1 ± 15.8 | 125.3 ± 15.9 | 124.4 ± 15.6 | 0.76 |
DBP, mean (±SD) | 79.4 ± 11.7 | 79.5 ± 11.8 | 79.2 ± 11.6 | 0.92 |
Tumor pathology | ||||
PTC | 162 (93.1) | 130 (93.5) | 32 (91.4) | 0.57 |
FTC | 10 (5.7) | 7 (5.0) | 3 (8.6) | |
Other | 2 (1.1) | 2 (1.4) | 0 (0.0) | |
Recurrence, yes | 26 (14.9) | 23 (16.5) | 3 (8.6) | 0.24 |
Lymph node, yes | 89 (64.5) | 69 (62.7) | 20 (71.4) | 0.39 |
RAI dose | ||||
50 m/Cu | 40 (23.0) | 36 (25.9) | 4 (11.4) | |
100 m/Cu | 110 (63.2) | 84 (60.4) | 26 (74.3) | 0.18 |
150 m/Cu | 24 (13.8) | 19 (13.7) | 5 (14.3) | |
Thyrogen, yes | 7 (4.0) | 6 (4.3) | 1 (2.9) | 1.00 |
Pretherapy TSH (mIU/L), mean (±SD) | 83.9 ± 23.4 | 85.8 ± 21.7 | 76.6 ± 28.4 | 0.04 |
Pretherapy TSH (mIU/L) categorical | ||||
<50 | 18 (11.0) | 12 (9.2) | 6 (17.6) | 0.11 |
50–75 | 27 (16.5) | 19 (14.6) | 8 (23.5) | |
>75 | 119 (72.6) | 99 (76.2) | 20 (58.8) | |
Smoking, yes | 39 (22.5) | 31 (22.3) | 8 (23.5) | 0.88 |
Alcohol, yes | 23 (13.3) | 19 (13.7) | 4 (11.8) | 1.00 |
Dental problems, yes | 3 (2.2) | 3 (2.8) | 0 (0.0) | 1.00 |
Days off LT4, mean (±SD) | 26.8 ± 19.4 | 24.9 ± 16.3 | 33.0 ± 27.9 | 0.63 |
Days after surgery, mean (±SD) | 42.4 ± 38.3 | 44.3 ± 42.0 | 35.1 ± 15.9 | 0.12 |
Days off total, mean (±SD) | 39.7 ± 36.2 | 41.0 ± 39.5 | 34.7 ± 18.5 | 0.18 |
WBS result | ||||
Negative | 19 (15.0) | 17 (17.5) | 2 (6.7) | 0.16 |
Positive | 98 (77.1) | 71 (73.3) | 27 (90.0) | |
Distant | 10 (7.9) | 9 (9.2) | 1 (3.3) | |
Treatment received | ||||
NSAIDs | 20 (11.5) | 3 (2.2) | 17 (48.6) | 0.19 |
Steroids | 5 (2.9) | 2 (1.4) | 3 (8.6) | |
Both | 6 (3.4) | 0 (0.0) | 6 (17.1) |
Of the 174 patients, 35 (20.1%) [95% CI (15–27)] had sialadenitis within the first 48 hours of iodine intake. Out of these, 17 (48.6%) received nonsteroidal anti-inflammatory drugs (NSAIDs) for symptomatic management, 3 (8.6%) were treated with steroids, 6 (17.1%) got dual therapy (NSAIDs and steroids), and 9 (25.7%) were not treated pharmacologically (Table
The association between sialadenitis and the studied parameters is outlined in Table
There were twice as many non-Lebanese subjects who developed salivary gland dysfunction (30.4%), as there were Lebanese (15.3%), with a
Because of the significant association with nationality, the analysis was stratified by nationality and the abovementioned risk factors were studied (Table
Characteristics of subjects who received radioactive iodine for differentiated thyroid cancer, stratified by nationality as Lebanese versus non-Lebanese.
Nationality | |||
---|---|---|---|
Lebanese | Non-Lebanese | ||
Gender | |||
Male | 31 (26.3) | 19 (33.9) | 0.30 |
Female | 87 (73.7) | 37 (66.1) | |
Age, mean (±SD) | 44.2 ± 15.2 | 39.8 ± 12.2 | 0.06 |
BMI, mean (±SD) | 27.4 ± 5.2 | 30.9 ± 7.6 | 0.004 |
SBP, mean (±SD) | 123.5 ± 15.3 | 128.6 ± 16.4 | 0.05 |
DBP, mean (±SD) | 78.1 ± 11.5 | 82.2 ± 11.8 | 0.03 |
Tumor pathology | |||
PTC | 108 (91.5) | 54 (96.4) | 0.42 |
FTC | 8 (6.8) | 2 (3.6) | |
Other | 2 (1.7) | 0 (0.0) | |
Recurrence, yes | 17 (14.4) | 9 (16.1) | 0.77 |
Lymph node, yes | 54 (56.8) | 35 (81.4) | 0.005 |
RAI dose | |||
50 mCi | 30 (25.4) | 10 (17.9) | 0.38 |
100 mCi | 74 (62.7) | 36 (64.3) | |
150 mCi | 14 (11.9) | 10 (17.9) | |
Thyrogen, yes | 4 (3.4) | 3 (5.5) | 0.68 |
Pretherapy TSH (mIU/L) mean (±SD) | 85.0 ± 22.2 | 81.5 ± 25.9 | 0.39 |
Pretherapy TSH (mIU/L)–categorical | |||
<50 | 9 (8.2) | 9 (16.7) | 0.25 |
50–75 | 18 (16.4) | 9 (16.7) | |
>75 | 83 (75.5) | 36 (66.7) | |
Smoking, yes | 32 (27.4) | 7 (12.5) | 0.03 |
Alcohol, yes | 21 (17.8) | 2 (3.6) | 0.01 |
Dental problems, yes | 2 (2.1) | 1 (2.3) | 1.00 |
Days off LT4, mean (±SD) | 20.6 ± 16.1 | 37.1 ± 20.8 | 0.02 |
Days after surgery, mean (±SD) | 40.3 ± 28.1 | 47.5 ± 56.3 | 0.45 |
Days off total, mean (±SD) | 37.2 ± 27.5 | 45.2 ± 50.7 | 0.19 |
WBS result | |||
Negative | 12 (13.0) | 7 (20.0) | 0.31 |
Positive | 71 (77.2) | 27 (77.1) | |
Distant | 9 (9.8) | 1 (2.9) | |
Treatment received | |||
NSAIDs | 11 (9.3) | 9 (16.1) | 0.60 |
Steroids | 3 (2.5) | 2 (3.6) | |
Both | 2 (1.7) | 4 (8.9) |
Independent risk factors for sialadenitis, using multivariate analysis as outlined in Table
Independent predictors for sialadenitis.
Sialadenitis (reference: no) | ||
---|---|---|
Variables | OR (95% CI) | |
Gender | 2.34 (0.92–5.97) | 0.07 |
Nationality | 2.34 (1.07–5.13) | 0.03 |
WBS positive | 3.99 (1.13–14.16) | 0.03 |
Variables included in the model were gender (reference: male); age; nationality (reference: Lebanese); lymph node involvement (reference: no); RAI dose (mCi) (reference: 50 mCi); days off LT4; and WBS result (reference: negative).
Patients admitted to receive RAI at our institution were predominantly women with PTC, and the majority had lymph node involvement. The prevalence of sialadenitis was 20.1% and its independent risk factors were being non-Lebanese and having a positive postiodine WBS. In turn, subjects who were from other nationalities had more lymph node involvement and were kept off LT4 for longer times.
TSH level per se did not clearly stand out as a predictor for sialadenitis. However, TSH levels in our study were found to be higher than the recommended level of 30 mIU/mL or above.
The reported prevalence rate of sialadenitis is highly variable in the literature. For instance, An et al. found that symptomatic late onset sialadenitis occurred at a rate of 10.2% [
Concerning risk factors, although sialadenitis might not be very common with lower, often diagnostic, doses of RAI [
The novelty in the current study is that it targets early sialadenitis predictors, and no such reports have been conducted in the Eastern Mediterranean Region to our knowledge. Most of our patients had pretreatment TSH above 75 mIU/ml, way above the needed target to achieve adequate iodine uptake. The latter fact indicates that most patients are rendered symptomatically hypothyroid for longer than necessary. In fact, a goal TSH of equal or above 30 mIU/L is generally adopted in preparation for RAI therapy or diagnostic testing as per the 2015 American Thyroid Association (ATA) guidelines for thyroid nodules and DTC [
Finally, sialadenitis rates are considerable and the burden of the symptoms affect patients’ quality of life; hence one can identify high-risk patients, look for early signs and symptoms and implement when appropriate preventive measures. The 2008 European Association of Nuclear Medicine (EANM) guideline for 131I therapy recommends sufficient hydration with use of lemon candy, sour candy, or chewing gum in the 24 h following 131I administration to increase salivary flow and to reduce radiation exposure of the salivary glands [
Our study has the following limitations: first, being retrospective by design, data was collected from medical records and medication orders. Mild symptoms could have been neglected and/or not mentioned, and therefore sialadenitis may have been underdiagnosed.
In addition, the diagnosis of sialadenitis was based on subjective reports by the patients. Some symptoms of sialadenitis, such as neck pain, might have alternative etiologies. One example is higher concentrations of radioiodine in the thyroid of patients with relatively large residual thyroid tissue increase the risk of neck pain after treatment, in support of remnant tissue thyroiditis [
In addition, this study only assessed early sialadenitis and results would not apply to the long-term effects. Although early sialadenitis may be transient in nature, it is still considered bothersome by many patients and can persist in some cases causing chronic symptoms.
Finally, it is worth noting that though dosimetry would be very useful to estimate RAI doses to the salivary glands, at our institution, it is not included in the RAI therapy protocol. As a consequence of the retrospective design of our study, it was not possible to implement it beforehand, and we were consequently unable to assess exact doses to the salivary glands and further elucidate the RAI dose-dependence of the risk of sialadenitis, as previously discussed.
Around 20% of patients admitted to receive RAI for DTC develop sialadenitis at our institution, which is comparable to rates reported in the literature. Having a positive posttreatment whole-body scan, more lymph nodes involvement, and longer LT4 withdrawal time are risk factors for salivary glands dysfunction. Therefore, we emphasize on giving special attention to patients with more aggressive disease. We also recommend avoiding rendering patients excessively hypothyroid as measures to prevent early development of sialadenitis.
The data generated in the study are included in this article. The database is available upon request.
The authors have no conflicts of interest.
Ruba Riachy was responsible for cowriting the protocol, data collection, analysis, and manuscript. Nisrine Ghazal was responsible for data collection, analysis, and manuscript. Mohamad B Haidar was responsible for reviewing the nuclear imaging and writing sections of the manuscript. Ahmad El-Amine was responsible for addressing reviewers’ comments, updating the literature review, and editing pertinent changes in the manuscript. Mona P Nasrallah was responsible for the idea and cowriting of the protocol, data analysis, and manuscript.
The investigators are grateful to Dr. Hani Tamim for providing guidance and insight into the statistical analysis and to Ms. Maha Makki for conducting the analysis.