Incidence and Risk Factors of Contrast-Induced Sialadenitis after Therapeutic Neuroendovascular Procedures

DISCUSSION

On the basis of a previous study, although CIS was not clearly defined, the documented frequency of adverse reactions to iodide contrast agents deemed appropriate for CIS was approximately 1%–2%.² However, given that not all CIS-associated symptoms were included and considering that a meta-analysis analyzing CIS-related studies from 1956–2018 found only 77 cases,¹ the reported incidence of CIS is likely lower than the actual occurrence rate. In addition, in a 2-month prospective observation study,¹⁴ CIS was confirmed in 4 patients exposed to iodide contrast agents. In the present study, the incidence of CIS after therapeutic neuroendovascular procedures was 4.2%. In most reported cases of CIS, the condition occurred after intravenous procedures, likely due to the indirect effect of contrast agents on the salivary glands after systematic distribution. However, in cases of therapeutic neurovascular procedures, as in this study, the contrast agent is believed to directly affect the salivary glands through the arteries, leading to CIS. Therefore, we hypothesized that CIS occurrence is not rare, particularly after therapeutic neurovascular procedures.

The mechanism underlying CIS remains to be fully elucidated. One suspected mechanism involves the toxic accumulation of iodide within the salivary ductal mucosa, resulting in edema, duct obstruction, and enlargement of the salivary glands.¹⁵ Approximately 98% of the iodide contrast is eliminated by the kidneys, and only 2% is excreted by other organs including the salivary glands, lacrimal glands, hepatobiliary system, and sweat glands.¹⁶ Consequently, some researchers hypothesized that the risk of CIS increases in individuals with impaired renal function because of reduced elimination of iodide contrast agents,¹⁷ leading to elevated serum iodide levels, occasionally measuring up to 100 times the plasma level (>10 mg/100 mL).¹³ However, considering the similarity of plasma iodide levels in asymptomatic patients, there may be an inherent idiosyncratic component to CIS.¹⁸ This is consistent with the findings of the present study, as renal function was not significantly associated with CIS.

Based on this toxic accumulation mechanism, CIS may occur more frequently in direct contrast administration via the ECA, which supplies blood to the salivary gland, or with the administration of high volume of contrast. In the present study, CIS occurrence was associated with the guiding catheter tip being located within the ECA. According to previous studies, the submandibular gland was the most affected solitary gland in patients receiving intravenous contrast administration (48%–100%).^1,14 The submandibular gland accounts for the highest proportion of saliva excretion (69%), which may elucidate its preferential involvement in the reported cases.¹ On the other hand, in the present study, which specifically focused on therapeutic neuroendovascular procedures, the most common site of occurrence was the parotid gland, which was vascularized via the superficial temporal, maxillary, and transverse facial arteries. Although the submandibular gland is also supplied by the ECA, such as the submental artery (branch of facial artery) and sublingual artery (branch of lingual artery), guiding catheter tips located within the ECA were often positioned more distally relative to the lingual artery, resulting in more blood supply from ECA branches located even more distally than the distal tip end. Therefore, the parotid gland, which directly receives more contrast agents during procedures, may be more prone to developing CIS after therapeutic neuroendovascular procedures. Multivariate logistic regression analyses showed that the volume of contrast administered was also statistically significantly associated with CIS occurrence with a cutoff value of 205 mL However, a relatively low OR was observed (1.007 [1.003–1.011]). Similar to previous reports, we hypothesized that there is a greater association between CIS occurrence and the directly supplied toxicity of contrast agents on the salivary glands via the ECA rather than an increase in blood contrast concentration because of a higher volume of contrast agent being administered. In line with this, contrast type (nonionic monomer versus nonionic dimer) also showed no association with CIS occurrence in this study.

In this study, female sex was identified as a risk factor of CIS occurrence. On the other hand, a previous meta-analysis found that men had a slightly higher occurrence rate of CIS compared with women (61% versus 38%), but there was no significant predilection for sex.¹ Both our study and the previous meta-analysis had a limited number of patients (33 and 77 cases, respectively) with CIS; therefore, there may be statistical differences. However, because a previous study reported that adverse reactions to iodide contrast agents are more common in women (OR = 2.40),¹⁹ female sex was identified as a risk factor of CIS occurrence in the present study.

The diagnosis of CIS is primarily clinical but can be supplemented with imaging modalities.⁶ Ultrasound (US), CT, and MR imaging are the most commonly used diagnostic tools for suspected cases of CIS. US findings typically reveal a notable swelling of salivary glands with hypoechoic septa, heightened vascularity, and dilated ducts with no indications of sialolithiasis or infection.^1,6 CT and MRI examinations reveal gland enlargement without any evidence of fat stranding surrounding the glands or inflammatory changes in the surrounding fat and subcutaneous tissues.¹ Because the patients in the present study had no symptoms associated with CIS before the procedures, in the event of CIS-related symptoms emerging after the procedure, routine imaging tests were not conducted because of the low likelihood of causes other than CIS.

The natural history of CIS is typically benign and self-limited. In most cases, symptoms occurred in a median of 16 hours (ranging from immediate onset to 5 days) after the administration of contrast agent, and symptoms generally resolve over a median duration of 3 days (ranging from immediate resolution to 72 days).^1,5,14 Compared with these results, our study showed a slightly faster onset time of symptoms, with an average of 9.89 hours, and similar timing of symptom improvement, with a mean duration of 4.06 days. The relatively faster onset of symptoms in this study may be attributed to the direct impact of the contrast agent on the salivary glands through the ECA.

While there is currently no standardized treatment protocol for CIS, the management of CIS typically focuses on symptomatic relief, including hydration and administration of NSAIDs, antibiotics, and oral anti-inflammatory drugs. Although therapeutic interventions including medication can help alleviate symptoms, no statistically significant difference was observed in the time to symptom recovery between individuals who received therapeutic intervention and those who did not.^1,6,14 In line with this finding, this condition is not caused by allergic reactions, and pretreatment is ineffective in preventing recurrent CIS episodes.¹ All patients in the present study experienced resolution of symptoms with conservative treatment with medications prescribed according to the advice of an otolaryngologist.

Our study had several limitations. First, the study was conducted retrospectively at a single institution and included only 33 patients with CIS, potentially leading to limited statistical power. Second, in our quantitative assessment of contrast volume as a risk factor of CIS, the optimal cutoff value of contrast volume being administered was 205 mL. Although the specificity (0.87) appears acceptable, the sensitivity (0.47) was low. This is likely attributed to the limited frequency of CIS occurrences, resulting in insufficient data, and may have yielded a false-negative prediction. Third, CIS was defined solely on the basis of clinical symptoms without radiologic evaluation in this study. Consequently, some cases that were not suitable for CIS may have been included, and those with ambiguous clinical presentations of CIS may not have been included. Finally, in this study, undiluted contrast agents were used, and only therapeutic neuroendovascular procedures involving relatively high amounts of contrast agents were included. Consequently, the incidence of CIS may have been higher than typically reported. Larger-scale prospective multicenter studies are needed to verify the findings of the present study.