Salivary Gland Scintigraphy: The Use of Semiquantitative Analysis for Uptake and Clearance

MATERIALS AND METHODS

Imaging Technique

Imaging was performed using a mobile small-field-of-view scintillation camera (Starcam 300; General Electric Medical Systems) equipped with a low-energy all-purpose parallel-hole collimator. The patient was supine, and the camera was positioned for an anterior head-and-neck projection. Dynamic imaging was performed in a 64 × 64 pixel matrix at 30 s per frame starting immediately after a bolus intravenous injection of 185 MBq (5 mCi) ^99mTc-sodium pertechnetate. Imaging continued for 30 min after injection. At 15 min after injection, each patient drank the freshly squeezed juice of half a lemon in water, using a straw and without moving, while imaging was continued. The study was reviewed to check for patient motion and for suitability for further analysis (Fig. 1). Afterward, standard imaging of the thyroid proceeded as set out in our protocol manual, using the same camera, to address the question for which the patient was referred.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Reframed dynamic images (1 min per frame) show parotid and submandibular glands. Washout after lemon juice is seen halfway through study (3rd frame of 3rd row).

Analysis

Regions of interest were drawn on the dynamic images of the parotid glands (right and left) and submandibular glands (right and left) (Fig. 2). Time–activity curves were generated for each region. These were analyzed as follows: First, the initial slope of the time–activity curve for each salivary gland was calculated from a linear fit of the rapidly rising part of the curve 0–240 s after injection of the radiotracer. This value was considered the uptake rate parameter of the salivary gland and was expressed as count rate per second (cps/s). Second, the response of the salivary gland to lemon juice was noted on the time–activity curves as a sharp decline of the activity in the gland with subsequent slow buildup. The washout fraction in each salivary gland was defined as the amount of cleared radioactivity from the gland divided by the postclearance activity (nonmobilizable uptake). A parametric image of the washout fraction in each gland was generated by subtraction of the image immediately after washout from the image at peak activity, followed by division of the result by the former image (peak − post/post).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

(A) Regions of interest drawn around individual salivary glands. (B) Time–activity curves of parotid and submandibular glands. (C) Linear curve fit for right parotid curve from 0 to 240 s. (D) Parametric washout fraction image. Ct = count; Subm = submandibular.

The parotid and submandibular glands and the right and left sides were compared for these parameters using the paired t test. A P value < 0.05 was considered statistically significant.

RESULTS

The time–activity curves of the individual salivary glands showed an early fast-rising uptake followed by a slow-rising component to nearly a plateau in all glands (Fig. 2).

In Table 1, the results of the calculated parameters of salivary gland function (i.e., uptake rate and washout fraction for the parotid and submandibular glands) are exhibited.

The average value for uptake rate was 0.10 ± 0.09 cps/s. The uptake rates were similar for all salivary glands and did not significantly differ between parotid and submandibular glands or right and left sides. The P values for the comparisons using the paired t test were as follows: P = 0.17 for parotid versus submandibular, P = 0.28 for right versus left parotid, and P = 0.81 for right versus left submandibular.

The washout fraction showed similar values for the right and left parotid glands (P = 0.21) and for the right and left submandibular glands (P = 0.09). However, a significant difference was noted between the parotid and submandibular glands (P = 0.005), with the parotid glands exhibiting a higher washout fraction (1.40 ± 1.60) than the submandibular glands (0.77 ± 0.41). A parametric image of the washout fraction of the salivary glands is shown in Figure 2.

Of the 2 patients with suspected salivary gland pathology, the first showed normal scan findings and parameters within the range of the controls. In contrast, the second showed a markedly depressed washout fraction (parotid, 0.4; submandibular, 0.15), whereas the uptake rate (0.07 in this patient) remained within the average range shown above.

DISCUSSION

In this study, a semiquantitative method for analysis of ^99mTc-sodium pertechnetate salivary scintigraphy was used to assess the function of, as well as image, the parotid and submandibular glands. Semiquantitation was performed using a standard dynamic imaging protocol that included administration of a sialogogue halfway through the dynamic study. Two parameters of function were calculated, namely uptake rate and washout fraction, providing information that could be useful to categorize the effects of insults to the salivary glands.

To establish the method of quantitation and provide guidelines for its use, including expected values and normal range, studies of 21 healthy volunteers were performed. These subjects were selected from a pool of patients referred for thyroid scintigraphy who did not have a history or clinical evidence of salivary gland dysfunction. The possibility that thyroid disease could affect the results of the quantitation method was considered minimal, since only relative quantitation was performed and thyroid uptake would usually be a small fraction of the total injected dose (0.7%–3.0%) (13).

The uptake rate parameter was derived from the early rising part of the individual salivary gland time–activity curve. It represented a simple measure of how well the gland could extract the radiotracer from the circulation, thus avoiding the need for absolute quantitation of uptake and ignoring the subsequent fate of the radiotracer in the gland (dealt with using the second parameter). The linear fit of the time–activity curves from 0 to 240 s was chosen because progressive accumulation, approximating a straight line, was seen in the salivary glands during that time. A diseased salivary gland would be expected to show a slowly rising curve with a resulting decreased slope, compared with normal.

The washout fraction parameter would be most useful to track subtle changes in salivary gland function, providing a marker for follow-up with time. The parameter calculation was based on the assumption that once the plateau of the time–activity curves was reached (usually between 6 and 10 min after injection), radioactivity in the salivary gland was located in 2 separate pools: a mobilizable one in the excreted saliva and a nonmobilizable one remaining in the parenchymal cells. The 2 pools were separated using intervention with a sialogogue at 15 min after injection. The ratio of the mobilizable radioactivity to the remaining activity gave a measure of the specific function, denoted as the washout fraction, of the individual salivary gland. Our observation that the washout fraction of the parotid glands was higher than that of the submandibular glands agreed with other reports in the literature and might have had to do with the type of saliva produced (parotid, serous, vs. submandibular, mucous) (14). This parameter helped us recognize and confirm the disturbance in the second patient and provided a baseline for future follow-up.

A review of previous reports on quantitation of salivary gland function, comparing the 2 parameters introduced in this study with others (3,4,10,12,15,16), showed that most functional indices were derived from individual salivary gland time–activity curves generated from a dynamic study similar to that outlined here. In certain cases, the functional indicators were somewhat difficult to define and would vary with the amount of injected radioactivity, requiring absolute quantitation of the uptake for reproducibility (12). The semiquantitative reporting of the uptake and clearance data in the salivary time–activity curve (i.e., the slope of the curve and the ratios of counts), as presented in this study, would make the data easier to reproduce using different imaging equipment or amounts of administered radioactivity.

The clinical experience with the 2 patients suspected of salivary gland disease in this study encourages the use of the semiquantitative method when salivary function impairment is suspected, the rationale being that function is assessed from 2 different aspects, for changes occurring in disease that are related to separate impairment of blood flow and secretory stimulation (8). The application of the semiquantitative method to a larger number of patients with salivary dysfunction is required, however, to establish the usefulness of the method in clinical practice.

CONCLUSION

Salivary scintigraphy using the described semiquantitative method including the uptake rate and washout fraction parameters would provide an objective means of diagnosis and a reproducible tool for follow-up of salivary function impairment.