Article Text

A pilot study of the salivary scintigraphy diagnostic performance in a Spanish population with Sjögren's syndrome diagnosed by the European criteria
  1. Division of Nuclear Medicine, Hospital Universitario Miguel Servet, Zaragoza, Spain
  2. Member of the Spanish Society of Nuclear Medicine
  3. Department of Internal Medicine
  4. Hospital Universitario Miguel Servet
  1. raulmartinez{at}

Statistics from

The European classification criteria (ECC) include salivary scintigraphy (SSC) for diagnosing Sjögren's syndrome (SS).1 The performance of this test has been established without considering the ECC for either confirming or excluding SS.1-6 This pilot study aimed at evaluating the performance of the qualitative reading of SSC, performing a clinical measurement of SSC, and establishing the most discriminatory scintigraphic parameters for diagnosing SS in a Spanish population fulfilling the ECC.

This cross sectional study included consecutive patients referred for SCC: 15 healthy volunteers (eight women, seven men; mean (SD) age 50.6 (17.5)), 16 patients with SS based on the ECC (15 women, one man; age 58.1 (10.4)), and 15 xerostomic patients who did not fulfil the ECC, as controls (seven women, eight men; age 53.3 (19.0); AIDS (n=3), chronic parotitis (n=2), sarcoidosis (n=2), or taking drugs that cause dryness (n=8)).

Patients underwent a sufficient number of tests included in the ECC set, if not all, to confirm or exclude SS. Scintigraphy was performed in patients and volunteers: image acquisition started two minutes after injection of technetium-99m pertechnetate, 60 second frames were continuously obtained for 16 minutes, and lemon juice given orally at 9.5 minutes. Data of the ECC set, except for scintigraphy, drug history, extraglandular manifestations of SS, associated connective tissue disease, and history of exclusion criteria, were collected by a questionnaire and from the medical records. SS was diagnosed according to the ECC.

Qualitative reading comprised visual evaluation of tracer accumulation and excretion by the parotid and submandibular glands in either the scan or the time-activity curves. Scintigraphy was positive for SS if both a sicca syndrome pattern7 and a curve M, F, or S8 were detected in at least two glands; it was negative if either positive in only one gland or normal (N curve8and normal pattern on the scan7 in all glands; fig 1). Qualitative reading showed an excellent normalcy fraction (100%) and high sensitivity (87.5%) in detecting SS, but specificity was lower (66.7%). Sensitivity and specificity were as previously reported.1-6 Some authors obtained better specificity by including healthy people as controls.5 Predictive values (positive predictive value 73.7%, negative predictive value 83.3%) differed from those of other studies1-6; they are influenced by the prevalence of SS.

Figure 1

Dynamic scan and time-activity curves of a normal scintigraphic study. Both parotid and submandibular glands show regular size and morphology. Tracer uptake by the salivary glands is uniform and similar to thyroidal uptake, and fast prompt excretion of the tracer into the oral cavity follows excretory stimulus. The time-activity curves show a phase of increased counts that corresponds to active radioisotope uptake: the parotid curves present a marked increase (RP = right parotid; LP = left parotid), higher than that of the submandibular glands (RS = right submandibular; LS = left submandibular). Stimulation with lemon juice is instantly followed by profuse excretion by both the parotid and submandibular glands, as note by a sharp fall in the curves of the glands and a spike in the oral time-activity curve (OC = oral cavity).

Clinical measurement was made on each gland curve measuring gland size (area), tracer accumulation, and stimulated excretion; these scintigraphic parameters were tabulated for all, parotid, submandibular, right and left glands. Clinical measurement in normal subjects generated a normal database that could be used to evaluate the inter- and intrapersonal variation of gland area and function in patients. ROC curves were plotted from parameters in patients, and optimal thresholds computed.

Optimal thresholds of area agreed well with the qualitative reading for diagnosing SS, and increased accuracy as a result of improved specificity with loss in sensitivity or normalcy fraction. They highly discriminated between patients with SS and controls, because areas significantly and exclusively decreased in patients with SS compared with normal subjects and controls. The amount of acinar mass lost by the gland, is reflected by the smaller size of the gland on SSC. As patients with SS, normal subjects, and controls had similar ages, the gland size reduction may be considered as a characteristic scintigraphic abnormality of SS. However, any conclusion must await a further extensive study.

Optimal thresholds of gland function were insufficiently accurate for diagnosis of SS, as they failed to discriminate between SS and controls, probably owing to different degrees of dysfunction in the patients with SS and the small sample size. We noted that tracer accumulation by the parotid gland decreased in sicca syndrome, but failed to show similar changes in submandibular uptake.9 ,10 A decreased excretion fraction in all glands compared with normal is characteristic of sicca syndrome and not exclusive to SS as has been described.9 ,10


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.