1990, The British Journal of Radiology, 63, 190-196

Salivary gland scintigraphy—a suitable substitute for sialography? By *W. J. Pilbrow, MB ChB, DMRD, FRCP, tS. M. Brownless, MA, MSC, " J . I. Cawood, BDS, RCSEd, *A. Dynes, DCR, DRI, \ J . D. Hughes, BSc, MB ChB, DMRD and tH. R. Stockdale, BSc, PhD, MIPSM 'Departments of Radiology and Nuclear Medicine, Countess of Chester Hospital, Liverpool Road, Chester, **Maxillo-Facial Unit, Chester Royal Infirmary, and tDepartment of Medical Physics, Royal Liverpool Hospital, Prescot Street, Liverpool L69 3BX (Received June 1989 and in revised form October 1989) Abstract. By comparing 27 patients who had both scintigraphy and sialography in the assessment of salivary gland disease, scintigraphy has been shown to correlate well with abnormal sialograms. It is suggested that scintigraphy could become the initial screening procedure in the assessment of salivary gland disease. A normal scintiscan is unlikely to miss significant pathology (as demonstrated by sialography), but sialography must always be performed if there is a suspicion of duct obstruction on scintigraphy. Patients suspected of focal salivary gland pathology such as tumour have not been investigated. The series documents the findings in patients who presented with facial pain, swelling or xerostomia suggesting sialadenitis, duct occlusion or Sjogren's syndrome.

The primary radiological investigation of a salivary gland is sialography. It is invasive and even with excellent technique the appearances bear only a poor relationship to the functional or secretory aspect of the gland. Any changes present are not necessarily recent, and the technique must therefore be viewed with care even in a symptomatic patient. It is rare for sialography to be of "special interest" in a non-specialized centre and is frequently performed by different operators often at junior level. Salivary gland scintigraphy has been available for a number of years and, with a technique using time/activity curves, is an excellent assessment of gland accumulation and secretion in Sjogren's syndrome (Sugihara & Yoshimura, 1988) and "problem" sialograms (Van den Akker & Busemann-Sokole, 1985). It has a lower morbidity than sialography, assesses all four major salivary glands at one sitting, and is not so operator dependent. The authors describe the use of scintigraphy in the assessment of salivary function in 36 patients (27 of whom had sialography also) who presented with a variety of features, mostly salivary gland swelling, pain or xerostomia. Focal pathology is excluded from the series. By comparing sialography and scintigraphy, an evaluation is made whether scintigraphy might offer advantages over sialography as the primary screening procedure. Materials and methods

Of 36 patients who were examined, all but one were referred from a maxillofacial surgeon (Case 3 was referred from a general physician). Some 42 scans were performed, with six patients scanned twice; 25 were female and 11 male (average age 44 years, range 9-70 years). Most patients were scanned erect with the nose and chin touching the collimator face. In this Address correspondence to: Dr William J. Pilbrow, Consultant Radiologist, Radiology Department, Chester Royal Infirmary, Nicholas Street, Chester CHI 2AZ. 190

position 200 MBq 99mTc pertechnetate was injected intravenously and sequential scintigraphy commenced at the same time. Recently, the dose has been reduced to 100MBq 99m Tc. Occasionally, patients were scanned supine. Dynamic images were obtained and recorded every minute over a 20 min period. Time/activity curves were recorded on film. The patients remained stationary during this period and at 10 min they were given 10 ml of Carbex solution (Ferring Pharmaceuticals) by mouth through a straw. Carbex is an excellent secretagogue and each 10 ml dose contains l g of citric acid. The scintigraphic technique was standard for all the patients, and was not modified according to differing presenting features. Sialography was performed in a standard manner with plain radiographs obtained prior to the cannulation of the appropriate duct and introduction of Lipiodol Ultra-fluid (May and Baker). All but six of the sialograms were performed by one experienced radiologist who has regularly performed six to eight sialograms per month over the last 15 years. Every scintiscan performed since the start of the service (1985) is included in the paper. Sialography preceded scintigraphy in all cases but one (no. 22), and the time interval between the two examinations is given in brackets in Tables II, III and IV. In eight of the 27 patients the two were performed within 20 days of each other, 11 within 40 days, and 20 within 14 weeks. The result of the sialogram was known at the time of reporting the scintiscan, and one observer (different from the sialographer) reported all the scintiscans. The method of reviewing the scintiscan was essentially unchanged throughout the series. The dynamic scan was viewed and the trapping, accumulation and secretion of each of the four main salivary glands assessed. The time/activity curves were correlated with the dynamic scan and in no instance was there marked disagreement between the two sets of data. The final assessment of each of the four glands was based on the curves as presented by Mita et al (1981) (Fig. 1), but not all cases fitted conveniently The British Journal of Radiology, March 1990

Salivary gland scintigraphy

Table I. Scintigraphy alone (nine patients) Age

Presenting features

Scintigraphic findings

Correlation with clinical features

1 Female 2 Female

44 70

Xerostomia, xerophthalmia Xerostomia

Poor Fair

3 Female

70

Xerostomia, xerophthalmia

4 Female

15

5 Female

64

Prior to removal of submandibular glands for persistent drooling Xerostomia

Normal Poor accumulation; left parotid gland and left submandibular glands Poor accumulation in three glands Poor accumulation in right parotid gland

Good

6 Female

27

7 Female

59

8 Female

28

?Block in right submandibular duct Rheumatoid arthritis, xerophthalmia Left parotitis

9 Male

62

Xerostomia, xerophthalmia

Poor accumulation in all four glands Poor accumulation/secretion in right submandibular gland Poor accumulation/secretion in both submandibular glands Poor accumulation in left parotid gland Poor accumulation in all four glands

Sex

into a particular curve. Patients who attended for a second follow-up scintiscan were noted to have slightly differing curves overall compared with the original, presumably representing a different physiological state at the time of the second scan; for instance a more dehydrated state would result in less production of saliva and a correspondingly poorer trapping accumulation and secretion by all four glands. However, the criteria for determining normality or abnormality was by comparison with the paired gland rather than with any absolute appearance. The dynamic scan excluded artefact due to rotation of the head, by confirming that each of the glands lay equally from the midline structures, and excluded a partially tissued injection by comparing the salivary gland activity with the adjacent

Good

Good Fair Good Good

thyroid tissue, important in the group I patients in whom a Sjogren's appearance could be simulated by an inadequate blood level of the radiopharmaceutical.

Results The 36 patients were split into four groups as follows: (1) Scintigraphy alone—nine patients (Table I). (2) 'Failed' sialograms—five patients (Table II). (3) Good correlation between the two techniques in depicting abnormality—nine patients (Table III). (4) Non-correlation between the two techniques in possibly abnormal glands — ten patients (Table IV).

Table II. Failed sialograms (five patients) Sex

Age

10 Male (two scans)

11 Female

50

12 Female

16

13 Male (two scans)

40

14 Female

59

Presenting features"

Scintigraphy

Small ectopic opening on cheek, draining saliva; ?state of right parotid gland (74 days)

All four glands normal; second scan after repositioning of duct: better accumulation pattern but secretion phase poorer Good accumulation; poor secretion No accumulation; no secretion

Right submandibular gland sialadenitis (4 days) Left submandibular gland sialadenitis (4 days) Intermittent obstruction left parotid duct (7 days) Induration of left submandibular gland area — previous tumour (pleomorphic adenoma) removed in 1975 (2 days)

Poor accumulation; no secretion; second scan: no change All four glands normal

"Interval between sialogram and scintiscan in brackets. Vol. 63, No. 747

191

W. J. Pilbrow et al

Table III. Good correlation between scintiscan/sialogram (nine patients) Sex

15 Male (two scans) 16 Male (two scans)

Scintiscan"

Sialogram

Age

Clinical

61

Left submandibular gland Good accumulation; poor Large calculus, sialadenitis; second scan—improvement in secretion calculus secretion (76 days) following removal of duct calculus Good accumulation; poor Dilated duct, calculus; second scan—improvement in secretion Obstruction left secretion (40 days) following removal of duct calculus submandibular duct Left parotitis Good accumulation; poor Dilated duct, sialadenitis; second scan—improvement in secretion secretion (80 days) following removal of duct calculus Good accumulation; poor Right submandibular Sialadenitis gland sialadenitis secretion (90 days) Right parotitis No accumulation; no Dilated duct, calculus secretion (90 days) Left parotitis Good accumulation; no secretion (not known) Sialadenitis Left submandibular gland Poor accumulation; poor Calculus, sialadenitis sialadenitis secretion (14 days) ?Block in right Poor accumulation; good Sialadenitis submandibular duct secretion (6 days) ?Block in right Good accumulation; poor Occluded duct submandibular gland secretion (40 days)

44

60 17 Male (two scans, see Fig', 3)

18 Female

64

19 Female (see Fig. 2) 20 Female

56 33

21 Female

13

22 Male

50

23 Male

50

"Interval between scintiscan and sialogram in brackets.

Table IV. Poor correlation between scintiscan and sialogram (ten patients) Age

Clinical

Scintiscan"

Sialogram

24 Male

33

Normal (4 months)

Dilated duct sialadenitis

25 Female

30

Right submandibular gland sialadenitis Right parotid gland pain

26 Female (see Fig. 4)

30

27 Female 28 Female

40 58

29 Female

35

30 Female

55

31 Female (two scans)

40

32 Female

40

33 Female

30

Sex

Normal (11 months)

Dilated duct and intraglandular duct system Right parotid gland Normal right parotid Narrowed duct system and extraswelling gland; accumulation is vasation into gland tissue; ? Sjogren's normal in other glands syndrome also (11 days) Right parotitis Normal (22 days) Dilated intraglandular ducts, sialadenitis Indurated left Poor accumulation; good Normal submandibular gland secretion (62 days) Left submandibular gland Normal (55 days) ?Duct occlusion; ?technical pain Left submandibular gland Poor accumulation; good Normal sialadenitis secretion left submandibular gland (80 days) Right parotitis Right parotid gland Normal showed poor secretion; second scan—slight improvement in secretion (7 days) Left parotid gland Poor accumulation in left Normal swelling parotid gland; poor secretion; ? Sjogrens syndrome (11 months) Right parotitis Poor accumulation by all Normal four glands; ? Sjogren's syndrome (80 days)

"Interval between scintiscan and sialogram in brackets. 192

The British Journal of Radiology, March 1990

Salivary gland scintigraphy

M

Figure 1. Curve N represents a normal trapping, accumulation and secretion pattern with good response to the secretagogue. Curves M, F and S represent a progressively poorer facility by the gland to trap, accumulate and secrete the radiopharmaceutical. V , secretory stimulation.

The three remaining patients had normal scintigraphy and sialography and are not considered further, since the combination of normal scintiscan and sialogram strongly suggests the presenting features are unrelated to salivary gland disease. Discussion

Any comparison between two imaging techniques usually depends upon one of them being a "gold standard" and the other is related to it in a true-positive, true-negative, false-positive, false-negative fashion. Neither sialography nor scintigraphy is the "gold standard", and in fact they indicate different aspects of salivary gland anatomy and function—as indicated by sialographic visualization of the duct system and the accumulation and secretion of a radiopharmaceutical as shown by scintigraphy. However, the issue is not whether one of the studies shows pathology or normality to a degree of accuracy that invalidates the use of the other, but whether scintigraphy should replace sialography as the primary screening technique. Mishkin (1981) has described the mechanism by which the gland traps, accumulates and secretes the isotope. The intralobular ductule cells have a capacity for trapping and concentrating group VII anions, and these cells concentrate iodine and pertechnetate. These ions, mucus, proteins and enzymes are released as a fluid into the secretory ducts. This is mediated in part by the myoepithelial cells, but it may well be that other Vol. 63, No. 747

factors are also operational in producing the secretory response (Garrett & Emmelin, 1979). It is the lymphocytic periductal infiltrate and duct cell hyperplasia of Sjogren's syndrome that narrows the ducts and produces glandular atrophy with subsequent poor trapping, accumulation and secretion on scintigraphy. The various types of time/activity curve seen in normal and abnormal states are shown in Fig. 1. Basically, the transition from north to south reflects a diminishing functional and secretory capacity by the affected gland (Mita et al, 1981). Scott et al (1984) have demonstrated by animal experimentation that pertechnetate uptake is determined by the mass of acinar rather than solely ductal tissue in the gland, and that scintigraphy is only sensitive to differences exceeding 25% of the gland mass. Sugihara and Yoshimura (1988) mention four cases which showed an F or S type curve and underwent histological examination following removal; all proved to be considerably damaged, suggesting that whatever the cause, if the glandular tissue is destroyed beyond a certain point the time/activity curves tend to F or S type. This article also showed good correlation between sialography and scintigraphy in Sjogren's syndrome and it seems unnecessary to perform sialography on a patient suspected of Sjogren's syndrome in whom the scintiscan demonstrates a typical F or S type curve affecting two or more glands. Nine patients in the series had scintigraphy alone without sialography, and Table I assesses the correlation between clinical and scintigraphic findings. Six of the nine patients were referred with features suggesting Sjogren's syndrome, usually xerostomia, of whom three had good scintigraphic correlation and two had fair. Not all patients with xerostomia have Sjogren's syndrome and some patients with only mild xerostomic salivary disturbance have markedly abnormal scintigraphic studies, indicating the sensitivity of the technique (Fox et al, 1985). A number of patients in the present series had abnormal scintigraphy of clinically asymptomatic glands, possibly indicating subclinical Sjogren's syndrome (Case 33). Two cases in Table I had a symptomatic single gland problem. Scintigraphy correlated well with the clinical findings in both patients. The final case in this group demonstrates the use of scintigraphy to assess the functional status of the parotid glands prior to removal of the submandibular glands in a patient with cerebral palsy who had persistent drooling despite redirecting the parotid ducts. This functional information can only be provided easily and accurately by scintigraphy. Five of the 27 patients in whom it was attempted had unsuccessful sialography, with a failure to cannulate the duct or inadequate filling. This usually indicates duct pathology, and in three patients abnormal scintigraphy confirmed this with poor or no secretion of the isotope. However, another patient had a normal scintiscan, suggesting that the failure to perform a sialogram was purely technical. A 9-year-old boy with an ectopic duct orifice opening onto his cheek was examined by scintigraphy pre- and post-repositioning of the parotid duct 193

W. J. Pilbrow et al

ft?

(b) Figure 2. Owe 19, see Table III. (a) The right parotid sialogram demonstrates a duct calculus with marked proximal dilatation of the intraglandular duct segments, (b) The corresponding scintiscan curve (2) reveals no trapping, accumulation or secretion of the isotope, indicating a severely damaged gland (F type curve).

opening. Sialography via the ectopic duct orifice was cely valid since each test provides different information. unsuccessful owing to a poor seal at the opening on the If scintigraphy does not miss any significant pathology cheek. This group of five patients accounted for 18% of (as shown on sialography) and adds further informaall the sialograms attempted and without scintigraphy tion, then it should be considered as the primary no functional or secretory information of the glands screening procedure since it has a lower morbidity than would have been obtained. sialography and will provide functional and secretory All the patients in Table III had an abnormal sialo- information on all four glands. The ten patients in this gram with a calculus and/or sialadenitis with or without group have the longest interval between scintigraphy duct occlusion. All the patients with occluded ducts had and sialography compared with other groups, and this an abnormal scintiscan, mostly in the secretion phase, might account for some discrepancy. However, salivary but only two had accumulation impairment. Scinti- gland disease is unlikely to alter markedly, once estabgraphy detected all duct occlusions and gave valuable lished. Since scintigraphy correlated well with the sialoinformation regarding function (Figs 2 and 3). If the graphic changes of duct occlusion in Table III, the gland is poorly functioning or non-functioning on scin- normal scintiscan in Case 29 is considered to be the true tigraphy and is a source of infection the surgeon may normal. Clinical follow-up over a year has failed to decide to remove it. Six patients had sialadenitis demon- detect any developing submandibular gland pathology. strated on sialography of whom three had a duct occlu- Cases 24 and 27 represent false-negative scintigraphy. sion problem. On scintigraphy four had a good Both patients had a clinical history suggesting sialadeniaccumulation phase but only one had a good secretion tis, confirmed on sialography. However, scintigraphy phase. Previous papers have shown that sialadenitis showed a normal trapping and accumulation pattern produces an increase in the accumulation phase (van but confirmed the absence of any duct obstruction. den Akker and Busemann-Sokole, 1985), but this is not Case 25 is considered a false-positive sialogram. The confirmed in the present series. Perhaps our relatively patient's facial pain resolved following treatment with a short accumulation phase of 10 min has not allowed this bite-raising appliance, suggesting a temporomandibular increase to take place. Possibly the poor secretory joint problem. Case 26 is difficult to assess. Clinical response in these glands rejects damage to those follow-up was not possible as the patient left the district elements which mediate normal secretion of saliva. A shortly after the two examinations, but the normal longer accumulation phase might provide more infor- scintigraphy probably excludes a diagnosis of Sjogren's mation, but it seems unreasonable to expect a patient to syndrome, and the sialography is considered to be remain stationary against the collimator face for more falsely positive. than 20 min. A Cephalostat type of stabilizing device The remaining five cases in this group all had might prove useful to immobilize the head, but there abnormal scintigraphy and normal sialography. Case 31 was no instance of a failed scintiscan resulting from has been followed up in clinic for 16 months, and is patient movement in the present series. considered to have had falsely negative sialography. She The final group of 10 patients in Table IV is the most has continued to have episodes of parotitis with probdifficult to assess. Any direct comparison between scinti- able iritis. However, diagnostic tests for Sjogren's graphy and sialography in a symptomatic gland is scar- syndrome were all negative. Case 28 was found at 194

The British Journal of Radiology, March 1990

Salivary gland scintigraphy

wmm

iimm w\ •

(c)

Figure 3. Case 17, see Table III. (a) The left parotid sialogram demonstrates a small filling defect within the duct with proximal dilatation of the intraglandular duct segments, (b) The corresponding scintiscan curve (1) reveals a good trapping and accumulation curve but no secretory response to Carbex. This indicates good functional status but an obstructed duct pattern, (c) Following surgery, with removal of a small duct calculus, the left parotid gland now has a normal secretory phase (1).

surgery to have an enlarged lymph node compressing the left submandibular gland (histologically showed sarcoid), but the underlying submandibular gland was considered normal. Scintigraphy therefore is considered to be falsely positive, although the compression by the

enlarged node may have contributed to the poor function. Case 32 has an 11 month interval between the two tests, but since a labial gland biopsy and progressive xerophthalmia confirmed the diagnosis of Sjogren's syndrome, scintigraphy is regarded as the true-positive

f\P i

IT

Figure 4. Case 26, see Table IV. (a) The right parotid sialogram reveals a narrowed duct system with minimal extravasation of contrast medium into gland tissue: ?Sjogren's syndrome, (b) The corresponding scintiscan curve (2) indicates good trapping, accumulation and secretion by the gland. The sialographic appearances are therefore falsely positive for Sjogren's syndrome. Vol. 63, No. 747

195

W. J. Pilbrow et al

examination. Case 33 was followed for 6 months with clinical improvement. She was discharged without any definite diagnosis, although tests for Sjogren's syndrome were not performed. The clinical history of Case 30 suggested sialadenitis, and the scintiscan demonstrated a degree of functional impairment despite a normal sialogram. No definite diagnosis was reached. Discussion-summary

The article presents data on 36 patients with features of salivary gland disease who had salivary gland scintigraphy; 27 of the 36 patients also had sialography of the symptomatic gland. In five patients, sialography was technically unsatisfactory. In the majority of cases there was good agreement between scintigraphy, sialography and clinical findings. All duct occlusion problems were accurately detected by scintigraphy, and follow-up scans in three patients demonstrated improvement in the secretory pattern following removal of the duct obstruction. In ten cases there was poor correlation between scintigraphy and sialography. Scintigraphy is an accurate test of the function and secretion of salivary glands and should be considered as a substitute for sialography in the assessment of possible sialadenitis, duct occlusion or Sjogren's syndrome.

196

Acknowledgments The authors wish to thank Miss G. Pilbrow for assisting in the preparation and typing of the manuscript.

References Fox, P. C , VAN DER VEN, P. F., SONIES, B. C , WEIFFENBACH,

J. M. & BAUM, B. J., 1985. Xerostomia: evaluation of a symptom with increasing significance. Journal of the American Dental Association, 110, 519-525. GARRETT, J. R. & EMMELIN, N., 1979. Activities of salivary

myoepithelial cells. Medical Biology, 57, 1-28. MISHKIN, F. S., 1981. Radionuclide salivary gland imaging. Seminars in Nuclear Medicine, 11, 258-265. MITA, S., KOHONO, M., MATUOKA, Y. & IRIMAJIRI, S.,

1981.

Diagnostic availability of Rl—sialography in Sjogren's syndrome. The Ryumachi, 11, 305-316. SCOTT, J., CAWOOD, J. I., GRIME, J. S., CRITCHLEY, M. & JONES,

R. S., 1984. Histological evaluation of quantitative scintigraphy of the salivary glands in a primate model. International Journal of Oral Surgery, 13, 45-52. SUGIHARA, T. & YOSHIMURA, Y., 1988. Scintigraphic evaluation

of the salivary glands in patients with Sjogren's syndrome. International Journal of Oral Maxillofacial Surgery, 17, 71-75. VAN

DEN AKKER,

H.

P.

&

BUSEMANN-SOKOLE,

E.,

1985.

Absolute indications for salivary gland scintigraphy with " m Tc pertechnetate. Oral Surgery, 60, 440-447.

The British Journal of Radiology, March 1990

Salivary gland scintigraphy--a suitable substitute for sialography?

By comparing 27 patients who had both scintigraphy and sialography in the assessment of salivary gland disease, scintigraphy has been shown to correla...
1MB Sizes 0 Downloads 0 Views