The role of salivary duct morphology in the aetiology of sialadenitis: statistical analysis of sialographic features

Abstract

Morphologic characteristics of salivary ducts may contribute to stagnation of saliva. The authors hypothesized that some features might contribute to development of submandibular and parotid sialadenitis. 106 digital subtraction sialograms (DSS) were retrospectively reviewed for: degree of sialadenitis, length of Wharton’s and Stensen’s ducts (SD), and angle of Wharton’s duct (WD) genu. Student’s t test was used for independent samples to statistically compare normal and sialadenitis groups. The effect of independent variables (age, gender, side, degree of sialadenitis) on the dependent variable (length or angle) were tested using regression analysis. Submandibular duct sialadenitis was mild (67%), moderate (8%), or severe (25%); parotid duct sialadenitis was mild (57%), moderate (18%), or severe (25%). Mean length of normal WD was 58.2 mm, and 56.3 mm with sialadenitis. Mean length of normal SD was 52 mm, and 53 mm with sialadenitis. The mean angle of WD genu was 115° in normal ducts, and 119° with sialadenitis. None of the independent variables affected variation in length or angle. There were no statistical differences in duct length or measured angle between normal and sialadenitis groups. There is a wide variation in salivary duct morphology but this does not appear to be associated with the cause of sialadenitis.

Acute and chronic sialadenitis is a frequent complication of sialolithiasis, the most common salivary gland disease. It has been suggested that anatomic features such as the genu of Wharton’s duct (WD) and the length of the salivary ducts may be aetiological factors in the formation of sialoliths and sialadenitis since they may aid in stagnation of saliva and consequent superimposed infection. The authors tested the hypothesis that long salivary ducts, more acute genus in WD, and more obtuse angles between accessory gland ducts and Stensen’s duct (SD) would all be contributory factors in the development of submandibular and parotid sialadenitis.

Materials and methods

All imaging was performed during routine sialographic work-up of consenting patients based on clinical indications. These indications were the commonly encountered ones that might indicate the presence of sialolithiasis or sialadenitis: acute swelling or a palpable mass in the submandibular or parotid regions, or gradual progressive or chronic enlargement of a salivary gland. 106 unilateral non-consecutive digital subtraction sialograms (DSS) performed on patients with these clinical indications over a 7 year period at the authors’ institution were retrospectively reviewed. The cohort of patients was retrospectively chosen from all the patients having a sialogram for the above mentioned indications, by the exclusion of observable sialolithiasis on the sialograms, and exclusion of those patients with imaging or known clinical manifestations of inflammatory conditions of the salivary glands other than possible simple infective sialadenitis (e.g. Sjogren’s syndrome, sarcoidosis, or HIV sialopathy).

All procedures were performed by four radiologists using the same standard technique : after obtaining informed consent the patient was placed in the supine position, and a 0.012 in. (for WD) or a 0.016 in. (for SD) catheter was connected to a disposable syringe containing contrast medium. Its end was introduced into the respective punctum of WD or SD. The head of the patient was fixed and the correct position checked on fluoroscopy. 2 ml of non-ionic contrast medium was gently administered whilst acquiring digital subtraction images at 3 frames/s, until contrast medium was seen to fill the ductal system. Only lateral images with angles of the mandibles superimposed on the same projection were obtained on a routine basis, but frontal or angled images were occasionally required for improved interpretation of findings.

According to the guidelines of the revised Declaration of Helsinki this study reported the analysis of data (images) obtained from ‘routine sources’ where consent of individual patients and ethical approval for research analysis was therefore not considered necessary. Two radiologists retrospectively analysed all anonymized lateral DSS images that demonstrated the full extent of the ductal system. The authors reviewed the images and used electronic callipers on a PACS review workstation to measure the following parameters: degree of sialadenitis, length of WD or SD, angle of WD genu, and the angle of confluence of the accessory parotid duct with SD. Student’s t test was used for independent samples to statistically compare normal and sialadenitis groups, with significance set at P < 0.05. The authors tested the effect of independent variables (age, gender, side, degree of sialadenitis) on the dependent variable (length or angle) using regression analysis in Microsoft ® Excel. One-way analyses of variance (ANOVA) were used as regression models when the explanatory variables were dummies or qualitative in nature (gender, side, degrees of sialadenitis) and the regressands were quantitative (length or angle). Linear regression analyses were used when the regressor (age) was quantitative.

Results

Retrospective evaluation of the images available revealed that, based on the clinical indication inclusion criteria and the exclusion criteria stated above, 39 of 106 (36.8%) cases exhibited sialographic features of sialadenitis. Sialadenitis was diagnosed and classified by the presence of sialectasis, that is pruning (mild sialadenitis) or beading (moderate sialadenitis) of distal ducts, and irregular main duct segmental dilatation and narrowing (severe sialadenitis). Conversely, 67 of 106 (63.2%) patients had normal appearances on their sialograms despite their clinical presentation. There were therefore four separate patient cohorts that, in turn, allowed two analysis groups comprising: normal submandibular ducts (24) versus submandibular sialadenitis (9), and normal parotid ducts (43) versus parotid sialadenitis (30). The mean age of patients in these groups was 62, 51, 54, and 48 years, respectively. The gender (M:F) percentage distribution in these groups was 75:25, 75:25, 65:35, and 80:20, respectively. All DSSs were of good diagnostic quality, with sufficient image pixel shifting that counteracted any motion artefacts, and which optimally outlined the course of the salivary ducts free of surrounding bones. In those patients demonstrating sialectatic features on their sialograms the degree of submandibular sialadenitis was mild in 67%, moderate in 8%, and severe in 25% ( Fig. 1 a and b ). Similarly, parotid sialadenitis was mild in 57%, moderate in 18%, and severe in 25% ( Fig. 2 a and b ). The results are presented in detail as mean, standard deviation, and range, as displayed in Table 1 . The mean length of normal WD was 58 mm, and 56 mm with sialadenitis. The mean length of normal SD was 52 mm, and 53 mm with sialadenitis. The mean angle of WD genu was 115° in normal ducts, and 119° with sialadenitis. 59% of patients had an accessory parotid gland (43% patients with one accessory duct, and 16% patients had two accessory ducts). The mean angle of confluence of the accessory parotid duct with SD was 53° in normal ducts, and 56° with sialadenitis. There were no statistical differences in the duct length ( P = 0.28 for SMD, P = 0.08 for PD) or measured angle ( P = 0.30 for SMD, P = 0.41 for PD) between normal and sialadenitis groups. The results of regression analyses are presented in Table 2 . None of the independent variables affected variation in length or angle.

Fig. 1
(a) DSS showing an example of mild submandibular sialectasis. Patient facing the left. (b) DSS showing an example of severe submandibular sialectasis. Patient facing the right.

Fig. 2
(a) DSS showing an example of moderate parotid sialectasis. Patient facing the right. (b) DSS showing an example of severe parotid sialectasis. Patient facing the left.

Table 1
The results are presented as mean, standard deviation, and range when comparing normal ducts and those affected by sialadenitis, in the SMDs and PDs.
Submandibular ducts Parotid ducts
Normal ( n = 24) Sialadenitis ( n = 9) Normal ( n = 43) Sialadenitis ( n = 30)
Mean ±SD Range Mean ±SD Range Mean ±SD Range Mean ±SD Range
Age (year) 62 16.5 18–85 51 13.4 39–91 54 14.9 16–77 48 15.8 21–89
Length (mm) 58 8.8 42–76 56 8.3 48–77 52 9.6 23–73 53 8.7 31–71
Angle of SMD genu (°) 115 16.2 80–144 119 28.6 54–152
Angle of confluence of accessory PD with main PD (°) 53 18.2 23–122 56 16.5 19–91
SD, standard deviation; SMD, submandibular duct; PD, parotid duct.

Table 2
Results of regression analyses, with significance set at P < 0.05. The coefficient of determination ( R 2 ) is also displayed for results of testing the effect of age on the dependent variables. None of the independent variables affected variation in length or angle.
Submandibular ducts Parotid ducts
Normal ( n = 24) Sialadenitis ( n = 9) Normal ( n = 43) Sialadenitis ( n = 30)
Length Angle Length Angle Length Angle Length Angle
Age 0.12 ( R 2 = 0.09) 0.09 ( R 2 = 0.10) 0.53 ( R 2 = 0.04) 0.77 ( R 2 = 0.01) 0.04 ( R 2 = 0.04) 0.13 ( R 2 = 0.13) 0.06 ( R 2 = 0.18) 0.95 ( R 2 = 0.01)
Gender 0.51 0.17 0.47 0.24 0.75 0.23 0.53 0.24
Side 0.47 0.59 0.90 0.49 0.57 0.32 0.87 0.52
Degree of sialadenitis 0.85 0.42 0.93 0.99
Only gold members can continue reading. Log In or Register to continue

Jan 24, 2018 | Posted by in Oral and Maxillofacial Surgery | Comments Off on The role of salivary duct morphology in the aetiology of sialadenitis: statistical analysis of sialographic features
Premium Wordpress Themes by UFO Themes