Localization of Salivary Glands for Botulinum Toxin Treatment: Ultrasound Versus Landmark Guidance

ABSTRACT

Background

Sialorrhea is a troublesome symptom in a variety of neurological diseases. Recently, local injection of botulinum toxin into the salivary glands was approved for treatment of sialorrhea, and injection guidance by anatomical landmarks was suggested.

Objective

To compare the accuracy of ultrasound versus previously proposed anatomical landmarks for localizing the salivary glands.

Methods

In a cross‐sectional study in 21 adults, landmark positions (LM) of the parotid gland (PG) and the submandibular gland (SG) were identified following published recommendations. The ultrasound position (US) was defined as the position representing the maximum gland thickness. The distance between positions, gland thickness, and optimal injection depth were recorded by US.

Results

Gland thickness differed significantly between LM and US positions (PG, 4 vs. 17.8 mm; P < 0.001; SG, 3.5 vs. 13.6 mm; P < 0.001). The spatial deviation between the recommended LM and identified US positions in the horizontal plane was 21 mm to the posterior direction for the PG and 19.6 mm for the SG. The deviation in vertical orientation was small for both glands; however, there was a positive correlation between the distance from the SG to the mandibular bone with age. The optimal injection depth measured by US was 11.8 mm for the PG and 13.6 mm for the SG. This showed to be positively correlated with the body mass index.

Conclusions

The position of the salivary glands differs from proposed landmarks and depends on the individual age and body weight; therefore, we recommend ultrasound guidance for injection.

Sialorrhea is a troublesome symptom of multiple neurological conditions (e.g. Parkinson's disease1 or amyotrophic lateral sclerosis2), but is also commonly observed in children suffering from cerebral palsy.3 Besides the risk of inflammation and infection of the perioral region, sialorrhea causes sleep, speech, and eating disturbances, aspiration pneumonia, and stigmatization and social embarrassment.

Sialorrhea can be reduced by local application of botulinum toxin (BoNT) in the salivary glands.4 Consequently, incobotulinumtoxin A has recently been approved by the U.S. Food and Drug Administration (FDA) and the European Union for treatment of sialorrhea based on a positive phase III trial.4 For optimal efficacy of BoNT treatment as well as to avoid side effects, an accurate targeting of the salivary glands is mandatory. Following the FDA‐approved incobotulinumtoxin A product information, the parotid gland (PG) shall be injected at “…the midpoint on the line connecting the tragus to the angle of the mandible, approximately the site of the ear lobe. Deliver injection 1 cm anterior to this…” and the submandibular gland (SG) at “…the midpoint between the angle of the mandible and the tip of the chin. Inject 1 finger breath medial to the inferior surface of the mandible at this point” in reference to McGeachan et al.5 However, current guidelines recommend the use of ultrasound (US),6 a view that is corroborated by the finding of significantly higher accuracy in targeting the SG under US guidance compared to anatomical landmarks (LM).7 This prompted us to compare the LM approach5 and the US approach in adult volunteers to assess the accuracy of targeting methods to provide relevant information for clinicians treating sialorrhea with BoNT.

Participants and Methods

Twenty‐one individuals participated in this cross‐sectional study (14 female; mean age 49 years [range 18–85]; mean body mass index (BMI) 24.4 [range 19.0–37.5]). Subjects with a history of intervention to the face or neck, neurodegenerative diseases, or under anticholinergic medication were excluded. All subjects gave written informed consent. The study was approved by the local ethics board; all experiments were conducted in accord with the Declaration of Helsinki.

Anatomical LMs were marked bilaterally following previous recommendations5 (Fig. 1) with participants sitting upright. If glandular tissue was identified at the LM position, thickness of the gland was recorded (Fig. 2). The probe was then moved to the position with the largest thickness of the respective gland defining the US position. The spatial distance between both positions was measured. Next, the distance from skin to the middle of the gland (depth) and the gland diameter (thickness) were assessed. For the SG, we additionally measured the distance between the US position and the mandible (Fig. 3). Imaging was performed using a MyLab Alpha machine with a 3‐13‐MHz (ESAOTE, Genua, Italy) linear probe.

Figure 1.

(A) Location of the proposed LM positions and direction of US probe movement. (B,C) Spatial deviation of the US position from the LM position (dashed cross) recorded in breaks of 5 mm.

Figure 2.

(A,B) Gland thickness at the LM and the US positions measured by US. (C,D) Representative US imaging taken at the LM positions. The arrow indicates the probe placed horizontally in center on the LM position. ^*** P < 0.001.

Figure 3.

(A) Image of the SG at the US position. The arrow indicates the distance between the mandibular bone and the US position. (B) The distance between the mandible and the US position correlates with age (r = 0.59; P < 0.001).

Statistical Analysis

Statistics were done with R statistics (R Foundation for Statistical Computing, Vienna, Austria), using paired t tests for group comparison, Pearson's coefficient analysis for correlations, and multiple linear regression analysis for dependency of factors.

Results

Deviation of Injection Position

For the PG, the mean (standard deviation; SD) deviation of the US versus the LM injection point was 21 mm (4.5) to posterior in the horizontal and 0.2 mm (5.2) in the vertical plane. Likewise, the mean deviation was 19.6 mm (6.7) horizontally and 0.4 mm (7.6) vertically for the SG. Hence, the injection position identified by US was always located posteriorly of the LM in the horizontal plane. In the vertical plane, however, the gland position revealed huge variability (Fig. 1).

Gland Thickness and Injection Depth

The PG was found in 74% and the SG in 62% at the LM position. Because of our methodological approach, the glands were found in 100% at the US position. The mean (SD; range) thickness of the PG at LM position was 4 mm (3.3; 0–10); the respective values for the SG at the LM position were 3.5 mm (3.4; 0–10). At the US position, the thickness of the PG was 17.8 (4.8; 12–30) and 13.6 mm (2.9; 10–21) for the SG. For both glands, the difference of thickness between positions was significant (PG, t ₍₄₁₎ = 21.975; P < 0.001; Cohen's d = 3.39; SG, t ₍₄₁₎ = 13.861; P < 0.001; Cohen's d = 2.14; Fig. 2). Optimal injection depth (SD) at the US position for the PG was 11.8 mm (2.9) and for the SG 13.6 mm (2.8).

Effect of Age on BMI

For the SG, a correlation analysis between the factors “distance between the mandible and the US position” and age and BMI revealed a strong correlation with both factors (age, r = 0.59; P < 0.001; BMI, r = 0.46; P = 0.002; Fig. 3). A multiple linear regression analysis with distance as the dependent variable and age and BMI as independent variables was significant (r² = 0.38; F ₍₃₉₎ = 12.1; P < 0.001). Here, age showed to be a significant model predictor (estimated coefficient: 0.132; P = 0.002) whereas BMI did not (estimated coefficient: 0.223; P = 0.197). Age also correlated with the distance between the cranial margin of the SG and the mandible, suggesting an increasing dissociation of the gland from the mandible with age (r = 0.66; P < 0.001). Regarding injection depth, BMI was positively correlated to optimal injection depth in both glands (PG: r = 0.8; P < 0.001; SG: r = 0.43; P = 0.005). The horizontal deviation did not correlate with either age or BMI.

Discussion

In this study, we demonstrate that US is more precise in targeting the optimal injection point for the parotid and submandibular glands in comparison to the recommended LM for injections of BoNT. For both glands, the optimal position was located posteriorly with a mean horizontal deviation of 21 mm for the PG and 19.6 mm for the SG, resulting in a significant difference in thickness between the LM and US positions for both glands. Moreover, there was no glandular tissue detected at all at the PG‐LM position in 26% of cases and at the SG‐LM position in 38% of cases. Hence, US guidance clearly accounted for individual anatomical variance whereas the LM approach did not.

The result of a more posterior localization of the SG is in keeping with the findings of a study assessing the anatomical localization of the SG in cadavers.8 They defined a safety zone for injecting the SG at 20% to 35% of the distance on a line drawn from the mandibular angle to the chin, and, in fact, 83% of the US positions in our study are located within this proposed safety zone. One may now conclude that a correction of the suggested landmarks to a more posterior point would be sufficient, but even by injecting within this safety zone a comparison of guidance techniques in cadavers showed a significantly higher injection accuracy of US versus LM guidance for the SG (92% vs. 50%).7 The poor target rate with LM guidance in the case of the SG is most likely attributable to the individual anatomical variance. There is, on the one hand, a known variability in volume of the SG, which differs between 1.6 and 15 mL.9 On the other hand, we found an age‐dependent increase of the distance of the SG to the mandible. Concerning the PG, we also found variability with regard to the optimal vertical injection point. However, the exact injection position in the vertical plane may usually have lesser impact on the target rate than a deviation in the horizontal orientation given the huge craniocaudal orientation of the gland. Finally, all these considerations neglect the question of how deep to inject. This study shows a positive correlation between the optimal injection depth and BMI, indicating individual differences also in this regard.

Our results are limited by the fact that because of our study design, blinded definition of the US position or blinding for age or BMI during data acquisition was impossible, raising the question of a potential bias.

In conclusion, the question of whether ultrasound guidance is actually needed for successful and safe treatment in everyday clinical practice is under debate, with some researchers arguing that guidance by anatomical LMs is equally effective to US guidance. It remains elusive whether injections need to be placed in the center or just somewhere in the gland or whether more than one injection point per gland would improve the clinical outcome. This is in line with ambiguous data in the literature.3, 4, 10, 11, 12

Our study proves a clear advantage in localizing the salivary glands using US versus LM guidance. It is tempting to suggest injections at landmarks located posterior to those proposed previously. However, our data suggest variability also in the vertical axis as well as concerning the injection depth. Therefore, we recommend the use of US guidance for BoNT treatment of salivary glands.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.

S.L.: 1A, 1B, 1C, 2A, 2B, 3A

N.B.: 1B, 1C, 2C, 3B

A.S.: 1A, 2C, 3B

T.B.: 1A, 1B, 1C, 2C, 3B

Disclosures

Ethical Compliance Statement: This study was approved by the ethics committee at the university hospital Schleswig Holstein (UKSH; IRB No 19‐061). Written informed consent was obtained from all participants for anonymized analysis of their data. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest: The authors report no sources of funding and no conflicts of interest.

Financial Disclosures for previous 12 months: Sebastian Loens was, in part, employed within the “DysTract – GermanDystonia Translational Research and Therapy Consortium” funded by the Federal Ministry of Education and Research (BMBF). Norbert Brüggemann received speaker's honoraria from UCB and AbbVie. He received nonfinancial support from Bayer. N.B. is funded by the Deutsche Forschungsgemeinschaft (DFG; FOR 2488, GRK1957), the Collaborative Center for X‐linked Dystonia‐Parkinsonism, and the Else‐Kröner Fresenius‐Stiftung (HA17_2017). He is a consultant for Censa. Armin Steffen received study grants by Inspire Medicals. He received travel cost and speaker's honororia by Inspire Medical, Merz Pharmaceuticals, and Intersect. He is a consultant for Merz Pharmaceuticals. Tobias Bäumer received honiara from Pharm Allergan, Ipsen Pharma, and Merz Pharmaceuticals and is a consultant for Pharm Allergan and Merz Pharmaceuticals. He received a grant from the DFG (FG2494).