Ultrasound-Guided vs Anatomical-Localization of Injectables
Comparing accuracy of ultrasound-guided vs anatomical localization of botulinum toxic injections into cadaver salivary glands.
Richard Trosh, MD, and Stuart Isaacson, MD
PUBLISHED February 05, 2020
Stuart Isaacson, MD: When we use Myobloc, we have to decide which glands to inject. We have to decide which dose to inject. We have to decide whether to do it with anatomical localization or with ultrasound guidance. There’s a lot of information that someone might need to know. Can anyone inject these salivary glands, or should it be done only by specialists? Because we’re wondering whether everyone who takes care of people with Parkinson disease should be asking about sialorrhea. Can anyone taking care of people with Parkinson treat sialorrhea with a drug like Myobloc?
Richard M. Trosch, MD: I think they could. If you look at both the Xeomin trials and the Myobloc trials, they use either ultrasound guidance or anatomic guidance. The results were pretty good in both groups and not really significantly different. It can be done, and fairly safely, without ultrasound guidance. I’ve used an ultrasound for about 9 years. For about 3 years before that, I was using just anatomic guidance, and my results were good. But I’m more comfortable now with ultrasound. And we’ve had this discussion on whether to use ultrasound or not to use ultrasound. Good results can be attained.
I would caution physicians that the more difficult gland to find is the submandibular gland. Parotid is a lot larger. It’s the largest of our major salivary glands, and it’s more predictable in its size and shape. The submandibular gland is more variable, and where you find it tends to be farther back. The older you are, it tends to be smaller in some people, and that factor can be as much as 10-fold in terms of size, the submandibular gland where the parotid doesn’t vary as much. So it’s certainly an easier target. You also can get into more trouble with injection of the submandibular gland because of the risk of dysphagia. I’ve not seen dysphagia occur when injecting only parotids. You certainly could have dry mouth if you gave too much but probably not trouble swallowing.
But when you’re injecting submandibular, there are other structures there that you could be in if you’re too deep or in the wrong position such as the tongue, the mylohyoid muscle or other structures could complicate things. So you have to be very careful with submandibular. I tell physicians, if you don’t have ultrasound or you’re inclined to use it, limit your injections to the parotid gland and just start at low doses. If the low dose doesn’t work, the next time you’ll increase the dose. I tell patients this. The first time I see them I tell them that we’re going to start at a low dose and to think of it as a test dose. If you have no adverse effects, then that’s good. We’ll increase the dose next time if it’s not efficacious. If you don’t have benefit the first time, that’s OK, because I’d rather you not have benefit and have no adverse effects than have you have an adverse effect such as dry mouth or some other problem.
If someone starts at a low dose and limits their injections parotid and stays within anatomic guidelines of where the gland is, I think they’ll be safe doing it with just anatomic guidance. But I see advantages to using ultrasound. Ultrasound can avoid structures. The facial nerve runs right about here, just under the auditory meatus. The carotid artery runs just deep to the more inferior portions of the parotid. When you’re dealing with the submandibular gland, you have the facial artery running right through it, so you see these structures on ultrasound. The other thing you’ll see—and it’s actually fairly large when you’re injecting the parotid—is the parotid duct. If you’re injecting there, your patient would tell you: “I taste something salty in my mouth.” That’s the saline that’s used in the product. It’s not dangerous, but your toxic injectable is wasted.
For reasons of safety, I like ultrasound. There is a cadaver study that was done in South Korea in which they had injectors, who were experienced, perform these injections using anatomic guidance or ultrasound. They used a dye, then they resected the parotids and submandibulars and saw how accurate they were. These were the same injectors comparing how good they were with anatomic versus ultrasound. When it came to the submandibular, they missed 50% of the time. When it came to the parotid, they missed about 21% of the time. When they used ultrasound, they were about 97% accurate. So even for good injectors who are used to using ultrasound, without ultrasound it is a lot less accurate procedure.
Stuart Isaacson, MD: It’s interesting. In the trials, some patients had ultrasound, some had anatomical localization, and the efficacy, safety, and tolerability were the same.
Richard M. Trosch, MD: They were.
Stuart Isaacson, MD: So even though you may not be as accurate from your study, in the clinical trials it was pretty similar.
Richard M. Trosch, MD: It was. But there were some methodological problems with the study. One is the vast majority of the toxic injectable was given into the parotids, which is easy to find and—at least in the case of the Myobloc trials—a very small dose into the submandibular. If you miss there, it probably had less effect. I think 250 units here, 500 units here is the starting dose. So they started at 1500 units, they went up to 3500 units. But the lion’s share was parotid anyway. So whether you hit here didn’t really make a difference.
There were also problems with needle sizing. We used the ½-inch needle, and to hit the submandibular, you need a 1-inch needle. It’s about 2 cm in. So even the group that was using ultrasound guidance wasn’t really injecting the submandibular where it counted. And the parotids are pretty large. The other problem I find is with repeated injections: the gland, in a lot of patients, atrophies. It’s big and fat and easier to find the first time you do it, and then for subsequent injections, you’re looking around a little to find where.
For patients I’ve been injecting for years, I really need it because I just can’t go anywhere in here and find it. I may have to go a little more posterior or inferior. And in the Myobloc trials and the Xeomin trials, we looked only at the first injection, and these were de novo patients or patients where the toxic injectable had worn off and there were 16 weeks between injection. In that situation, it may be a little easier. And where there are repeated injections over a period of time, ultrasound guidance becomes more important.
Stuart Isaacson, MD: There are some things that are very clear about sialorrhea, and there are some things that are still unclear. The debate and controversy persists about these. I tend to use ultrasound when patients don’t have a good response to my anatomical localization, and you tend to use it as the mainstay. I guess we all have our own clinical patterns.