Nelson: I think most of us would agree that third nerve palsies are not an easy situation to deal with due to the multiple muscles involved. What is your general approach to the treatment of a patient who has a third nerve palsy?
Deutsch: We often see these patients after the neuro-ophthalmologist and the neurologist have seen them. I think it's always wise to wait at least 6 months with an adult patient, such as one who has an acquired third nerve palsy from an aneurysm, because they often do recover some function.
Nelson: What do you tell the patient about your approach and whether you ultimately may wind up doing a surgical procedure?
Deutsch: It's important to guide their expectations because sometimes they're not happy even if you can get them straight. The best you can hope for is getting them a small area of single vision. I recently had a patient with third nerve palsy who had been told to wait for 2 years. We did a procedure and she doesn't really fuse, but in the 2 years she developed the ability to not pay attention to the second image and she was thrilled because she looks more or less normal. But if she had been expecting to be restored to normal function, I don't think she would have been happy.
Lingua: Being honest about what you can truly offer the patient is particularly important in third nerve palsies, where they often struggle with the most functional positions of gaze, governing walking, eating, and reading. We tell them that we hope to bring them into as good alignment as possible, but it is likely they will need some form of ancillary treatment, either prisms or partial occluders (eg, spot occluders or sector occluders), to help them look as normal as possible and function as well as they can. But going back to the way they used to be is not a reasonable goal.
Greenwald: I agree that being on the same page in terms of expectations is critically important, as it is with most adult strabismus. If there is an understanding that an approximation of normal is the goal, then I feel I can help them with that. If they want vision that's going to allow them to function in a manner similar to what they had before this occurred, it's not going to happen.You can often do more with a partial third nerve palsy than you can with a complete palsy. Those are some of the most challenging and gratifying patients that I have worked with, particularly if they don't have major eyelid problems. They have reasonable ductions in all or most directions, so I'd work on the other eye and limit the rotation of that eye to match the paretic eye. That sometimes can be a hard sell with the patient, but I think particularly if you start out early on and you have a waiting period, you can bring them around. Those patients often end up being happy with their range of binocular fusion.If complete ptosis is present, I think it's generally best just to leave things as they are. I vividly recall a teenage girl with complete third nerve palsy including complete ptosis who came from a distant state with her family for ptosis repair by Dr. Marshall Parks when I was a fellow. (Dr. Nelson and I were both fellows with Dr. Parks that year.) He lifted the eyelid to show her what it would be like after surgery, and she noted double vision. Dr. Parks told her, “Yes, you will see double afterwards.” And she said, “If that's what's going to happen, I won't have the surgery.” This precipitated a big argument between the girl and her parents. Dr. Parks dictated in his note afterward that the family must go home and reach an agreement.There are worse things than having a complete ptosis. It's not the most desirable, but a black patch is still acceptable for a male. A hairstyle that covers the eye is sometimes more acceptable for a female than what you can do with surgery.
Nelson: Let's discuss some specific cases. The first patient is a 27 year old with closed head trauma 9 months ago who has a left exotropia of 35 prism diopters and a left hypertropia of 15 prism diopters, and is unable to adduct the left eye hardly at all beyond the midline with poor depression. How would you treat this patient?
Lingua: Especially with closed head trauma and a hypertropia, I like to see some intorsion on attempting depression showing me that the fourth nerve is intact and know that I'm only dealing with a partial third palsy. If there's been some fairly good regeneration, I think they do well with recess-resect procedures. In this case where the horizontal deviation is greater than the vertical, I would do a horizontal recess-resect, releasing any positive forced ductions, and a full tendon-width vertical transposition of those horizontal recti for the vertical.
Nelson: Would you operate on the ipsilateral medial rectus?
Lingua: If there has been recovery of the third nerve, we should be able to see brisk adduction saccades to the OKN drum when the eye is in abduction where any restriction of the lateral rectus will not tether the saccade. Even if it can't move on duction testing past the midline, if as I'm watching with the OKN drum I see that they are able to generate an adduction saccade that's 50% or more than the yoke lateral rectus, I'm going to assume there's an adduction potential that's hidden by contraction of that lateral and I think they'll do well with a recess-resect. If as I'm watching with the OKN drum they pursue into abduction and when I ask them to do an adducting saccade it's a slow drift, then I assume that there has been no medial rectus regeneration and I'm going to have to do a transposition of the vertical recti nasally, without a medial rectus resection.
Nelson: If you're going to do a recess-resect, would you do a large recession of that lateral rectus?
Lingua: Yes. I think any time you have a partial paresis of the muscle, you have to accomplish the same for its antagonist. In this case, I would put that lateral rectus back 20 mm from the limbus.
Greenwald: I would take a similar approach with a couple of caveats. First, I don't hesitate to do a large resection on a totally paralyzed medial rectus muscle. I think you still get a significant effect. At a minimum this will prevent the eye from healing after a large recession of the lateral in a position that's still extremely exotropic. So I would start with a large recess-resect procedure. I'll resect up to 12 mm on the medial rectus if it's technically possible to do that, which it often is, and put the lateral rectus back up to 12 mm from the insertion or 20 mm from the limbus. Infraplacement or supraplacement of the two horizontal rectus muscles can be incorporated for mild to moderate vertical deviation. I wouldn't do anything more than that with the first procedure because the likelihood that you're going to have to do something more is high. I generally reserve superior oblique tendon transposition or attaching the lateral rectus to the lateral orbital wall for subsequent surgery.
Deutsch: A 35 prism diopter exotropia doesn't indicate complete third nerve palsy to me. The numbers would be much larger if there were no recovery of the medial rectus and I agree that it is difficult sometimes to see that recovery because there may be tightness in the lateral rectus. In this case, I might do something slightly different. I agree with a large recession of the lateral rectus, but I might be tempted to do a recession of the superior rectus with a transposition somewhat nasal of the muscle to try to get some forces generated that way and maybe an augmentation with transposition and see what I got there, preserving the medial rectus and knowing that it's possible at some point to go back and do that medial rectus.
Nelson: How much of a recession of that lateral rectus would you do?
Deutsch: Going from the insertion, perhaps 14 mm.
Nelson: The next patient is a 50 year old who had a benign brain tumor removed 9 months ago, but is left with a right complete third nerve palsy of 70 prism diopters and inability to elevate, depress, or adduct the right eye event to the midline. How would you handle this case?
Greenwald: I would encourage the patient to prepare to just live like this unless there is some significant recovery, which there still could be after 9 months. I would be reluctant to try to fix this surgically. I've personally seen recovery occur more than a year after the onset of the problem. I wouldn't from the outset say you've got to wait at least 2 years, but if there was any sign even after a year that there was some improvement occurring, I would want to wait longer until it was clear that it was not going to get better.
Nelson: What if there really wasn't that significant a ptosis?
Greenwald: In that case, I would try to bring the eye as close as possible to the center. With this magnitude of deviation, I would certainly consider transposing the superior oblique to the medial rectus insertion as part of the initial procedure. I don't have much experience with the lateral transposition to the orbital wall. It's appealing in theory, but I haven't found a need to do it in practice.
Nelson: How large a recession would you do?
Greenwald: In a case like this, I would typically put it on a hang-back suture. I don't use a hang-back suture often, but in this situation I would keep it attached with a suture just in case I need to get it back later.
Nelson: And a large resection.
Lingua: I think much like the story you told of that younger girl, this patient would need to understand that, in my opinion, he'd never be comfortably binocular. If we did anything at all, it would be purely for cosmesis to center the eye, but he would always be dealing with some method of occlusion, perhaps even a contact lens to maximally blur that eye, to ameliorate the diplopia. But he wouldn't ever be using the eyes together.In this case, there's no vertical deviation so I'm wondering whether the fourth nerve was also involved. I would consider trying to show the patient what it might be like to have a completely still eye using tape to elevate the eyelid or by doing a botulinum toxin injection to see if there was any relaxation of the lateral that could occur, or was it restricted into abduction. Back in the 1980s when I was part of the original trials of botulinum toxin injection for strabismus, posterior migration of the drug would give the patient a complete ptosis and the eye wouldn't move. So it could be a way to show him that we could center the eye but it's not going to move well and he's going to see double anytime he moves. I haven't done this, it's just an idea.
Nelson: Would you do similar to Dr. Greenwald and do a large recess-resect?
Lingua: If I was going to try to center the eye surgically, I know that with the third nerve completely out I'm only going to do well if I can also take out the sixth nerve, which means I need to extirpate the anterior lateral rectus past the intermuscular septum to get the eye to center in primary gaze. In some of these more aggressive procedures for nystagmus that I've been looking into when you extirpate past the intermuscular septum, you'll get a −4 underaction of that muscle and that's the only way I think you can get that eye to permanently be released of its lateral rectus tone with secondary positive forced ductions.I agree that a resection of the denervated medial is of value to hold the eye in. In longstanding sensory exotropia, after multiple surgeries on the horizontal recti to no avail, I have also performed a nasal anteriorization of both obliques to decrease abduction tone that may be the cause of their recurrent exotropia. So in this case I would consider an anterior transposition of the superior oblique to the nasal border of the superior rectus insertion. If that fourth nerve is intact, that gives you both adduction and depression. The position is a much better place to put the superior oblique than the superior border of the medial where you only get adduction on attempted depression. It also improves the appearance such that when he looks down he appears to converge.
Deutsch: I agree with Dr. Greenwald that I have not done suturing to the periosteum, but it sounds like an interesting possibility. On rare occasions I have done the superior oblique transposition, and I think it does make sense for this large and poorly moving eye. Most authors who have published large series for this large a third nerve palsy seem to believe that it is not helpful to do the medial rectus resection. I don't think we would lose much by doing it, but again education is important.As I get older I realize that 50 is really young and that appearance is important because he may be in business. If he has to wear an occluder contact lens, so be it if he feels more confident and can go out into the world presenting a more confident appearance. I think that's very important and shouldn't be lost.
Nelson: Let's assume that all three of you did some kind of procedure, and the patient comes back with an exotropia of 35 prism diopters. What are you going to do now?
Greenwald: The second operation is more challenging in these patients. Assuming there was little adduction, I would go back and find the lateral rectus, detach it, and try to inactivate it further. If the medial rectus hadn't been resected and the superior oblique had not been transposed, I would definitely do a resection of the medial rectus. I usually move the superior oblique and reattach it just above the medial rectus. Or if I'm resecting the medial rectus at the same time, I would actually consider compressing the medial rectus together in the lower half of the insertion while I'm attaching the superior oblique tendon to the upper half of the medial insertion. I think there's a good chance of making the eye look centered again but, as the others have remarked, it's not going to move from side to side.
Deutsch: The first thing I'd say is look how much improvement you've achieved compared with before. I don't think surgery on the other eye will help much. If I had done all that I could do, I don't think I would go back and do much more. There are cosmetic options with glasses, prisms, and contacts, but it's frustrating for the patient.
Lingua: I would do one more surgery. I wouldn't do anything on the fellow eye unless there was partial recovery of the third nerve. But on this eye, I wouldn't hesitate to do a nasal transposition of the nasal half of the vertical recti and isolate those halves, and make sure I have good visualization so I can be sure to preserve the temporal ciliary vessels. I know that they're not innervated so I'd have to resect them a good 8 mm and I would bring them over to the superior and inferior border of the medial rectus that I'd previously resected. I would adjust the position under anesthesia, which is probably fair because, on awakening, his innervated eye position isn't going to be a lot different. I'd be sure I wasn't inducing an unwanted vertical deviation because sometimes they'll get a hypertropia if the vertical recti are unequally resected. I would stop there.
Nelson: The next case is a 35 year old who had closed head trauma 1 year ago and now has a third nerve palsy. He has a left exotropia of 70 prism diopters and a left hypertropia of 18 prism diopters with an inability to elevate, depress, or adduct beyond the midline. How would you handle that?
Lingua: This is another case where I'd agree with Dr. Greenwald. I am often surprised at how much regeneration there might be that I'm not seeing. So I would again start with a large recess-resect with an infraplacement full tendon width to release any forced ductions and to see if there's been any recovery, with the patient well aware that I may have to go back and do a secondary transposition procedure on that eye or a counter paresis procedure on the other eye.
Greenwald: I'd take a similar approach, although I'm surprised there is so little active elevation given the fact that the eye is hypertropic. I agree with Dr. Lingua, this looks like an inferior division palsy and I would expect there to be some function of the superior rectus that would make me think about transposing it over to the nasal side. I would not have any enthusiasm for transposing a muscle that wasn't innervated. Again, you have to ask yourself what the goal of surgery is and what the patient expects.
Deutsch: I might be tempted to do an extremely large recession of the lateral rectus, perhaps again suturing into the periosteum and taking that superior rectus and moving it over to the border of the medial rectus.
Nelson: The last case is a 2 year old with a chin-up position of 15 degrees since infancy and moderate ptosis of the left eye. He has an exotropia of 15 prism diopters and a left hypotropia of 25 prism diopters, and is unable to elevate that eye even to the midline with mild adduction deficit. How would you handle this patient?
Greenwald: This certainly is consistent with a partial congenital third nerve palsy that would account for the ptosis, the vertical deviation, and the exodeviation. I think this child will need a combination of surgery on both eyes. It's difficult to do a good preoperative assessment in a young child and I'd be inclined to do less surgery for the initial procedure than if this were an older patient on whom I could do a good preoperative assessment. Also, as a general rule, the most important requirement in a young child is to make sure that they're fusing part-time to protect visual function.One thing you might consider if this child is using a chin-up position to fuse, which it sounds like he might be, is not doing anything definitive until he's 3 or even 5 years old. Whatever treatment he would have would involve recessing the inferior rectus of the involved eye, recessing the lateral rectus of the involved eye, and also doing something on the other eye.
Nelson: What would you do on the other eye?
Greenwald: Probably recess the lateral rectus for the medial rectus underaction and the superior rectus for the limited elevation.One thing I'd like to comment on is that if you're operating on the non-paretic eye, you should operate for the secondary deviation. Basically, if the patient can get the paretic eye into the primary gaze position and the deviation in the other eye is something you can correct with surgery on that eye, that's a good procedure to do. So if he can only get the eye to the center by generating a 70 or 80 diopter deviation of the other eye, you can't expect other eye surgery to be adequate. But if he can get the paretic eye to the center with 35 or 40 diopters in the other eye, that's a doable procedure.
Deutsch: I'm a little confused about this case. When he adopts his head position, would you say it's more for ptosis?
Nelson: I would say it's more for the hypotropia.
Deutsch: So when he adopts his head position, he's straight. He can fuse.
Nelson: When you put his head straight, he has a left hypotropia of 25 prism diopters. That's why he has a chin-up position and an exotropia of 15 prism diopters.
Deutsch: He has an exotropia but he can fuse that in his head position. So this case is different from the other cases you've presented for two reasons. First, this is a child. Second, he has fusion. I agree with Dr. Greenwald that getting repeated measurements and planning your approach is important. As long as he is able to fuse with a 15 degree head turn, it's important to see how much torsion there is from fourth nerve action. I think addressing the hypotropia is most important. I would recess the inferior rectus, but I would be careful and have repeated measurements in as many gaze positions as I could get prior to planning the surgery.
Lingua: I would try to do everything I could, approaching it a little like double elevator palsy. There's a V-pattern exotropia. If he's chin up and he's able to fuse in down gaze, I wouldn't be surprised if I found some positive forced ductions under anesthesia for that inferior rectus. So I'd plan to do a horizontal recess-resect for the 15 diopters of exotropia and recess the inferior rectus if the forced ductions were positive. I would be doing three neighboring muscles on a 2 year old. I don't think that's a problem: resect and supraplace the medial, recess and supraplace the lateral for the 15 diopters of exotropia, and recess that inferior rectus as need be, for approximately 20 of the 25 of vertical. Recessing the inferior rectus may decrease the esodeviation in down gaze, ameliorating the V pattern. If the hyperdeviation persisted, then, as Dr. Greenwald noted, depending on how much the right eye went up when he fixed with that left eye, I'd be thinking about a sizeable, probably 12 mm, posterior fixation of the right superior rectus. Last, but not least, I'd refer the patient for repair of the eyelids.
Nelson: These are not easy cases, but thank you for sharing your expertise.