We're going to talk today about complications related to the diagnosis and surgery of patients with Chiari malformation and how possibly to avoid them. I should add at the beginning that my focus is on surgery for adult patients with Chiari malformation. So complications can be thought of as being acute and delayed. They can also be thought of as being related to what we do or what we fail to do. Of the acute complications, the most common ones are aseptic meningitis, leakage of spinal fluid from the incision, rarely a reaction to the durograft, and quite rarely acute hydrocephalus occurring immediately after surgery. Aseptic meningitis is the most common complication acutely and is more common when the dura has been opened either deliberately or inadvertently. And today we very rarely deliberately leave the dura open. Most people will place a durograft if they open the dura. But the most common cause of aseptic meningitis, number one, number two, number three, is blood in the subarachnoid space. And as you will see later on, we go to great lengths to try to minimize the amount of blood that enters into the subarachnoid space during surgery. Occasionally, protonaceous transudate from the muscle in the wound edge can cause aseptic meningitis. And very rarely, an inflammatory reaction to the durograft material can produce it. Many years ago, I was asked to see a patient who was said to have had aseptic meningitis following her surgery. And I had the temerity or lack of judgment, as you might want to say, to think that possibly I could do something to help this patient. When I opened, I found this, which is just scar tissue covering the entire area. And I had to make the decision that there was nothing I could do other than to replace a durograft with the one that we commonly use, which I will discuss later. Cerebrospinal fluid leakage is the most common complication of posterior fossa decompression for chiari malformation. In this connection, it is important to think what type of material is used for the duroplasty. We use autologous pericranium, but I've used bovine material, synthetic material, and currently actually use a combination of autologous pericranium with a synthetic material, which I sew together to form a kind of sandwich. And the synthetic material is placed against the brain side, whereas the pericranium is used to effect a watertight closure by sewing it to the dura. Any of these materials have their advantages and disadvantages, but I have found over the years that this combination of autologous pericranium with a plastic liner gives very satisfactory results. And I will come back to the issue of why I like to use a liner. Just to jump ahead a little bit, the most important reason that I understand now is that it benefits the compliance of the cerebrospinal fluid compartment in the sense that a very distensible durograft will prevent effective pulsatile pressures of the CSF from helping to maintain the tonsils in normal position and helping to put pulsatile pressure against the peel surface of the cord and thereby diminish the syrinx formation tendency. There are technical factors regarding dural closure that I think help to minimize the risk of spinal fluid leakage. The most important is taking care in coagulating bleeders within the dura. And as you'll see in a moment, the important thing is not to do full thickness coagulation of the dura, but rather to place the bipolars between the leaves of the dura, which is where, in fact, the bleeders actually are. We challenge the closure with a Valsalva maneuver, sometimes repeating that to make sure that there's no spinal fluid leakage from the suture edge. And we cover the dural closure with a fiber and glue sealant. As I said, leakage from the suture line may be due to necrosis of the dura and suture pullout at that point. And so we make a point, as I already mentioned, to avoid full thickness bipolar coagulation of the dura and coagulate between the leaves of the dura. Pseudomeningoceles are another complication of spinal fluid leakage, although in a delayed manner. And it used to be thought that pseudomeningoceles are irrelevant and can be ignored. But as you'll see in a moment, we've come to understand that they actually may have a more important consequence and very often require correction. As I already mentioned, in connection with the use of a plastic liner for the pericranial graft that we use, this helps to maintain the pulsatile pressure of the cerebrospinal fluid, and thereby act as a force, maintaining the cerebellar tonsils in a more normal position, as well as working against the peel surface of the cord. This is just a typical pseudomeningocele, but of interest is that this patient also has a distended dural graft or dura, which we'll come back to a little later. And this is just a cartoon to illustrate how the pseudomeningocele acts to dissipate the pulsatile force and prevent it from acting as it should against the cerebellar tonsils and the surface of the cord. This is a patient I saw many years ago, who was referred to me because she had a residual cervical thoracic syrinx, in spite of the fact that she had had a posterior fossal decompression for her carrion malformation. And as you can see, this patient has a small pseudomeningocele in the area of her decompression. I decided to re-explore this patient, and what I found was a couple of small holes in the synthetic material used for her duroplasty. And I elected at that point to do nothing but replace the dural graft with the type of graft that I described to you before, autologous pericranium with a synthetic lining material. And when I did that, the syrinx disappeared making it a proof of concept that if you can eliminate the dissipation of the pulsatile force and allow the pulsatile force to act on the pure surface, the syrinx will disappear. Another possible source of pseudomeningocele is that maybe the underlying cause was not a technical factor such as we discussed just now, but that it was the fact that this patient has idiopathic intracranial hypertension or as it's still referred to pseudotumor cerebri. This is a classical scan on the patient with a pseudotumor. Note the small ventricles and the absence of a subarachnoid space and the brain looks full. And this is this patient's pseudomeningocele. This is a patient I actually treated myself a number of years ago, and she had a classical history compatible with Chiari malformation, strain-related headaches, tinnitus, some tingling in her limbs. She was not overweight. On examination, she had some unsteadiness, but no nystagmus. I did not observe any spontaneous venous pulsations in the sitting position. And her imaging studies were typical. She had pointed tonsils extending to the upper margin of C1. Ventricles were normal size. She didn't have a syrinx. And I did the usual procedure, a posteophosphate decompression, except she had a lot of bleeding from the dural edge. And I decided at that point to leave the arachnoid intact and do a duraplasty over the intact arachnoid. This patient continues to have headaches a year after surgery. And it is my practice in that situation to bring the patient in and study them and do an intracranial pressure monitor procedure. During her intracranial pressure monitoring, she had a cardiopulmonary arrest and was immediately resuscitated. Her pressure was elevated, and we did an emergency ventricular peritoneal shunt. And except for the fact that the patient had a permanent small quadrantic visual field defect, she remained fully intact neurologically to this day. This is what you may see in a patient with pseudotumor, pointed tonsils, typical appearance of a Chiari. And after surgery, nothing much changes. She still has pointed tonsils and no cisterna magna. So the lesson from this that I learned is that you always have to consider the possibility of a pseudotumor in every patient. And if necessary, have them undergo an ophthalmologic evaluation, non-invasive phonography, either MR or CT phonography. And if there's any question, have the patient undergo an ICP measurement to make sure that they have normal CSF pressures. I had another somewhat different situation, again, a patient with a typical Chiari history, typical imaging for a Chiari malformation. And once again, his headaches did not resolve within a year of surgery. And I brought him back to do an ICP measurement as soon as he was sat upright. He had a dramatic drop in CSF pressure. This led me to do a myelogram, which disclosed an anterior thoracic dural defect. And as soon as we did a repair of that defect, his headaches disappeared, his cerebellar tonsils went back into normal position, and he made a full recovery again to this state. Tethered cord is also related to tonsil descent and can produce a situation very similar to Chiari malformation. Other delayed complications that I'd like to discuss briefly are cerebellar ptosis, dural actasia, and symptoms that worsen in relation to instability. Cerebellar ptosis is a condition in which the cerebellar hemisphere descends into the bony decompression performed in the course of a patient's posterior fossa surgery. And this can be associated with recurrent symptoms because the descent of the cerebellum will occlude the spinal fluid pathways, eliminate the cisterna magna, and recreate conditions that are very similar to those which we tried to correct in the first place. It is important to keep in mind that the object of the surgical procedure is to unblock the foramen magnum and create a pulsatile CSS space, a cisterna magna, and not to unroof the entire posterior fossa. Here's a patient I saw who has a typical Chiari appearance with a cervical thoracic syrinx, and she underwent surgery. As you can see at the upper edge of this imaging study, her cerebellar hemisphere has descended to fill the decompression site, and there is no cisterna magna. And she still has a somewhat smaller but significant syrinx cavity. This is a close-up of the area of the posterior fossa decompression, and you can see that the cerebellum has basically filled the decompression site. And her bony decompression extended almost to the turcular. What I did after evaluating this patient was to place a small metal plate in the posterior fossa region to keep the cerebellum in better position. We created a small cisterna magna, and by doing this, her syrinx got considerably smaller, didn't disappear completely, as it often does not do in an adult, but it became significantly smaller. Colitosis may occur because a particular patient's Chiari anomaly is related to low insertion of the tentorium, and it is important to assess this prior to surgery. Patients with a low insertion of the tentorium are easy to spot because their tentorium is essentially parallel to the clivus. I did not make that observation at the time, and when I did a standard posterior fossa decompression, the cerebellum in this patient with a very small posterior fossa again sank down to the C1 level. And to correct this problem, I placed a small titanium plate to maintain the cerebellum in normal position, very similar to the patient that you saw just before. Dural ectasia is a situation in which the dura or dural graft distends because there's inadequate support, and that is in fact the reason, as I mentioned before, that I like to use a synthetic liner for the autologous pericranium to minimize the likelihood of expansion of the dura. This is a patient similar to one that you saw once before, where the dura has distended, and the major manifestation that is seen in patients who have this is headache. The headache, I believe, is due to the fact that the distended dural graft or material that is used will tug against the normal dura to which it is sutured, and of course a normal dura is innervated and will then give rise to headache. And as a means of evaluating these patients, I give them a sponge and ask them to hold this in place with an ACE bandage beginning in the morning before they arise and see whether this relieves their headache. And I've had a number of patients who have had such significant relief of their headache by doing this that they don't even want to take the bandage off. And once again, what we do is re-explore these patients. Here is the appearance of the distended dura in a patient, and we put in a small titanium plate to deal with this problem. So the lesson of this is that you don't want to take off too much bone. As I said before, the objective is to uncover the cerebellar tonsils and not the entire cerebellum. And the amount of bone that should be removed can be estimated by taking a look and a measurement of the sagittal image of the patient's MR scan. Making the craniectomy too small can also result in treatment failure, and one would have to take the patient back and enlarge the decompression to uncover the tonsils. This is an introduction to a topic of what are sometimes referred to as complex Chiari patients. And you will note that this patient's clivus is almost horizontal in position, and there's soft tissue or some type of tissue behind the dens compressing the brainstem. This patient has what is, in today's terminology, called a Chiari 1.5, and it is very important to assess these patients before doing surgery and treat them accordingly. One has to identify instability, and preferably before surgery rather than operating on them and finding out that they don't have a satisfactory result or have a complication. Important in this connection is to look at the clivoaxial angle in the patient whose imaging I just showed you had a clivoaxial angle that was well below 130, and therefore abnormal. This is just an illustration, again, how we measure the clivoaxial angle. And the other important measurement is the grab oaks measurement, and again, the patient I showed had a larger than normal measurement for this parameter. This is a patient with horses and difficulty swallowing, and again, you see an abnormal clivus. There's some peculiarity about the termination of the clivus and what looks like an os odontoidium, but the clivus is below 135 degrees and is obviously an abnormal tissue behind the dens compressing the medulla, and one can easily account for this patient's horses and difficulty swallowing. This patient requires a fusion in a slightly extended position in addition to decompression of the posterior fossa to unroof the tarsals. This is a patient who had undergone a prior posterior fossa decompression and was a young woman who had persistent swallowing difficulty and required a feeding tube, and again, note that this patient has a shallow clivoaxial angle. She does not have an abnormal grab oaks measurement, but she clearly also has basilar invagination, and we performed a cranioservical fusion on this patient with a complete resolution of her problem. She was fused in slight extension and has been able to get rid of her feeding tube and lead a normal life. This is a patient with vocal cord paralysis, again, basilar invagination and obvious indentation of the brainstem by tissue behind the dens, and this patient also required decompression and fusion in slight extension to alleviate the problem. Lastly, I want to touch very briefly on the problem of patients whose syringomyelia fails to resolve or recurs or occurs, I should say, later, even though the patient has had what I consider to be an adequate posterior fossa decompression. This is a young girl who continued to have a syrinx. She does have straightening of the cervical spine, but if you look very closely, her persistent syrinx may be related to the fact that she really doesn't have an adequate cisterna magna, and we performed a reoperation on this patient, shrunk the cerebellar tonsils with bipolar coagulation, and she had significant improvement in her syrinx cavity. This is an elderly patient on whom I had performed a posterior fossa decompression, and her tonsils look in good position. I did extensive studies and could not find any kind of membrane formation, and because she was well in her 70s, I preferred to do a syrinx to subarachnoid shunt, which I rarely recommend, but in this case, it was the procedure of choice. So finally, the important thing is to choose your patients correctly, because the smaller the indication, the greater the complication. Thank you.

complications

diagnosis

surgery

Chiari malformation

aseptic meningitis

spinal fluid leakage