A neurosurgeon from the University of Pittsburgh believes that some cases of failed Chiari decompression surgery may be due to a partial misdiagnosis and co-existence of another condition, idiopathic intracranial hypertension (see Side Bar). Dr. Ghassan Bejjani along with neuro-opthamologist, Dr. Kimberly Cockerham, and their colleagues reported on 6 cases of failed decompression surgery in the February, 2003 issue of the European journal Acta Neurochirurgica.

Bejjani's study looked at Chiari patients whose primary symptoms returned after decompression surgery. The six cases included 5 women and 1 man ranging in age from 19 to 43 years. Each patient had a Chiari malformation verified by MRI and neurological symptoms consistent with Chairi and severe enough to warrant intervention. As to be expected, the most common symptom was headache along with neck pain, visual disturbances, and weakness and numbness in arms and legs. None of the six had a syrinx.

The patients all underwent similar decompression surgeries involving a suboccipital craniectomy, C1 laminectomy, and duraplasty. For all six patients there was some symptom relief initially, but symptoms began to return 1 to 9 months later. The patients underwent follow-up MRI's, including cine-MRI to look at CSF flow. The MRI's were reviewed by a neuroradiologist, who did not know the purpose of the study or the clinical reports of the patients, to assess whether there was adequate decompression. The radiologist looked at whether the tonsils had moved up or down after surgery, the shape of the tonsils, and whether there was CSF flow behind the tonsils. There was very little movement of the tonsils as compared to before surgery, but in 4 of the patients the tonsils were more rounded in shape. In addition, there was CSF flow behind the tonsil in all 6 patients. While not conclusive, these findings are suggestive that surgically, the decompression was adequate and should have relieved the symptoms.

In addition to the radiological review, 4 of the 6 patients underwent an eye exam by Dr. Cockerham. All four patients showed signs of elevated intracranial pressure (ICP). Specifically, the patients lacked what are known as spontaneous venous pulsations (SVPs). In most people, the veins in the retina (in the back of the eye), fluctuate in size periodically. In a person with elevated ICP, these fluctuations do not occur. Besides the absence of SVP's, two of the patients showed additional signs of elevated ICP.

Following the MRI's and eye exams, all six patients underwent a lumbar puncture to both measure their ICP and relieve it if it was elevated. Most of the patients were near or above the pressure level considered normal and all patients reported a temporary (about one week) improvement in symptoms following the lumbar punctures. Because of this, four patients chose to have ventriculo-peritoneal shunts placed in an effort to drain CSF and lower their ICP. Two patients chose to undergo periodic lumbar punctures along with taking acetazolamide, a drug which can lower ICP. Following these treatments, all patients reported significant improvement in their symptoms and were stable at least 16 months later.

Why did CSF drainage - using shunts or lumbar punctures - work where decompression surgery failed? The researchers cite several possibilities, including surgical scarring disrupting CSF flow and inadequate decompression, but speculate that at least some of the patients also suffered from idiopathic intracranial hypertension - a condition where ICP is abnormally high for unknown reasons. Bejjani points out that in the case of surgical scarring, the lumbar punctures should not have relieved symptoms as the procedure is below the level of CSF blockage. While it is difficult to define an adequate decompression, the MRI's for most patients in this study showed signs that are typical of an adequate decompression.

In support of the idiopathic intracranial hypertension (IIH) theory, Bejjani points out that IIH symptoms, as reported, are similar to Chiari and could be difficult to distinguish. In addition, the eye exams and elevated pressure readings after surgery are suggestive of IIH. Adding to this theory is that some of the patients were overweight, a predisposing factor for IIH. Perhaps the strongest piece of evidence that supports the IIH theory however, is that the patients responded to treatments designed to drain cerebrospinal fluid and didn't respond to the decompression surgery.

An association between IIH and Chiari has been identified and discussed previously in the medical literature, but the exact nature of the relationship is not known. Does a Chiari malformation cause elevated ICP, or can sustained high pressure actually cause the cerebellar tonsils to herniate? Much more research will be required to sort out the link between the two conidtions.

Failed decompression surgery can be devastating to a patient and happens much too frequently. While there are likely many reasons why surgeries fail - as reported elsewhere in this issue - for one group of patients, even if its a small group, the reason may be because of a second, often treatable, condition.