Search

Early in the course of treating arteriovenous malformations with radiosurgery, it was noted that the nidus isocenter and diameter, as identified by stereotactic angiography, often differed from that identified by stereotactic enhanced computed tomography (CT). To assess the sources of discrepancy between the arteriographic and CT representations of the nidus, dosimetry (from treatment records stored on an optical disk) was reviewed in 81 consecutive cases. In 44 cases, the isocenters differed by an average of 3.6 mm and the collimator size differed. Fourteen nidi were larger on CT (average, 2.6 mm), and 30 were smaller on CT (average, 4.0 mm). Overall, the angiographic and the CT nidus differed in 75% of the cases reviewed. Sources of error in the angiographic nidus determination included overlapping vessels, bony structures, fine filamentous arterioles, and irregular shapes.

An anterior fossa dural arteriovenous malformation was treated with radiosurgery. The lesion was irradiated with 3000 cGy to the 80% isodose line of an 18-mm collimator. Angiography revealed complete thrombosis 3 years after treatment. The case is described in detail, and the pertinent literature is reviewed.

After a stereotactic brain biopsy of intracranial lesions, many neurosurgeons routinely obtain a noncontrast computed tomographic scan to rule out hemorrhage. We have observed there is often a significant delayed contrast enhancement on this postoperative scan, and the pattern of this enhancement frequently differs from the preoperative pattern. To assess these changes and to determine if they are related to lesion pathology or other clinical factors, the pre- and postoperative scans of 89 patients who underwent stereotactic biopsies were reviewed. Despite an average time delay of 153 minutes between scans, 81% of the lesions enhanced similarly or better on the postbiopsy image. One lesion enhanced only on the postoperative image, and in two cases, new lesions were seen postoperatively. Ring and other enhancing lesions tended to fill in with time, becoming homogeneous. The area of contrast enhancement increased in 83%. The borders of the lesions tended to become less sharp with time. No patterns of delayed enhancement, which give histologically definitive diagnostic information, were found. The tendency toward an increased area of contrast enhancement with time suggests that computed tomographic scans taken at the time of contrast injection may not show the true extent of the lesion. For patients undergoing computed tomography-guided stereotactic biopsies or radiosurgery of poorly enhancing lesions, the localization scan might be delayed after the administration of contrast medium to improve resolution and target selection.

Fifteen cats were irradiated with an animal LINAC (linear accelerator) radiosurgery device. After 6 months, they were studied with gadolinium-enhanced magnetic resonance (MR) imaging, gross pathology after Evans blue perfusion, and histopathology. The lesion size was relatively constant in the cats receiving a dose of 7500, 10,000, 12,500, or 15,000 cGy to the 84% isodose line of a 1-cm collimator. The composition of the lesion correlated with dose: the lower-dose lesions were found histologically to contain areas of edema, demyelination, reactive gliosis, and vascular proliferation; the higher-dose lesions contained increasing amounts of hemorrhage and coagulative necrosis. The borders of the lesions were sharply demarcated from surrounding normal brain. Gadolinium-enhanced MR studies correlated well with areas of blood-brain barrier breakdown, as confirmed by areas of Evans blue staining. This model could serve as the basis for further studies of the biological effects of LINAC radiosurgery.

Radiosurgery has been shown, in multiple reports using various radiosurgical methodologies, to be a safe and effective means of treating AVMs. The major drawback of this treatment method is that it does not provide, as does open surgery, immediate protection against re-hemorrhage. For those lesions in the appropriate size range (< 3.5 cm in diameter) which are not amenable to treatment with surgical and/or endovascular techniques, radiosurgery has become an accepted and valuable part of the neurosurgical treatment armamentarium.

Between May, 1988, and August, 1991, 80 patients with arteriovenous malformations (AVM's) were treated radiosurgically at the University of Florida. A mean dose of 1650 cGy was directed to the periphery of the lesion, which almost always corresponded to the 80% isodose line. The mean lesion diameter was 23 mm. Seventy-six patients were treated with one isocenter. Angiography, performed at 1 year after radiosurgery in 41 of the 48 eligible patients, revealed an overall complete thrombosis rate of 39%. The 1-year thrombosis rate was highest in those patients with relatively small AVM's. Angiography was performed at 2 years posttreatment in 21 of the 25 eligible patients, demonstrating an overall complete thrombosis rate of 81%. This incidence did not correlate with lesion size: that is, large lesions (up to 35 mm in diameter) seemed just as likely to thrombose. Two patients (2.5%) experienced hemorrhage at some time after radiosurgical treatment, and both recovered. Two patients (2.5%) have sustained mild, but permanent, radiation-induced neurological complications.

A case of a 60-year-old man with metastatic adenocarcinoma of the colon who developed bilateral ophthalmoplegia acutely is presented. Bilateral cavernous sinus metastasis was suggested by magnetic resonance imaging and proven by endoscopic sphenoidal biopsy. The patient was treated with radiotherapy (2000 cGy) with near resolution of his symptoms. This patient represents the first reported case of acute bilateral ophthalmoplegia resulting from hematogenous metastasis to both cavernous sinuses, diagnosed antemortem. A review of the more common causes of acute bilateral ophthalmoplegia is presented.

Stereotactic techniques have been utilized for more than 40 years in clinical neurosurgical practice. There has been considerable experience with stereotactic techniques in the drainage of intracerebral hemorrhages, although its effectiveness remains controversial. Stereotactic craniotomy, often incorporating advanced computer guidance, is increasingly applied to effect complete resection of small, deep brain lesions. Radiosurgery has become a viable alternative for the treatment of arteriovenous malformations that, in most cases, are untreatable by other means. Currently, stereotaxis is a valuable part of the therapeutic armamentarium available to cerebrovascular neurosurgeons.

Individual differences in open-field activity and emotionality (number of defecations and urinations), voluntary wheel running, voluntary and forced maximal sprint running speed on a photocell-timed racetrack, swimming endurance, and maximal oxygen consumption (VO2max) were studied in 35 random bred male ICR mice. With the exception of latency in the open field and voluntary speed on a racetrack, all measurements were significantly repeatable on two successive trial days. Maximal oxygen consumption (corrected for body size) was positively correlated with amount of wheel running during each day of a 7-day test, and the correlation became stronger throughout the testing period; however, none of the daily correlations reached statistical significance. The first factor from a principal components (PC) analysis showed positive loadings (component correlations) for all measures of speed in the open field, for both voluntary and forced maximal speeds on the racetrack, and for VO2max, but a negative loading for emotionality. Wheel running and VO2max loaded positively on PC 2. Only swimming endurance loaded strongly on PC 3; this trait was uncorrelated with any other measure of physiology or behavior. These results suggest that measures of both voluntary and forced locomotor speed, as well as amount of voluntary wheel running, may be related to aerobic physiological capacities in untrained mice.

Planning radiosurgical treatment for cerebral arteriovenous malformations requires accurate definition of the true tridimensional size and shape of the nidus. Over- or underestimation of these parameters may result in undue irradiation of normal brain tissue or suboptimal irradiation coverage of the malformation leading to treatment failure. Angiography is not an ideal database for radiosurgery of arteriovenous malformations. Its shortcomings include planar representation of a tridimensional volume and simultaneous visualization of feeding arteries and draining veins overlapping with the nidus and obscuring its outline. Two illustrative clinical cases of these angiographic inadequacies are presented. Stereotactic, contrast-enhanced computed tomography may provide, in selected cases, better spatial definition of the nidus and superior anatomic detail for the final design of the radiosurgical isodose distribution.

Radiosurgery has become a widely accepted method of treating intracranial targets with a highly focused single dose of radiation. The technique allows the precise treatment of small targets through the use of stereotactic techniques. The ability to accurately focus and target tissues with both patient and gantry movements allows the production of sharp dose gradients. This permits large target doses and allows for the sparing of critical tissues which may only lie a few millimeters from the edge of the target volume. In order to carry out this technique, routine radiotherapy simulation and treatment planning must be replaced by virtual simulation and non-coplanar treatment planning. These procedures carry with them increased quality assurance responsibilities.

The University of Florida radiosurgical project began in 1986 with the following design criteria: the most accurate radiosurgical device possible, state-of-the-art computer hardware and software for dose planning, and a number of collimators sufficient to treat any lesion homogeneously. In this article we have reviewed how these goals have been met. Physical aspects of this device (accuracy, dose gradient, and dose-planning speed) as well as clinical results compare favorably with any other radiosurgical experience. We believe that LINAC radiosurgical systems are advantageous in terms of cost, variety of collimator sizes available, and currently available sophistication of computerized dose planning. In the near future, the development of conformal treatment may significantly change the entire field of radiosurgery by offering heretofore unobtainable dose plans for irregularly shaped lesions. In addition, LINAC systems may be adapted for stereotactically focused fractionated radiation therapy and for radiosurgical treatment of lesions elsewhere in the body. Accuracy and computer sophistication notwithstanding, we cannot emphasize strongly enough our belief that the least important determinant of radiosurgical results is the machine used to deliver the radiation. It is absolutely vital that all groups undertaking radiosurgery include neurosurgeons, radiation physicists, and radiation therapists who have spent considerable time studying and learning the myriad details necessary to produce consistently good results. All radiosurgical patients must be followed up carefully and studied so that we can learn how to better apply this technique. Only patients who are not candidates for conventional surgery should be treated radiosurgically, at least until much more is known about long-term success and complication rates. A patient never should be treated radiosurgically simply because the referring or treating neurosurgeon is uncomfortable with proven conventional procedures. All groups performing radiosurgery should strive to adhere to the highest possible standards. We are all responsible for verifying the adequacy of our radiosurgical systems. We are all responsible for selecting our patients well, treating them with a team approach that applies the latest available knowledge of our field, following up closely, and reporting our results honestly and thoroughly so that all can benefit. We owe this, at least, to our patients and to neurosurgery.

During the past decade, multiple linear accelerator (LINAC)-based radiosurgical systems have been developed. By measuring beam energy, beam accuracy, and dose gradient, it has been possible to demonstrate that these systems may compare favorably with the gamma knife. Clinical results also appear quite comparable in terms of efficacy and safety. LINAC systems appear to be advantageous in terms of cost, variety of collimator sizes available, and currently available sophistication of computerized dose planning.

Previously reported suboccipital transcerebellar stereotactic biopsy methods, performed with the patient in the prone position, have required general endotracheal anesthesia. A technique is described for performing such biopsies with the patient in the lateral decubitus position, under local anesthesia. Phantom planning and routine computerized tomography graphics allow the selection of a safe entry point and intra-axial trajectory to the lesion. The time required for data acquisition and the operative procedure itself compares well with that of more routine biopsy techniques.

Stereotactic radiosurgery has become a major force in the treatment of arteriovenous malformations (AVMs) of the brain. After treatment, obliteration of flow through the malformation occurs in 75%-85% of cases within 2 years, assuming the entire AVM nidus can be encompassed by the radiation field. Because the follow-up period is relatively long, a noninvasive means to assess residual transnidus blood flow is desirable. The authors report favorable findings after a comparative analysis of 85 posttreatment magnetic resonance images and 27 follow-up cerebral arteriograms in 34 patients treated with stereotactic radiosurgery. The authors found that transnidus flow can be determined from apparent signal intensity differences between tandem two-dimensional gradient-recalled echo images obtained first without and then with gradient moment nulling (flow compensation), with empirically derived pulse parameters. This method provides a means to monitor the reduction in AVM matrix size and to assess the extent of persistent arteriovenous shunting (ie, blood flow) across the nidus.

Somatosensory evoked potentials (SEPs) were monitored during 53 procedures for aneurysms of the middle cerebral artery (MCA). "Significant" changes were reported to the surgeon, who took corrective action when possible. Changes in the SEPs were categorized as follows: Type I, no change; Type II, significant change with complete return to baseline; Type III, significant change with incomplete return to baseline; Type IV, complete loss with no return; and Type V, no response at baseline. Only 1 of 37 patients with a Type I SEP had a new neurological deficit, and this was a patient who could not be examined for several days after surgery because he was in a pentobarbital coma. All 4 patients with Type III and IV changes had new postoperative neurological deficits. Perhaps of greater importance, 4 of 5 patients with Type II changes had no new deficit. These patients all had changes in SEPs that were completely reversible by clip adjustment (2), prompt removal of temporary clips (1), and inducing hypertension after aneurysm trapping (1). These cases may, therefore, represent instances in which SEP monitoring allowed the clinicians to prevent a neurological deficit. The MCA supplies the area of the somatosensory cortex that controls the hand. Median nerve SEPs are, therefore, a theoretically ideal monitor during surgery for MCA aneurysms. This study suggests that the results of MCA aneurysm surgery may be accurately predicted and improved with SEP monitoring.

Stereotactic angiography has long been the imaging database for the radiosurgical treatment of arteriovenous malformations (AVM). The following analysis reveals systematic shortcomings in the methodology, resulting in errors in determining target shape, errors in determining target size, and errors in the identification of the true AVM "nidus."

Objective: To determine the value of multimodality-evoked potential recordings in predicting outcome in comatose children., Design: Prospective series and literature review., Setting: Pediatric ICU in a university hospital., Patients: Forty-one children with a Glasgow Coma Scale score of less than 8, who were admitted to the pediatric ICU between 1984 and 1989., Interventions: Forty-one patients underwent brainstem auditory-evoked potential testing within 72 hrs of admission. Of these patients, 37 also had somatosensory-evoked potential testing at the same time. Four patients did not receive somatosensory-evoked potential testing for various nonmedical reasons., Measurements and Main Results: Multimodality-evoked potential recordings were used to predict outcome in these comatose children. Outcomes were categorized as bad (death or chronic vegetative state) or good (all other outcomes). Survivor outcomes were determined at discharge and on subsequent follow-up visits from 1 to 3 yrs later. There were no false pessimistic predictions, and two false optimistic predictions in this series. A comprehensive literature review of coma outcome prediction, using multimodality-evoked potential recordings, revealed 20 series with 982 additional patients in whom the predictive errors of false optimism and false pessimism could be determined. Five cases of false pessimism and 99 cases of false optimism were identified in the 982 additional patients. If neonates are excluded, the false pessimism number is reduced to three., Conclusions: A bad outcome can be reliably predicted using multimodality-evoked potential recordings with little chance of a false pessimistic prediction. The acceptable error of false optimism occurs frequently, since patients often die of progressive neurologic and nonneurologic problems that may or may not be present at the time of the evoked potential recordings. Thus, in comatose children, multimodality-evoked potential recordings are a useful adjunct to clinical examination and other diagnostic aids in predicting outcome and in making decisions regarding the degree of intervention to offer.

Neurophysiologic tests currently used in the diagnosis and treatment of peripheral nerve trauma include EMG, NCV, late responses, SEPs, and intraoperative NAPs. The physiologic basis of these tests has been reviewed. Examples of typical applications have been provided. The interested reader is referred to the listed references for more detailed information.

Nucleus caudalis dorsal root entry zone lesions (open trigeminal nucleotomy) are a surgical procedure which can achieve pain control without major complications in the difficult clinical setting of deafferentation-type facial pain. Two patients are reported, who had relief of pain, but also experienced neurological complications. One patient succumbed to pulmonary complications, which provided the opportunity for anatomic analysis of the lesioned area, which is discussed in detail. Potential modifications of the surgical technique are suggested.

As radiosurgery evolves into a widely available treatment modality for a variety of intracranial lesions, the need for basic research concerning the radiobiology of high-dose single-fraction ionizing radiation becomes crucial. A device especially designed for experimental radiosurgery in the cat is described. It incorporates basic parts of the Kopf stereotactic frame for accurate target positioning. A motorized pendular movement of the machine is used to describe a radiation arc, while the radiation source (either a linear accelerator or a cobalt machine) remains stationary. The pathway of the different radiation arcs is modified by rotation of the animal platform around the machine isocenter. Mechanical accuracy tests have shown a maximal alignment error of 0.15 mm, comparing favorably with that reported for modern clinical radiosurgical systems.

Forty-three patients with chronic pain disorders of different causes were selected for spinal cord stimulation. All underwent implantation of a ribbon electrode through a small laminotomy, under general anesthesia. Thirteen patients (30%) failed to obtain significant pain relief during a period of trial stimulation, and their electrodes were removed. The remainder underwent a definitive implant and were followed for a mean of 13 months (range, 3-33 months). Nineteen of them (63%) continued to experience pain relief. A detailed analysis of this series, as well as a literature review, is presented.

The anterior and posterior commissures (AC; PC) are the most frequently used reference points for target determination in functional stereotactic surgery. Computerized tomography (CT) has been slow in replacing ventriculography for identification of these points, primarily because in most methods described to date, the AC/PC plane had to be identified by reformatted CT images or cumbersome mapping procedures. The authors describe their methodology for determination of functional CT-stereotactic coordinates using the Brown-Robert-Wells (BRW) stereotactic system, axial CT imaging, and a proportional method to rapidly scale standard stereotactic map coordinates to the diencephalic size of any individual patient. 1. CT gantry angulation coinciding with Twining's line allows rapid identification of the AC/PC plane on axial CT. 2. Determination of the AC/PC distance with the "measure distance" function of the scanner enables the surgeon to rapidly find the proportional coordinates for either V.O.P., V.im., or V.C. 3. This localization method takes little more time than a routine stereotactic biopsy. Its application is easy, demanding only a few minutes of CT-console time, employing the standard computer software available in every current CT scanner. The method was been successfully employed in 25 functional thalamic procedures.

During the past decade, multiple linear accelerator (LINAC) radiosurgical systems have been developed. By measuring beam energy, beam accuracy, and dose gradient, it is possible to compare them to the older radiosurgical device, the gamma knife. LINAC systems appear to be advantageous in terms of cost, variety of collimator sizes available, and currently available sophistication of computerized dose planning. In the near future, the development of conformal LINAC treatment may change the entire field of radiosurgery significantly by offering improved dose plans for irregularly shaped lesions. In addition, LINAC systems may be adapted for stereotactically focused fractionated radiotherapy and for radiosurgical treatment of lesions elsewhere in the body.

Peripheral nervous system grafts have long been known to promote CNS axonal elongation in the rodent. To investigate whether these principles also apply to a higher order animal, more suitable for physiologic studies, we performed autogenous sciatic nerve grafts to the adult cat spinal cord. This report is a brief, initial study of these grafts. In an effort to study both descending and ascending fibers as separately as possible, laminectomies were performed at both thoracic (T2-T9) and lumbar (L2-L4) levels. Aspiration lesions were made in the left side of the cord and the sciatic grafts were introduced. Three cats were studied with standard histological techniques, confirming previous reports of glial scar formation, collagen deposition, and cyst formation at the graft-CNS junction. In 7 other animals, at 146-179 days after the initial operation, the axonal origins were traced retrogradely using horseradish peroxidase and the tetramethyl benzidine technique. Reinnervation of the graft was found by intrinsic spinal cord neurons and dorsal root ganglia neurons. The number of reinnervating neurons was found to be greater than previously reported in the rodent. There was, however, no evidence of regeneration from brainstem nuclei when the grafts were placed at spinal levels where this has been seen in the rat. This may support an absolute maximum for regenerative distances vs a relative one. Although the amount of data was limited, it did support previous reports that peripheral axotomy enhances the regenerative response seen in primary sensory neurons.

Thirteen cases of cervical carotid dissecting aneurysm have been seen at the University of Florida during the past 3 years. In our experience, two groups of patients with this disease can be distinguished clearly by either clinical or angiographic criteria. One group of cervical carotid dissecting aneurysms is characterized by spontaneous onset, unilaterality, and the absence of pseudoaneurysm formation. These lesions have a strong tendency to resolve with appropriate medical therapy. The other group is characterized by an association with obvious predisposing factors, such as fibromuscular dysplasia, angiography, or trauma. These lesions are often associated with pseudoaneurysms and rarely resolve spontaneously. These two groups of lesions probably represent a spectrum of the same basic disease process. The natural history of patients who survive a cervical carotid dissection is usually one of stability or improvement. A trial of medical therapy (to prevent embolic symptoms) and repeat arteriography are indicated before the consideration of surgical therapy.

Vasospasm of the vertebrobasilar system was induced in seven dogs by the intracisternal injection of autologous blood. Somatosensory and brain stem auditory evoked potentials were recorded before and after the induction of angiographically confirmed vasospasm. Additionally, somatosensory evoked potentials were monitored during graded hypotension to 40 mm Hg. There was no significant alteration in the evoked potentials by vasospasm or hypotension. Detailed clinical examination and postmortem histopathological studies did not demonstrate any focal neurological deficit or infarction attributable to vasospasm. Previous studies have noted close correlations between decreased cerebral blood flow and evoked potential alterations. Induced hypotension to a mean arterial pressure of 40 mm Hg in the presence of documented vasospasm was not sufficient to cause evoked potential changes, focal neurological deficit, or pathological evidence of infarction in the canine model.

Total surgical excision is the main goal of therapy for intracranial meningiomas. The controversy today involves the efficacy of postoperative radiation therapy. To evaluate this question, 132 patients with benign intracranial meningiomas, treated between October 1964 and April 1985, were evaluated. All patients had a minimum 2-year follow-up. The actuarial local control rates at 10 years for the three treatment groups were as follows: subtotal excision alone, 18%; subtotal excision plus postoperative radiation therapy, 82%; and total excision alone, 77%. The actuarial determinate survival rates at 10 years were 49%, 81%, and 93%, respectively. Postoperative radiation therapy was also effective for patients treated at the time of the first recurrence, with an actuarial local control rate at 10 years after salvage treatment of 30% for patients treated with surgery alone and 89% for patients receiving postoperative radiation therapy at the time of salvage. This analysis suggests that radiation therapy has a significant role in the treatment of subtotally excised and recurrent intracranial meningiomas.

The implantation of radioactive sources into malignant intraaxial brain tumors, interstitial brachytherapy, is a theoretically attractive treatment for these nearly uniformly fatal lesions. Optimal application of this treatment modality requires careful preoperative planning, computer-assisted dosimetry, and computed tomography-guided stereotactic implantation of the isotope-bearing catheters. A method is presented by which available computed tomographic imaging software can be used with a standard radiotherapy treatment planning computer and the Brown-Roberts-Wells stereotactic system for preoperative imaging, dosimetry planning, and brachytherapy catheter insertion. Dedicated computer facilities are not required, making this method available to any neurosurgeon with access to a current generation computed tomographic scanner.

External ventricular drainage is an important therapeutic adjunct in neurosurgical practice. Unfortunately, this procedure has been associated with a significant incidence of ventriculitis. A major source for many of these infections has been bacterial contamination of the tract of the ventricular catheter, at the site where it enters the scalp. To prevent this problem, the authors have devised a new ventriculostomy technique that involves tunneling the ventricular catheter through the scalp, between the dermis and the galea. One hundred consecutive procedures in 66 patients are analyzed in this paper. The average duration of drainage was 6.2 days. There were no infections subsequent to the insertion of the ventricular catheter in this group of patients.

A case of an encysted fourth ventricle subsequent to meningitis is described. This clinical entity may cause a progressive neurological deficit secondary to mass effect and recurrent central nervous system infections. Direct drainage, as opposed to lateral ventricular shunting alone, is indicated.

Percutaneous discectomy is a viable alternative in the treatment of herniated intervertebral discs of the lumbar spine. Anatomical analysis of the retroperitoneal surgical path utilizing computed tomography suggests that the risk of vascular injury is negligible at the L-4, L-5 level, but substantial at the L-5, S-1 level. In addition, one-third of patients otherwise suitable for percutaneous discectomy have segments of bowel obstructing the surgical path. Obtaining an abdominal computed tomographic scan with the patient in the surgical position seems to be a valuable screening technique in the evaluation of candidates for this procedure.

A radioisotope scanning and clearance procedure has been developed for the postoperative assessment of third ventriculostomy patients. The patency of third ventriculostomy sites can be verified by this procedure, and the bulk flow clearance of CSF can be quantitated. 4 postoperstive third ventriculostomy patients have been evaluated by this procedure to date. In each case the third ventriculostomy site was shown to be patent. The bulk flow clearance of CSF ranged from 170 to 342 cm3/24 h. Comparison of these figures with estimates of CSF production suggests that the majority of CSF reabsorption in these patients occurs through the arachnoid villi.

Massive, multifocal bleeding after the technically successful removal of a cerebral arteriovenous malformation (AVM) represents a frightening and usually catastrophic complication. This phenomenon, termed normal perfusion pressure breakthrough by Spetzler et al., is caused by the diversion of blood flow from the AVM into adjacent, maximally dilated, and nonautoregulating small vessels. We recently encountered three cases of cerebral AVM that exhibited breakthrough bleeding immediately after an apparently successful operation. In all cases, the surgeon was unable to control the resultant hemorrhage by standard microsurgical techniques. High dose barbiturate anesthesia, combined with blood pressure reduction to the lower levels of the normal cerebral perfusion curve, was then initiated in each case. Controlled hyperventilation, steroids, and osmotic dehydrating agents were also used to control intracranial pressure elevation. After maintenance of this regimen for several days, the patients were returned to the operating room for hematoma evacuation. At this time, bleeding was controlled easily and breakthrough did not occur. Although this regimen requires intensive anesthetic, pulmonary, and cerebral monitoring, it successfully salvaged all three patients. Two made immediate, remarkable recoveries, and the third patient is slowly improving. This protocol therefore seems promising for the management of the massive, uncontrollable cerebral swelling or bleeding that may occur as a consequence of AVM removal.

A device is described that performs automatic artifact rejection for somatosensory evoked response analysis in intraoperative and other electrically noisy environments. Although based on amplitude discrimination, rejection is not triggered by the large stimulus-dependent voltage transients associated with somatosensory evoked potentials. The device readily interfaces with commercially available evoked potential equipment.

Despite the widespread availability of image-guided stereotactic biopsy, the need for such procedures in patients with "clear-cut" clinical and radiographic diagnoses is still frequently discussed. In an attempt to ascertain the incidence of unexpected pathological findings, 100 consecutive stereotactic biopsies have been reviewed. Twelve patients were found to have diagnoses of pathological conditions that preoperatively were considered unlikely, or not considered at all. The details of these "diagnostically discordant" cases are discussed.