Fiber tracking from DTI using linear state space models: detectability of the pyramidal tract.

Diffusion tensor imaging (DTI) is an emerging and promising tool to provide information about the course of white matter fiber tracts in the human brain. Based on specific acquisition schemes, diffusion tensor data resemble local fiber orientations allowing for a reconstruction of the fiber bundles. Current techniques to calculate fascicles range from simple heuristic tracking solutions to Bayesian and differential equations approaches. Most methods are based only on local diffusion information, often resulting in bending or kinking fiber paths in voxels with reduced diffusion properties. In this article we present a new tracking approach based on linear state space models encompassing an inherent smoothness criterion to avoid too wiggly tracked fiber bundles. The new technique will be described formally and tested on simulated and real data. The performance tests are focused on the pyramidal tract, where we employed a test-retest study and a group comparison in healthy subjects. Anatomical course was confirmed in a patient with selective degeneration of the pyramidal tract. The potential of the presented technique for improved neurosurgical planning is demonstrated by visualization of a tumor-induced displacement of the motor pathways. The paper closes with a thorough discussion of perspectives and limitations of the new tracking approach.

Development and characterization of an animal model of carnitine deficiency.

Mammals cover their carnitine needs by diet and biosynthesis. The last step of carnitine biosynthesis is the conversion of butyrobetaine to carnitine by butyrobetaine hydroxylase. We investigated the effect of N-trimethyl-hydrazine-3-propionate (THP), a butyrobetaine analogue, on butyrobetaine hydroxylase kinetics, and carnitine biosynthesis and body homeostasis in rats fed a casein-based or a vegetarian diet. The K(m )of butyrobetaine hydroxylase purified from rat liver was 41 +/- 9 micromol x L(-1) for butyrobetaine and 37 +/- 5 micromol x L(-1) for THP, and THP was a competitive inhibitor of butyrobetaine hydroxylase (K(i) 16 +/- 2 micromol x L(-1)). In rats fed a vegetarian diet, renal excretion of total carnitine was increased by THP (20 mg.100 g(-1) x day(-1) for three weeks), averaging 96 +/- 36 and 5.3 +/- 1.2 micromol x day(-1) in THP-treated and control rats, respectively. After three weeks of treatment, the total carnitine plasma concentration (8.8 +/- 2.1 versus 52.8 +/- 11.4 micromol x L(-1)) and tissue levels were decreased in THP-treated rats (liver 0.19 +/- 0.03 versus 0.59 +/- 0.08 and muscle 0.24 +/- 0.04 versus 1.07 +/- 0.13 micromol x g(-1)). Carnitine biosynthesis was blocked in THP-treated rats (-0.22 +/- 0.13 versus 0.57 +/- 0.21 micromol x 100 g(-1) x day(-1)). Similar results were obtained in rats treated with the casein-based diet. THP inhibited carnitine transport by rat renal brush-border membrane vesicles competitively (K(i) 41 +/- 3 micromol x L(-1)). Palmitate metabolism in vivo was impaired in THP-treated rats and the livers showed mixed steatosis. Steady-state mRNA levels of the carnitine transporter rat OCTN2 were increased in THP-treated rats in skeletal muscle and small intestine. In conclusion, THP inhibits butyrobetaine hydroxylase competitively, blocks carnitine biosynthesis in vivo and interacts competitively with renal carnitine reabsorption. THP-treated rats develop systemic carnitine deficiency over three weeks and can therefore serve as an animal model for human carnitine deficiency.

Improvement of surgical simulation using dynamic volume rendering.

In the last years high efforts have been taken to develop surgical simulators for computer assisted training. However, most of these simulators use simple models of the human's anatomy, which are manually created using modeling software. Nevertheless, medical experts need to perform the training directly with the patient's complex anatomy, which can be received, for example, from digital imaging datasets (CT, MR). A common technique to display these datasets is volume rendering. However, even with high-end hardware only static models can be handled interactively. In surgical simulators a dynamic component is also needed because tissues must be deformed and partially removed. With the combination of springmass models, which are improved by neuro-fuzzy systems, and the recently developed OpenGL Volumizer, surgical simulation using real-time deformable (or dynamic) volume rendering became possible. As an application example the simulator ROBOSIM for minimally invasive neurosurgery is presented.

ROBO-SIM: a simulator for minimally invasive neurosurgery using an active manipulator.

This application report describes the software system ROBO-SIM, which is a planning and simulation tool for minimally invasive neurosurgery. Using actual patient's datasets, ROBO-SIM includes all planning steps necessary. These are; defining the trepanation point for entry into the skull and the target point within the depth of the brain, checking the surgical track, performing virtual trepanations (virtual craniotomy), and defining sanctioned volumes for use with an intra-operative active manipulator. With the additional simulation part, neurosurgeons are able to simulate whole surgical interventions directly on the patient's anatomy using the same instruments as for the real operation. First tests with ROBO-SIM are performed on actual patient's datasets with ventricular tumours.

Fibrinolysis and aspiration of experimental intracerebral hematoma reduces the volume of ischemic brain in rats.

The hypothesis was tested in rats that brain ischemia by an intracerebral hematoma can be ameliorated by fibrinolysis and aspiration of the hematoma. Intraparenchymal blood clots were generated by the injection of 50 microliters of autologous blood into the right caudate nucleus in two portions seven minutes apart. Thirty or 120 min later 12 microliters recombinant tissue plasminogen activator (rtPA) or 0.9% NaCl were injected and after 30 min the resolved hematoma was aspirated. Six hours later cerebral blood flow (CBF) was determined by 14C-iodoantipyrine autoradiography. Tissue volumes of CBF < 10 ml 100 g-1 min-1 and CBF < 30 ml g-1 min-1 were determined. Clot and lesion volume were quantified histologically from serial sections stained for succinate-dehydrogenase (SDH) activity. In rtPA-treated rats the major part of the hematoma could be evacuated 30 min as well as 120 min after production of the clot. The volume of ischemic brain (CBF < 10) was significantly reduced (p < 0.05) in the rtPA group compared to saline-treated and control groups irrespective of the time of treatment. In contrast, no difference was found between the control group and the experimental groups when the volumes of brain tissue surrounding the lesion were compared which had values of CBF < 30 ml 100 g-1 min-1. In a rat model of intracerebral hemorrhage, treatment by local fibrinolysis followed by aspiration of the hematoma is effective in reducing the volume of ischemic brain tissue and of the remaining clot volume.

Virtual endoscopy for planning and simulation of minimally invasive neurosurgery.

OBJECTIVE: This article demonstrates the usefulness and the problems of present-state software for virtual endoscopy as a tool for the planning and simulation of minimally invasive neurosurgical procedures. METHODS: The software Navigator (General Electric Medical Systems, Buc, France) was applied for virtual endoscopic visualization of three-dimensional magnetic resonance data sets of healthy volunteers and neurosurgical patients, using a clinical magnetic resonance scanner (1.5-T Signa Hispeed; General Electric Medical Systems). Classical approaches for minimally invasive procedures were simulated. RESULTS: Virtual endoscopy provided impressive three-dimensional views of intracranial and intracerebral cavities, with visualization of many anatomic details of the brain's inner and outer surfaces. The method proved to be especially suited for the simulation and planning of operations of intraventricular lesions, for which the technical limitations of the present state of development of this method have fewer implications. However, the present state of technology, as described in this article, has two major shortcomings: 1) the blood vessels cannot be visualized together with the brain tissue and cranial nerves; and 2) different tissue compartments cannot be stained in their original coloring, which would facilitate their recognition and thus orientation in space by anatomic landmarks. Another important disadvantage at this stage is time consumption for many single working steps. CONCLUSION: Virtual endoscopy is a promising tool for teaching and training in intracranial neuroanatomy as well as for planning and simulation of minimally invasive (e.g., endoscopic), mainly intraventricular, operations. Direct clinical application is, at this stage of development, limited by several technical shortcomings of visualization and quantification of distances and modeling of surfaces.

Experimental intracerebral hemorrhage: description of a double injection model in rats.

For experimental purposes, the most common technique of producing an intracerebral hematoma in rats is the injection of unclotted autologous blood. All modifications of this model share the problem that size and extension of the hematoma are not reproducible, because the injected blood either ruptures into the ventricular system or it extends to the subarachnoid or subdural space. Therefore a double injection model of experimental intracerebral hemorrhage in rats has been developed using 19 male Sprague-Dawley rats. After inducing anesthesia a cannula was stereotactically placed into the caudate nucleus and an intracerebral hematoma was produced with the double injection method in which first a small amount of fresh autologous blood is injected which is allowed to clot (preclotting) in order to block the way back along the needle track; the actual hematoma is produced in a second step of the injection. The clot volume was measured on stained serial sections. A total injection volume of 50 microliters of autologous blood produced intracerebral hematomas of 41.1 +/- 10.0 microliters and of similar shapes. The double injection method allows to generate reproducible hematomas in rats. This new model of intracerebral hemorrhage will allow further investigation of fibrinolytic and cytoprotective therapies.

Cost-effectiveness in ambulatory care: alternative approaches.

By relating health care resources and their use to health outcomes, through a coherent macro resource allocation framework, one can examine the health care system for allocative efficiencies. In this article, costs and outcomes are analyzed in such a framework, scenarios for optimizing the use of health care resources--while still maintaining existing health outcomes--are explored, and the implications for ambulatory care are discussed. The research clearly shows that much can be done to make health care systems more efficient without jeopardizing health outcomes.

Ultrasound stereotaxic endoscopy in neurosurgery.

Stereotaxic endoscopy assisted by laser- and video technique can be used in a circumscribed number of intracerebral lesions. In the present series of patients, ultrasound imaging has been used as a stereotaxic method to guide the tip of the endoscope to a target area in the depth of the brain via a burr hole in the cranial vault. Initial experience has been obtained in the evacuation of various intracerebral and ventricular as well as cerebellar haematomas; moreover, ventricular tumours can be laser-coagulated and resected. In cases of cystic hemispheric tumours biopsies can be taken under visual control and the inner surface coagulated with laser. The method is less traumatic compared to conventional neurosurgery.

Unfavorable outcome following early surgical repair of ruptured cerebral aneurysms--a critical review of 238 patients.

Among 238 consecutive patients admitted early with ruptured cerebral aneurysms, surgical repair within 48-72 hours was feasible in 200 cases. Unfavorable outcomes among the latter 200 patients are analyzed and discussed in this paper. Preoperatively, 148 patients were in Hunt and Hess grades I-III, 33 were in grade IV, and 19 in grade V. After clipping of the aneurysm, all patients received a regimen of topical intracisternal and intravenous/peroral medication with the calcium antagonist nimodipine. The overall rate of unfavorable outcomes was 25%, ie, outcome with moderate or severe deficit or lethal outcome. The reasons for unfavorable outcomes among these 49 patients were the devastating effect of the bleed (severe subarachnoid hemorrhage or additional intracerebral hemorrhage) in 31 patients (15% of the 200 patients), a surgical complication in 11 (5.5%), preoperative rebleeding in three (1.5%), delayed ischemia from vasospasm in one (0.5%), and various others in three further patients (1.5%). Unfavorable outcome occurred in 11% of patients with preoperative grades I-III, in 52% of patients with grade IV, and in 16 of 19 patients with grade V. Among the 141 patients with subarachnoid hemorrhage but not intracerebral or intraventricular hematoma, 16 made an unfavorable outcome, ie, 11% versus 56% among patients with intracerebral hematoma/intraventricular hematoma on preoperative computed tomography scan. The present data seem to speak in favor of early surgery. Since half of the patients with intracerebral hematoma and poor outcome had suffered previous warning leaks, it appears to be a continuing challenge to diagnose warning leaks before a massive hemorrhage occurs.

Fibrin tissue adhesive sealant for the prevention of CSF leakage following transsphenoidal microsurgery.

Fibrin glue (Beriplast, Behring or Tissucol, Immuno) was used for 126 sublabial transseptal transsphenoidal operations in 119 patients from April 1981 to March 1987 in a variety of sellar pathologies together with septal bone and spongycel to seal the sellar floor and the anterior wall of the sphenoid sinus. The incidence of postoperative rhinorrhea was 1.6%. A review of the literature revealed an incidence of 1.5%-9.6% with the use of autologous tissue for sellar packing such as fat or muscle; fibrin glue combined with autologous grafts led to postoperative rhinorrhea in 1%-4.4%. The present results support the view that sellar and sphenoidal sealing with fibrin glue instead of muscle or fat tissue does not raise the incidence of postoperative rhinorrhea.

Cerebrovascular response to changes of cerebral venous pressure and cerebrospinal fluid pressure.

Using parietal cranial windows and multichannel videoangiometry, pial vessel responses were studied in cats during stepwise elevation of superior sagittal sinus pressure (Psss) to a level of 50 mmHg or reduction of CSF-pressure (PCSF). PCSF was monitored via a needle in the great cistern, known from previous studies to be identical to supratentorial CSF-pressure. During elevation of PSSS, large and small pial veins dilated by 14 +/- 6.1% from resting diameter. Small arteries remained unresponsive until they were dilated by 9 +/- 2.1% at the level of PSSS 50 mmHg. Large arteries dilated by 18 +/- 5.5% at the level of PSSS 50 mmHg. PSSS was always approximately twice as high as PCSF during increase of PSSS. During reduction of PCSF to -5 mmHg, pial veins also dilated, by 7.4 +/- 1% on the average. This observation suggests that normal PCSF is a result of mainly venous vascular pressure, and that the level of normal venous pressure is not dictated by PCSF but by the function and architecture of the cerebral vasculature. Since the rapid reduction of cerebral perfusion pressure CPP by elevation of venous pressure does not induce autoregulatory adjustment according to the level of CPP, but to the level of arterial transmural pressure, it is concluded, that the basic mechanism underlying autoregulation of CBF is myogenic.

Acute effect of neodymium yttrium aluminium garment laser on the cerebral cortical structure, blood-brain barrier, and pial vessel behaviour in the cat.

Experimental brain lesions were created by Nd:YAG laser (wave length 1.06 microns) irradiation on the cerebral cortex of anaesthesized adult cats with 20 Watts impacts of 0.5, 1.0, 2.5, and 5.0 seconds exposure time through cranial windows. Histological changes, disruption of the blood-brain barrier (Evans blue extravasation) and pial vessel reaction (large vessels more than 100 microns and vessels smaller than 100 microns) were studied under constant PaCO2, blood pH, and mean arterial pressure. Histological changes of the lesions consisted of a zone of dense coagulation, a pale zone of homogeneous coagulation and an oedematous zone. Evans blue extravasation was uniformly seen extending from the histologically changed area into the surrounding tissue in all experiments. Pial arteries in the area with morphological changes showed pronounced dilatation (100.0 +/- 7.2%) and one third of these arteries were closed by thrombi. Pial arteries in the area of Evans blue extravasation but outside of histological changes also dilated (large arteries 60 +/- 4.1%, small arteries 77 +/- 5.9%). Pial arteries outside of the Evans blue extravasation were affected transiently and only in a very small zone: Within a distance of 200 microns from the Evans blue extravasation, large arteries initially dilated by 41 +/- 8.3%; small arteries dilated within 400 microns (42 +/- 3.7%). Within 4 minutes after irradiation arterial dilatation was again significantly reduced (p less than 0.01). It is concluded that no important vascular changes occur beyond the zones of histologically altered brain tissue.

Hydroxyethyl-starch in transient experimental focal cerebral ischemia.

In a model of focal cerebral ischaemia in the cat (transorbital occlusion of the middle cerebral artery for 60 minutes, thereafter 6 hours reperfusion by clip removal), hydroxyethyl-starch (HAES) (ELOHES; Leopold Pharma GmbH, Graz, Austria) was administered intravenously before and during the ischaemic episode as a 6% or as a 10% solution in a randomised manner (6 animals each group). The size of the developing cerebral infarct was not significantly different when comparing the 6% and the 10% group with the controls (SALINE). Collateral circulation to the infarct border (pial arteries on the suprasylvian gyrus) was also not significantly different between the two groups, except for the first hour of reperfusion, where vessels of the 6% group were wider than vessels of the 10% group. At the infarct border (ectosylvian gyrus) small resistance vessels were significantly more dilated in the 6% than in the 10% group both during the occlusion period and during the reperfusion episode after removal of the clip. Pial arteries dilated less in both HAES-groups than in the controls. It can be assumed, that HAES-incuded decrease of plasma viscosity led to an elevation of blood flow velocity and blood flow quantity (CBF). But the latter might be counteracted by autoregulation of CBF, i.e. vasoconstriction. Thus, a possible positive effect of HAES might in part be counteracted by autoregulation, which explains that no significant therapeutic effect could be achieved.

Disturbed cerebrospinal fluid circulation after subarachnoid hemorrhage and acute aneurysm surgery.

In 138 patients with ruptured cerebral aneurysms operated on within 48 to 72 hours after subarachnoid hemorrhage, an external ventricular drainage catheter was inserted before craniotomy and was used intermittently during the first week after surgery. In 51 patients, intracranial pressure (ICP) was measured intraoperatively. The majority of patients showed increased ICP intraoperatively irrespective of the preoperative Hunt and Hess grade and the amount of subarachnoid blood accumulation or intraventricular blood clot. Intraoperative drainage of cerebrospinal fluid allowed easy access for aneurysm dissection by making the brain slack in more than 90% of patients. Postoperative ICP measurements revealed that significant brain swelling did not occur in the majority of patients. In 7 patients, persistently elevated ICP (greater than 20 mm Hg) was recorded. Nine patients (8%) developed shunt-dependent hydrocephalus; all of these patients had suffered an intraventricular hemorrhage. Measurements of the volumes of cerebrospinal fluid drained did not allow prediction of shunt-dependent hydrocephalus.

Practical application of intraoperative ultrasound imaging.

Intraoperative real-time ultrasound imaging (US) was used in over 500 patients to investigate which of the previously considered applications are of practical use in everyday neurosurgery. During all intraoperative applications for a wide variety of pathological conditions, small deep-seated as well as subcortically located lesions were detected with accuracy; in many instances they could be approached through smaller surgical exposures. US guidance was accurate in 209 cases for the stereotactic introduction of needles or endoscopes into various lesions: thus burrhole evacuation was performed on 148 intracerebral haematomas; in 16 patients endoscopic biopsy and resection of ventricular tumours was performed as well as biopsies of 39 hemispheric brain tumours and aspiration of a brain abscess in 6 instances. Seventy six gliomas were investigated by US imaging; the frequently unclear boundary between tumour and surrounding oedema was not better visualized than with CT or MRI. Moreover, US imaging for the detection of residual tumour towards the end of operation was unreliable. Postoperative imaging through burrholes or other cranial defects was mostly of unacceptably low quality for therapeutic decision making.

Intraoperative ultrasound (US) imaging. Comparison of pathomorphological findings in US and CT.

Experience with the use of intraoperative ultrasound (US) imaging in over 300 patients are presented in this paper and discussed with special reference to various pathomorphologies as well as their identifiability within the brain/intracranium. In 201 of these patients, the pathomorphological peculiarities in US could be compared with preoperative CT findings. As a general result, all investigated lesions could be identified during intraoperative US investigations with the exception of small aneurysms. Most of the lesions gave at least partly higher echosignals than normal brain tissue, except arachnoid cysts. Size and shape of lesions were comparable in US and CT with the exception of some gliomas; in the latter group, the diffuse image in US was more akin to the situation likely to be found by the surgeon during operation, whereas CT used to give a misleading picture of a more or less clearly delineated tumour. US allowed more accurate differentiation between intratumoural necrosis and cysts than CT: the latter was misleading in many instances. At the present state of development, real time US imaging does not allow a histopathological diagnosis. The ease of handling and the high quality of morphological imaging, however, warrant a number of practical applications in daily neurosurgical practice.