Vascular and Cerebrospinal Supply of the Spinal Cord

Wise Young, Ph.D., M.D.
W. M. Keck Center for Collaborative Neuroscience
Rutgers University, Piscataway, New Jersey 08854-8082

Started 13 January 2000, Last updated 23 January 2002

The anatomy of the spinal arterial and venous systems explain many of the symptoms and causes of spinal cord ischemia, a condition that affects thousands of people world wide. For example, one of the most important causes of spinal cord ischemia is occlusion of the aorta, the primary source of blood flow to the thoracic and lumbar spinal cord. A recent study suggests that the lumbosacral spinal cord is sensitive to prolonged periods of very low blood pressure. The anterior spinal artery, a thin and tortuous artery, supplies over 50% of the blood flow to the lower cervical and thoracic spinal cord; this artery is vulnerable to occlusion and inflammation. In this article, I will try to describe the vascular anatomy and cerebrospinal fluid supply of the human spinal cord, and consequences of obstructing these systems.

Spinal Cord Arteries and Ischemia

The spinal cord receives its blood supply from two main sources: the anterior spinal artery (ASA) and the posterior spinal arteries (PSA). The ASA is located along the anterior midline of the spinal cord and supplies over 50% of the blood flow to the spinal cord, including the anterior spinal cord and most of the gray matter in the spinal cord. The two PSA arteries runs longitudinally along the posterior surface of the spinal cord, alongside the spinal roots; they supply most of the blood flow to the posterior spinal cord.

In the cervical spinal cord, the ASA receives its blood flow from the vertebral arteries that come from the subclavian artery, enter the spinal canal in the upper cervical vertebral column, ascend to the anterior midline of the brainstem, and merge to become the basilar artery. Two small branches of the vertebral artery descend towards the anterior midline, merge, and form the ASA which descends to the lumbar cord.

Figure 1. A diagram of the anterior arteries of the spinal cord and brainstem (from, showing the (1) vertebral arteries, (2) anterior spinal artery, (3) posterior inferior cerebellar artery, (4) basilar artery, (5) anterior inferior cerebellar artery, (6) pontine arteries, and (7) posterior cerebellar artery. The spinal cord is the tubular structure with the butterfly shaped gray matter at the bottom. The winged structure is the cerebellum. Part of the carotid artery can be seen at the top of the picture. This blood flow system supplies most of the blood to the midbrain, brainstem, cerebellum, and cervical spinal cord. Hence, it has many sources of blood. At the level of the brain, there are two carotid arteries. At the level of the spinal cord, there are two vertebral arteries connected by the basilar artery. Two branches of the vertebral artery descend to join in the midline to form the ASA, a thin artery that runs from the cervical spinal cord to the lower lumbar cord.

The vasculature of the upper cervical spinal cord can be better appreciated from a three-dimensional perspective, as shown in figure 2. The following pictures are available from, a wonderful source of superbly rendered and accurate pictures of the anatomy of the human body. The vertebral arteries receive blood from the subclavian artery. Two branches of the vertebral artery head downward and merge to form the anterior spinal artery which courses down the anterior midline of the spinal cord. Note that the vertebral arteries also send two branches that become the posterior spinal arteries.

Figure 2. A 3-dimensional view of the anterior side of the brainstem and spinal cord of human (from The picture clearly illustrates the two vertebral arteries, the merger of the two vertebral arteries to form the basilar artery, the small branches of the vertebral artery that form the anterior spinal and the posterior spinal arteries, the posterior inferior cerebellar arteries, the anterior inferior cerebellar artery (not labelled) and the posterior cerebellar artery (not labelled). The brain stem can be clearly seen with two lobular protruberances seen on the anterior surface or the pons (not labelled). The midbrain is the larger structure on the upper end of the brainstem. The cerebellar hemispheres can be seen as two large winged structures on the side.

In the thoracic and lumbar spinal cord, segmental blood supply comes from the intercostal or lumbar arteries. The following figures illustrate the segmental arterial supply. Figure 3 shows a lateral view of the intercostal and lumbar arteries. Figure 4 shows a cut-away view of the spinal canal and the various branches of the radicular branch. Figure 5 shows a special intercostal artery, the artery of Adamkiewicz which supplies blood to the lower half of the anterior spinal artery. Figure 6 shows how this artery enters the spinal canal, courses up the spinal cord to merge with the anterior spinal artery. Figure 7 illustrates the relationship of the anterior and posterior spinal arteries.

Figure 3. Segmental arteries. This lateral view of the thoracic and lumbar spinal column shows the anterior longitudinal ligament (on the right) that holds the vertebral bodies together in the front, the spinal canal and the spinal cord in the canal, the foramina or openings of the spinal canal through which spinal roots emerge (not shown), and the posterior vertebral processes (on the left). The intervertebral discs are depicted in white while the spinal cord is lighter pink. The blood vessels are in red.

Coming directly from the aorta, the segmental arteries are called intercostal arteries in the thoracic region and lumbar arteries in the lumbar region. The thoracic vertebra are distinguished by the lateral protruding processes that interface with ribs.

Intercostal and lumbar arteries are networked longitudinally with each other both anteriorly and posteriorly. Thus, each intercostal or lumbar artery may supply the vertebral structures above and below the segmental level, providing a significant degree of redundant supply so that occlusion of one or more branches will not lead to ischemia of the vertebral column.

Small branches of each segmental arteries penetrate into vertebral bone through small openings. One branch heads posteriorly to supply the posterior vertebral processes. Another branch enters the spinal canal through the foramina between the vertebral levels. These segmental arteries are responsible for supplying blood flow to structures around and inside the spinal canal at each segmental level.

Figure 4. Branches of the segmental artery. Each intercostal or lumbar artery splits into several branches as it comes around the vertebral body (see figure 4). One large branch heads posteriorly where it supplies the posterior vertebral process. One branch enters the foramen alongside the spinal root and splits into dural branches and a radicular artery. At most levels, the radicular artery connects to the posterior spinal artery. The dural artery supplies the dura and, in the lower spinal cord and cauda equina, these arteries supply the roots and the dural sleeves.
Figure 5. The radicular arteries. The arteries that enter the spinal cord through the roots typically split into two branches. One branch join with posterior spinal arteries and is called radiculopial artery. However, one radicular artery that enter the spinal canal at T9-L1 levels, usually from the left (opposite of what is depicted here), forms a special radiculomedullary artery or the Artery of Adamkiewicz (also called the arteria radicularis anterior magna. The artery takes a characteristic hairpin turn as it enter through a spinal root, ascends on the anterior surface of the spinal cord, and merges with the anterior spinal artery, usually at T4-T6.
Figure 6. Pial and sulcal arteries.

The anterior spinal artery runs along the anterior midline of the spinal cord, in the anterior sulcus. The anterior sulcus usually extends close to the central canal (depicted but not labelled) in the center of the spinal cord. The radiculomedullary artery joins the anterior spinal artery. Branches of the anterior spinal artery penetrate into the spinal cord, forming the sulcal arteries that supply the anterior white matter and most of the gray matter of the spinal cord.

The radiculopial artery supplies the posterior spinal arteries that run medial to the posterior sensory rootlets that form the spinal roots. The radiculopial arteries contribute blood flow to a pial plexus on the spinal cord surface, sending arteries that penetrate into the spinal cord, particularly the posterior and lateral columns, as well as the dorsal horns.

Together these two longitudinal arterial systems supply the blood flow for the spinal cord. The two systems communicate with each other at several levels. The radiculomedullary artery contributes to the pial plexus and some anterior sulcal arteries may anastomose with pial arteries. Thus, when the radiculomedullary artery is occluded, some retrograde blood flow may be able to come from the posterior spinal artery. This may account for the relatively low incidence of spinal cord ischemia that leads to paraplegia in humans. Of course, blood flow from above, especially in the upper thoracic and lower cervical spinal cord may prevent ischemia.

In humans, the thoracic ASA receives much of its blood supply from the Artery of Adamkiewicz (sometimes called the Arteria Radicularis Magna) which arises from a left inferior intercostal artery or the first lumbar artery, typically enters the spinal cord through the left T9-11 foramen in 83% of people, takes a characteristic hairpin curve (see figure 4), and ascends to T4 to T6 to anastomose with the ASA. The ASA supplies the central gray matter and ventral white matter. This artery arises from T9-T12 about 75% of the time, T6-T8 15% of the time, L1-L2 about 10% of the time, and from the left side 80% of the time (see There may be as many as 16 other anterior radiculomedullary arteries (average 6-10) but none are as large as the Artery of Adamkiewicz. By contrast, there are 10-23 posterior radicular arteries or radiculopial arteries.

Although traditional neurosurgical dogma holds that the Artery of Adamkiewicz is provides most of the blood supply for the anterior spinal cord, Griep, et al. (1996) provides some data to refute this notion. They examined all patients that had undergone resection of thoracic and thoracoabdomenal aneurysms at Mount Sinai Hospital in New York from 1993-96 and that had somatosensory evoked potential monitoring while each intersegmental vessel were sequentially clamped. They were doing this procedure to assess which of the vessels could be sacrificed without causing spinal cord damage during the aneurysm repair. They found that no patient with less the 10 intersegmental artery severed developed paraplegia and that spinal cord ischemia were reversible in three patients when they improved perfusion by increase blood pressure and vasodilation. They suggest that spinal cord blood flow does not depend on a single artery of Adamkiewicz.

Spinal Cord Veins and Arteriovenous Malformations

The venous system of the spinal cord is important for several reasons. First and foremost, venous drainage is an important determinant of blood flow. Compromise of venous flow may result in delayed ischemia of the spinal cord. Second, enlarged veins are the primary manifestation of arteriovenous malformations where an artery is abnormally connected to a vein. When this happens the veins become enlarged and tortuous with a corkscrew pattern. The veins may become so massive that they cause spinal cord compression. More important, the arteriovenous malformation steals blood from the spinal cord, making it ischemic.

Figure 7. Spinal veins. The venous drainage of the spinal cord follows the arterial system. Like the arterial system, the veins are also segmentally organized into anterior and posterior systems. The anterior spinal vein is situated in the anterior sulcus, adjacent to the anterior spinal artery while the posterior spinal veins are situated close to the posterior spinal arteries. The veins form a pial plexus that drain blood from the posterior, lateral, and anterior spinal cord. The venous system, like the arterial system, has many redundant and alternative flow patterns that can withstand occlusion of one or more veins. In addition to the radiculopial veins and radiculomedullary vein, some of the veins may form anastomoses with each other in gray matter.

Arteriovenous malformation form when arteries directly connect to veins. When this happens, blood from the arteries go directly into veins, often leading to huge tortuous venous sinuses that can get to such size that they compress the spinal cord. Likewise, venous congestion can easily occur with increases in central venous pressure. These tortuous corkscrew shaped veins are a cardinal sign of arteriovenous malformations. Venous occlusion can occur, often occur hours or even days after an embolization procedure to treat arteriovenous malformation. The resultant ischemia may lead to significant spinal cord damage.

Figure 8. Human arteriovenous malformation. This picture is from John Ratliff, MD and Edward Connolly, MD (with permission, Department of Neurosurgery; Louisiana State University Medical Center and Ochsner Clinic, New Orleans, Louisiana, see their web site). The pictures shows enlarged tortuous venous sinuses with a characteristic corkscrew pattern. The dura has been opened (held with sutures) and the view is of the translucent arachnoid surface of the spinal cord. The large darkish red vessels are the enlarged veins. Note the presence of smaller posterior spinal arteries (brighter red due to oxygenation) along the lateral edges of the spinal cord.

Castro-Moure, et al. (1997) proposed a pathophysiological classification of human spinal cord ischemia. They divided vascular myelopathies into acute and chronic syndromes and then subdivided these further by the type of vascular damage: arterial, venous, or mixed. Among the venous syndromes is a condition called "congestive myelopathy" formerly referred to as varicosis spinalis or Foix-alajouanine syndrome caused by a spinal dural arteriovenous fistula, as described by Koch, et al. (1998). This syndrome occurs in predominantly male patients over 60 years of age, manifesting with increasingly paretic gait that is symmetrical and progress from distal to proximal, alongside the development of a sensory level, bladder and bowel incontinence. This is associated with increased magnetic resonance signal from gray matter, slight swelling of the affected region, and even frank infarction. Spinal angiography usually reveals meningeal arteries draining from a fistula into the spinal venous system, producing congestion and slowing of venous outflow.

Borden, et al. (1995) described a classification system of arteriovenous fistulous malformations (AVFM) into three types: Type I dural AVFM that drain directly into dural venous sinuses or meningeal veins, Type II malformations that drain into dural sinuses or meningeal veins but also have retrograde drainage into subarachnoid veins, Type III malformations that drain into subarachnoid veins without dural sinus or meningeal venous drainage. Heroes, et al. (1986) has proposed a addition Type IV malformation where an intrinsic spinal artery feeds the arteriovenous malformation. Rosenblum, et al. (1987) described 81 patients with spinal arteriovenous malformations: 33% had dural AV fistulas, 67% had intradural AVM's. The latter tended to occur in younger patients, were high flow lesions, and were associated with arterial of venous aneurysms, suggestive of congenital causes. Other vascular malformations include hemangiomas, capillary telangiectasis, and cavernomas (Jellinger, 1986).

Venous congestion is a common presenting sign of arteriovenous fistulas and venous occlusion. In fact, Willinsky, et al. (1990) have suggested that if the venous phase of the spinal circulation is normal, this alone rules out the presence of dural arteriovenous fistulas. Interrupting the arteriovenous fistula generally reverses the venous congestion and neurological symptoms (Muraszko & Oldfield, 1990). Venous congestion also occurs with other types of arteriovenous malformations, including hemangioblastoma (Ohtakara, et al. 2000). It may even occur with occlusions or abnormalities of the vena cava system that the spinal veins drain into (Tsulade, 1999) or with paraspinal arteriovenous malformations (Hui, et al., 1994). In fact, increased venous pressure associated with hyperventilation, abdominal compression, and combined hyperventilation and abdominal occlusion causes engorgement of the epidural venous plexus in normal human volunteers (Lee, et al. 2001).

Cerebrospinal Fluid Supply

The spinal cord is bathed in cerebrospinal fluid (CSF). Secreted by the choroid plexus in lateral, III, and IV ventricles in the brain, CSF percolates down the spinal cord in the space between the pia and dura mater. The average human has 80-150 ml of CSF, 20% of which is located in the brain ventricles, 20% in the subarachnoid space (underneath the pia), and 60% in the lumbar cisterns of the spinal cords. The choroid plexus makes approximately 500 ml of CSF per day, sufficient to replace the CSF contents by 3-4 times during the day. The spinal subarachnoid space may take up to as much as 25% total CSF uptake (Bozanovic-Sosic, et al. 2001).

CSF flows from the choroid plexus into the lateral ventricles, through the interventricular foramen of Monroe, into the third ventricles, out the cerebral aqueduct of Sylvius, and into the fourth ventricle. It then moves out the foramen of Lusha into the pontine cisterns and cisterna magna (the spaces below and above the brainstem and upper cervical spinal cord). The latter is a common place to insert a needle to collect cerebrospinal fluid when a lumbar puncture is contraindicated. Figure 9 shows a diagram of the different cisternal spaces and their relationship to the spinal cord subarachnoid space.

In the human, the dura is thick and opaque whereas. The arachnoid is thin and translucent. Note the presence of small trabeculoid connections between the arachnoid mater and the dura mater. The CSF occupies the subarachnoid space. The pia mater of the spinal cord generally forms fine adhesions with the dura at the sides of the spinal cord. These are called denticulate ligaments at each segmental level. Figure 10 illustrates the subarachnoid and subdural spaces of the spinal cord. The CSF also surrounds branches of the arachnoid blood vessels that penetrate into the spinal cord. These CSF spaces are called Virchow-Robbins space.

When a person is lying down, the CSF pressure is 4-16 mm Hg; the pressure of course increases as the person sits up, since the pressure reflects the column of fluid. CSF pressure is also influenced by venous pressure and typically pulsates with breathing and heartbeats. Average CSF movement in the posterior spinal subarachnoid space is towards the tail while the average CSF movements in the anterior spinal space and central canal tend to be toward the head. Therefore, intrathecally administered drugs in the posterior subarachnoid space move downward towards the caudal (tailward) spinal cord and then back towards the rostral (headward) end of the cord.

Figure 9. Cerebral ventricles, cisterns, and their relationship with spinal CSF spaces. CSF is produced by choroid plexus in the third ventricle, where it moves through the cerebral aqueduct of Sylvius into the fourth ventricle, exiting at the median aperture of fourth ventricle to the cisterna magna, located posterior and caudal to the brainstem and cerebellum. The CSF in the fourth ventricle also goes into the spinal canal. CSF goes down the posterior subarachnoid space of the spinal cord and returns to the brain through the anterior subarachnoid space, going past the pons to the interpeduncular cistern, the cranial subarachnoid space where the CSF is taken up into the superior sagittal sinus. CSF in the cisterna magna move into the cranial subarachnoid space. (adapted from )
Figure 10. A sagittal view of the human thoracic spinal cord, showing the (1) intervertebral discs, (2) vertebral bodies, (3) dura, (4) epidural space, (5) spinal cord, and (6) subdural space. A thick ligament, the anterior longitudinal ligament forms the anterior border of the spinal column. The posterior longitudinal ligament form the posterior border of the vertebral bodies.

In this particular case, there is apparently some protruberance of two ridges that are indenting the anterior spinal cord at two levels. At the upper compression level, there may be an enlargement of the central canal and a possible discoloration of the spinal cord at the upper compression. The discs are quite thin. This picture does not illustrate the separation of the dura and the arachnoid. Frequently, in postmortem specimens, due to CSF loss, the arachnoid collapses and adheres to the pia mater. The subdural space is usually apparent, however, due to the stiffness of the dura. The subarachnoid space, when filled with CSF, usually pushes the arachnoid up against the dura. The arachnoid mater contains granulations which can take up CSF. A better view of the subarachnoid and subdural spaces can be see below

(source: The Anatomy Project ).

Figure 11. A posterior view of a dissected human spinal cord. A laminectomy has been carried out to expose the T6-9 spinal cord. The dura and arachnoid mater has been removed to reveal a portion of the T6-7 cord. The thin filmy arachnoid mater was left at T8 and the thicker dura mater at T9. The pia mater (1) is on the surface of the spinal cord. The denticulate ligament (2) and the motor roots (3) are readily visible, as well as a plexus of posterior veins on the surface. The corresponding subarachnoid space (5) and the subdural space (6) can be clearly seen. Some epidural fat can be observed on the lower end of the exposed dural surface (labelled 8) on the inset diagram, as well ad the ligamentum flavum (9) which represents the posterior surface of the spinal canal. (source: Virtual Hospital).

Figure 12. Diagram of the dura mater, arachnoid, and spinal cord artery. The dura mater is the outer most layer. The arachnoid is next. Cerebrospinal fluid (CSF) located in the space between the arachnoid and pia mater. Thin strands of connective tissues, the arachnoid trabeculae connect the arachnoid and pia. The pial artery enters the tissue through the Virchow-Robbins space. CSF percolates through the Virchow-Robbins space, through the tissue, into the central canal. (source: Wise Young)

CSF enters the spinal cord through the Virchow-Robbins space that surrounds blood vessels penetrating into the cord, diffuse to the spinal canal where it is taken up, and flow upward towards the brain. Trauma frequently interrupts the central canal. The resultant pooling of CSF can lead to a formation of a large cyst at the injury site. In rats, the ependymal lining is frequently absent at the injury site and proliferating ependymal and other cells can fill the cyst with a loose cellular matrix upon which thousands of axons may grow (Beattie, et al. 1997), coming from both central and peripheral sources (Hill, et al. 2001).

The central canal above and below the injury site enlarge, forming syringomyelic cysts similar to what people develop after occlusion of CSF flow at the foramen magnum (Arnold-Chiari syndrome). These cysts or syrinxes can develop in as many as 20% of people after spinal cord injury (Nielson, et al. 1999). Few true incidence studies have been reported. El Masry, et al. (1996) reviewed 815 consecutive cases of spinal cord injury seen between January 1990 and December 1992 and found 28 patients with post-traumatic syringomyelic cysts (3.43%). The incidence was twice as common in patients with complete spinal cord injury than incomplete injuries. The cysts were detected as early as 6 months and as late as 34 years.

Syringomyelic cysts are a frequent complication of scoliosis. Hanieh, et al. (2000), for example, examined 35 children with scoliosis and found that 7 had syringomyelia and that 6 had Chiari malformations. Gupta, et al. (1999) likewise examined 25 patients (13 boys, 12 girls) with idiopathic scoliosis and found that 7 (28%) had syringomyelic cysts and that 5 had Chiari malformation. Emory, et al. (1997) reviewed 25 patients who were admitted with scoliosis and syringomyelia and found that 19 had chiari malformation. Tokunaga, et al. (2001) reviewed 27 scoliotic patients with syringomyelia and found that a majority showed a decrease by 50% or more in cyst size, suggesting that spontaneous shrinkage of syringomyelia is not unusual. These and other experiences suggest that the scoliosis may be secondary to syringomyelia and that many of these cases may not be idiopathic scoliosis.

Blockade of CSF flow in the subarachnoid space may cause syringomyelic cysts. In the posterior subarachnoid space, the pulsatile flow of CSF on average tend to flow downward (tailward) while it flows upward (headward) in the anterior subarachnoid space. If the posterior subarachnoid space becomes occluded due to adhesions between the arachnoid and pia, downward flow may be diverted into the spinal cord and central canal, causing edema (Klekamp, et al. 2001) and forming a syrinx in the proximal cord. If the anterior subarachnoid space is occluded, some of the upward flow may be directed into the central canal and a caudal syrinx may develop (Cosan, et al. 2000). Many of the cysts still show relatively high pulsatile CSF flow within the cyst (Asano, et al. 1996).

Syrinxes may extend many segments either in the rostral or caudal direction. They may cause ascending loss of function if they ascend, increased spasticity and eventually flaccidity in the lower limbs when cyst descends. Several studies have reported pulsatile CSF flow inside syringomyelic cysts (Brugieres, et al. 2000). In animal studies, even when the central canal does not communicate directly with the subarachnoid space (Stoodley, et al. 1999), substantial pulsatile flow may be present in the syringomyelic cyst, suggesting that CSF percolates through spinal tissues to reach the central canal. Human studies suggest that increased spinal cord edema can be seen on MRI (Fischbein, et al. 1999) before the development of a syrinx.

For many years, surgeons shunted syringomyelic cysts. However, shunting only eliminates the cysts temporarily (Schaan & Jaksche, 2001). Shunt obstructions occur in 50% or more of cases and require re-operation (Batsdorf, et al. 1998). Lee, et al. (2000) found that post-traumatic syringomyelic cysts can occur with or without spinal cord tethering but, when tethering is present, removal of the adhesions around the spinal cord and the arachnoid alone collapsed the cyst. If care was taken to prevent readhesion of arachnoid to pia, most patients had no further recurrence of the cyst. Thus, untethering the spinal cord and removing subarachnoid adhesions appears to provide the best opportunity for eliminating syringomyelic cysts with a low rate of recurrence.

Spinal Cord Ischemia

Spinal cord ischemia frequently occurs with occlusion and repair of thoracic aortic aneurysms (Estrera, et al. et. 2001). Unless the surgeons took care to shunt the occlusion and maintain perfusion of the lower spinal cord, a majority of patients became paraplegic after such procedures. The spinal cord ischemia is believed to result from occlusion of the artery of Adamkiewicz, rendering the thoracic spinal cord ischemic at T4. A recent study by Duggal, et al. (2002) suggests that the lumbosacral spinal cord is susceptible to ischemic myelopathy resulting from cardiac arrest.

The causes of ischemic spinal cord injury are often not known. Salvador de la Berrera, et al. (2001), for example, retrospectively reviewed cases of anterior spinal syndromes or aortic surgery or rupture in an active spinal cord injury unit, excluding cases of compressive, tumor, of inflammatory pathology. They found 36 cases of documented spinal cord infarctions. Of these, 36.1% were due to idiopathic causes. Aortic surgery accounted for 25%, systemic arteriosclerosis for 19.4%, and acute perfusion deficit for 11.1%. Although in-hospital mortality was high at 22%, discharge functional status were similar to traumatic spinal cord injury: 57% were wheelchair bound, 25% were ambulatory with supportive devices, and 18% were fully ambulatory.

Spinal cord ischemia is often misdiagnosed as transverse myelitis (TM). Tartaglino, et al. (1996) examined 19 patients with idiopathic transverse myelitis. Seven of 12 patients who had peripheral nerve conduction studies showed signs of peripheral neuropathy. On T2-weighted MRI scans, 13 of 18 patients had holocord abnormal signal intensity with gray matter involvement similar to that seen in spinal cord ischemia. Only 3 (16%) showed evidence of isolated white matter involvement consistent with transverse myelitis. In another study, Jeffrey, et al. (1993) examined 33 cases and classified 45% as parainfectious, 21% associated with multiple sclerosis, 12% associated with spinal cord ischemia, and 21% as idiopathic

Cord compression causes ischemia. Although narrow spinal canal, metastatic tumors, intrathecal hemorrhage (Nolli, et al. 2001), chest compression during cervical spinal surgery in the prone position (Bhardwaj, et al. 2001), cervical spondylosis (Mifsud & Pullicino, 2001), arachnoid cysts (Paramore, 2000) or spinal cord tethering (Kathbauer & Sieler, 1997) are not usually considered ischemia, their mechanism of injury is probably ischemia. Traditionally, neurosurgeons have believes that cord compression causes ischemia by increasing tissue pressure. When tissue pressure approaches blood pressure, blood flow stops. Perhaps the most unappreciated effect of cord is venous obstruction or congestion. Because veins are lower pressure systems, even slight compression may occlude venous drainage of the spinal cord.

Venous congestion is frequently present in mildly compressed spinal cords. For example, Voskuhl & Hinton (1990) described 11 patients with spondylotic compression of the cervical spinal cord, presenting with glove-distribution sensory loss of the hands. Myelography revealed cord compression. Decompression of the spinal cord relieved the symptoms and the authors noted that venous stagnation may have played a role in several cases. Cooper, et al. (1995) did autopsies on patients that required decompressive surgery for herniated intervertebral discs, reporting vascular congestion, dilatation, thrombosis, neovascularization, and endothelial abnormalities in spinal cord vasculature. Hoyland, et al. (1989) noted prominent venous congestion in human cadavers that had intervertebral disc hernation and proposed that venous obstruction may be an important pathogenic mechanism in the development of perineural and intraneural fibrosis. Spinal canal stenosis can also cause venous congestion. For example, Delamarter, et al. (1990) produced lumbar spinal stenosis in dogs and found that 25% stenosis was associated with moderate or severe venous congestion of the root and the dorsal root ganglion of the seventh lumbar nerve.

Venous occlusion can cause spinal cord infarctions. For example, Zhang, et al. (2001) described spinal cord infarction due to ligation of the dorsal spinal vein at T10-13 vertebral levels. One day after ligation, edema and hemorrhage were evident in the dorsal column but axons appear to be well-preserved. By the 3rd day, axonal loss was apparent in the dorsal column, followed by macrophage infiltration and astrocytic gliosis. By 14 days, the spinal cord was atrophic. The lesion was confined to the dorsal funiculus in all cases. Non-hemorrhagic venous infarctions of the spinal cord in humans have long been described (Kim, et al. 1984).

Summary and Conclusions

The spinal cord receives blood supply from multiple sources. Two longitudinal spinal arterial systems receive blood from the vertebral and segmental arteries: the anterior spinal artery (ASA) and the posterior spinal arteries (PSA). The PSA is a long, and tortuous vessel that supplies over 50% of the spinal cord, particularly the gray matter and ventral white matter of the thoracic spinal cord. It receives blood largely from the vertebral arteries at the upper cervical levels and the artery of Adamkiewicz at the lower thoracic levels. In contrast, the PSA runs alongside the posterior spinal roots, receiving blood from the vertebral arteries in the upper cervical cord and the segmental arteries. It supplies the posterior and part of the lateral columns.

Venous drainage of the spinal cord occur mainly through pial veins on the spinal cord surface. Although spinal cord ischemia (loss of blood flow) is usually attributed to restriction of arterial supply, venous congestion probably plays a more important role. This is evident from clinical observation of arteriovenous malformations that invariably produce venous engorgement, spinal neurological deficits associated with venous congestion resulting form arteriovenous malformations outside of the spinal cord, venous occlusion producing spinal cord infarcts in animal models, and mild cord compression causing severe venous congestion in the spinal cord and roots. Some investigators in fact believe that venous congestion is the major cause of spinal cord ischemia under many conditions.

Cerebrospinal fluid emanate from the choroid plexus in the third ventricle, flow down the spinal cord in the subarachnoid space, percolate through the Virchow-Robbins spaces adjacent to blood vessels into the spinal cord, diffuse to the central canal where it heads back up the spinal cord. After injury, obstruction of subarachnoid space due to adhesions or interruption of the central canal can lead to development of syringomyelic cysts that extend both up and down the spinal cord. Enlargements of these cysts can lead to cord compression and ischemia. In the past, most surgeons shunted these cysts with high rates of shunt blockade and cyst recurrence. Recent studies suggest that careful removal of adhesions and reconstruction of the subarachnoid space to allow fluid flow eliminate a large majority of the cysts.

Spinal cord ischemia can result from many causes. Obstruction of arterial blood supply to the spinal cord is the most commonly cited cause of spinal cord ischemia. Ischemic myelopathy of the spinal cord is frequently misdiagnosed as transverse myelitis. Also, many cases of spinal cord ischemia are attributed to other causes. For example, arteriovenous malformations cause ischemia. Compression of the spinal cord also can cause ischemia. Perhaps the most under appreciated cause of spinal cord ischemia is venous obstruction or congestion. Although the venous system is redundant, it is also a low pressure system that is susceptible to even mild compression and elevated venous pressures.

Spinal cord injury due to arterial, venous, and cerebrospinal fluid obstruction is far more common than we think. It is likely that they contribute significantly to the severity of spinal cord injury and failure of recovery in many causes of traumatic spinal cord injury where failure to decompress the injury site for days or weeks, poor maintenance of perfusion pressure, increased central venous pressure, and development of enlarging spinal cysts may contribute to lack of recovery or loss of function in chronic spinal cord injury

References Cited

• Armon C and Daube JR (1989). Electrophysiological signs of arteriovenous malformations of the spinal cord. J Neurol Neurosurg Psychiatry. 52 (10): 1176-81. Summary: A characteristic pattern of electrophysiological changes was found in 24 patients with confirmed spinal cord arteriovenous malformations (AVMs). The AVMs were limited to the thoracic cord in seven, involved the conus and the cauda equina in 10, and involved all levels in six. Of the patients, 88% had at least one definite electrophysiological abnormality: nerve conduction studies showed abnormal results in 43% (10 of 23), needle electromyography in 77% (17 of 22), and tibial somatosensory evoked potentials in 88% (7 of 8). The pattern of involvement was of scattered, multiple, bilateral thoracolumbosacral radiculopathies, consistent with axonal or neuronal destruction, associated with paraspinal fibrillations or abnormal activation of motor unit potentials. Electrophysiological abnormalities were seen in most patients with lower motor neuron clinical deficit. These abnormalities depended on the caudal extension of the AVM, on an arterial supply at T-10 or below, and on the duration of symptoms. In addition to the expected abnormalities in the distribution of the AVM location, four patients had electrical changes at a distance, which may have been due to venous stasis. <> Department of Neurology, Mayo Clinic, Rochester, Minnesota 55905.

• Asano M, Fujiwara K, Yonenobu K and Hiroshima K (1996). Post-traumatic syringomyelia. Spine. 21 (12): 1446-53. Summary: STUDY DESIGN: This study retrospectively analyzed patients who developed post-traumatic syringomyelia secondary to spinal cord injury. OBJECTIVES: To identify an indicator that would predict the outcome of surgical treatment for post-traumatic syringomyelia. SUMMARY OF BACKGROUND DATA: Syrinx-subarachnoid shunting was chosen as a surgical treatment for post-traumatic syringomyelia. No previous study has been published concerning magnetic resonance imaging findings' ability to predict surgical results before surgery. METHODS: Nine patients diagnosed by magnetic resonance imaging with post-traumatic syringomyelia were the subjects of this study. The magnetic resonance imaging findings and surgical results were analyzed. RESULTS: Neurologic deterioration was found in five patients. Ascending dissociated sensory disturbances and motor weakness were noticed to occur characteristically above the level of the spinal cord injury. The other four patients complained of a slight worsening of numbness without displaying neurologic deterioration. The five patients with neurologic deterioration were treated with a syrinx-subarachnoid shunting. Two of the five patients experienced sustained neurologic improvement after a midline myelotomy, which allowed the fluid within the syrinx to spout out and cause the expanded spinal cord to collapse. This was called a "high-pressure syrinx." In these two patients, the preoperative magnetic resonance imaging demonstrated a positive flow-void sign. On the other hand, drainage of the syrinx in the three patients with a negative flow-void sign did not collapse the expanded spinal cord, and the surgical results were considered fair. This was called a "low-pressure syrinx." CONCLUSIONS: Post-traumatic syringomyelia was classified into two types. A preoperative distinction could be made based on the presence or absence of the flow-void sign on a T2-weighted magnetic resonance image. <> Department of Orthopaedic Surgery, Hoshigaoka Koseinenkin Hospital, Osaka, Japan.

• Batzdorf U, Klekamp J and Johnson JP (1998). A critical appraisal of syrinx cavity shunting procedures. J Neurosurg. 89 (3): 382-8. Summary: OBJECT: This study was conducted to evaluate the results of shunting procedures for syringomyelia. METHODS: In a follow-up analysis of 42 patients in whom shunts were placed in syringomyelic cavities, the authors have demonstrated that 21 (50%) developed recurrent cyst expansion indicative of shunt failure. Problems were encountered in patients with syringomyelia resulting from hindbrain herniation, spinal trauma, or inflammatory processes. A low-pressure cerebrospinal fluid state occurred in two of 18 patients; infection was also rare (one of 18 patients), but both are potentially devastating complications of shunt procedures. Shunt obstruction, the most common problem, was encountered in 18 patients; spinal cord tethering, seen in three cases, may account for situations in which the patient gradually deteriorated neurologically, despite a functioning shunt. CONCLUSIONS: Placement of all types of shunts (subarachnoid, syringoperitoneal, and syringopleural) may be followed by significant morbidity requiring one or more additional surgical procedures. <> Division of Neurosurgery, University of California, Los Angeles, USA.

• Beattie MS, Bresnahan JC, Komon J, Tovar CA, Van Meter M, Anderson DK, Faden AI, Hsu CY, Noble LJ, Salzman S and Young W (1997). Endogenous repair after spinal cord contusion injuries in the rat. Exp Neurol. 148 (2): 453-63. Summary: Contusion injuries of the rat thoracic spinal cord were made using a standardized device developed for the Multicenter Animal Spinal Cord Injury Study (MASCIS). Lesions of different severity were studied for signs of endogenous repair at times up to 6 weeks following injury. Contusion injuries produced a typical picture of secondary damage resulting in the destruction of the cord center and the chronic sparing of a peripheral rim of fibers which varied in amount depending upon the injury magnitude. It was noted that the cavities often developed a dense cellular matrix that became partially filled with nerve fibers and associated Schwann cells. The amount of fiber and Schwann cell ingrowth was inversely related to the severity of injury and amount of peripheral fiber sparing. The source of the ingrowing fibers was not determined, but many of them clearly originated in the dorsal roots. In addition to signs of regeneration, we noted evidence for the proliferation of cells located in the ependymal zone surrounding the central canal at early times following contusion injuries. These cells may contribute to the development of cellular trabeculae that provide a scaffolding within the lesion cavity that provides the substrates for cellular infiltration and regeneration of axons. Together, these observations suggest that the endogenous reparative response to spinal contusion injury is substantial. Understanding the regulation and restrictions on the repair processes might lead to better ways in which to encourage spontaneous recovery after CNS injury. <> Department of Cell Biology, Ohio State University College of Medicine, 333 West 10th Avenue, Columbus, Ohio 43210, USA.

• Bhardwaj A, Long DM, Ducker TB and Toung TJ (2001). Neurologic deficits after cervical laminectomy in the prone position. J Neurosurg Anesthesiol. 13 (4): 314-9. Summary: New neurologic deficits are known to occur after spine surgery. We present four patients with cervical myeloradiculopathy who underwent cervical laminectomy, fusion, or both in the prone position, supported by chest rolls. Three patients were intubated and positioned while awake, whereas the fourth patient was positioned after induction. Surgeries were successfully carried out, except for transient episodes of relative hypotension intraoperatively. On recovery from anesthesia, all patients were noted to have new neurologic deficits. Immediate CT myelography or surgical reexploration was unremarkable. All patients improved gradually with administration of high-dose steroids and induction of hypertension. Use of the prone position with abdominal compression may compromise spinal cord perfusion and lead to spinal cord ischemia. The use of frames that prevent abdominal compression, as well as avoidance of perioperative arterial hypotension, is important in maintaining adequate spinal cord perfusion during and after decompressive spinal cord surgery. <> Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.

• Borden JA, Wu JK and Shucart WA (1995). A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment. J Neurosurg. 82 (2): 166-79. Summary: A classification is proposed that unifies and organizes spinal and cranial dural arteriovenous fistulous malformations (AVFMs) into three types based upon their anatomical similarities. Type I dural AVFMs drain directly into dural venous sinuses or meningeal veins. Type II malformations drain into dural sinuses or meningeal veins but also have retrograde drainage into subarachnoid veins. Type III malformations drain into subarachnoid veins and do not have dural sinus or meningeal venous drainage. The arterial supply in each of these three types is derived from meningeal arteries. The anatomical basis of the proposed classification is presented with several cases that illustrate the three types of dural AVFMs. A rationale for the treatment of spinal and cranial dural AVFMs according to their anatomical characteristics is discussed. <> Department of Neurosurgery, New England Medical Center, Boston, Massachusetts.

• Bozanovic-Sosic R, Mollanji R and Johnston MG (2001). Spinal and cranial contributions to total cerebrospinal fluid transport. Am J Physiol Regul Integr Comp Physiol. 281 (3): R909-16. Summary: In this study, we quantified cerebrospinal fluid (CSF) transport from the cranial and spinal subarachnoid spaces separately in sheep and determined the relative proportion of total CSF drainage that occurred from both CSF compartments. Cranial and spinal CSF systems were separated by placement of an extradural ligature over the spinal cord between C(1) and C(2). In one approach, two different radiolabeled human serum albumins (HSA) were introduced into the appropriate CSF compartment by a perfusion system (method 1) or as a bolus injection (method 2). Plasma tracer recoveries in conjunction with a mass balance equation were used to estimate CSF transport. In method 3, catheters connected to reservoirs filled with artificial CSF were introduced into the cranial and spinal CSF compartments. Incremental CSF pressures were established in each CSF system, and the corresponding steady-state flow rates were measured. Total CSF drainage ranged from 0.51 to 0.75 ml. h(-1). cmH(2)O(-1). Expressed as a percentage of the total CSF transport, the ratios of cranial-to-spinal clearance estimated from methods 1, 2, and 3 were 75:25, 88:12, and 75:25, respectively. Primarily on the basis of the data derived from methods 1 and 3, we conclude that the spinal subarachnoid compartment has an important role in CSF clearance and is responsible for approximately one-fourth of total CSF transport. <> Trauma Research Program, Department of Laboratory Medicine and Pathobiology, Sunnybrook and Women's College Health Sciences Centre, University of Toronto, 2075 Bayview Ave., Toronto, Ontario, Canada M4N 3M5.

• Brugieres P, Idy-Peretti I, Iffenecker C, Parker F, Jolivet O, Hurth M, Gaston A and Bittoun J (2000). CSF flow measurement in syringomyelia. AJNR Am J Neuroradiol. 21 (10): 1785-92. Summary: BACKGROUND AND PURPOSE: CSF circulation has been reported to represent a major factor in the pathophysiology of syringomyelia. Our purpose was to determine the CSF flow patterns in spinal cord cysts and in the subararachnoid space in patients with syringomyelia associated with Chiari I malformation and to evaluate the modifications of the flow resulting from surgery. METHODS: Eighteen patients with syringomyelia were examined with a 3D Fourier encoding velocity imaging technique. A prospectively gated 2D axial sequence with velocity encoding in the craniocaudal direction in the cervical region was set at a velocity of +/- 10 cm/s. Velocity measurements were performed in the larger portion of the cysts and, at the same cervical level, in the pericystic subarachnoid spaces. All patients underwent a surgical procedure involving dural opening followed by duroplasty. Pre- and postoperative velocity measurements of all patients were taken, with a mean follow-up of 10.2 months. We compared the velocity measurements with the morphology of the cysts and with the clinical data. Spinal subarachnoid spaces of 19 healthy subjects were also studied using the same technique. RESULTS: A pulsatile flow was observed in syrinx cavities and in the pericystic subarachnoid spaces (PCSS). Preoperative maximum systolic cyst velocities were higher than were diastolic velocities. A systolic velocity peak was well defined in all cases, first in the cyst and then in the PCSS. Higher systolic and diastolic cyst velocities are observed in large cysts and in patients with a poor clinical status. After surgery, a decrease in cyst volume (evaluated on the basis of the extension of the cyst and the compression of the PCSS) was observed in 13 patients. In the postoperative course, we noticed a decrease of systolic and diastolic cyst velocities and a parallel increase of systolic PCSS velocities. Diastolic cyst velocities correlated with the preoperative clinical status of the patients and, after surgery, in patients with a satisfactory foraminal enlargement evaluated on the basis of the visibility of the cisterna magna. CONCLUSION: CSF flow measurement constitutes a direct evaluation for the follow-up of patients with syringomyelic cysts. Diastolic and systolic cyst velocities can assist in the evaluation of the efficacy of surgery. <> Centre Inter-Etablissements de Rechereche en Resonance Magnetique, Bicetre Hospital, Paris Sud University, France.

• Cooper RG, Freemont AJ, Hoyland JA, Jenkins JP, West CG, Illingworth KJ and Jayson MI (1995). Herniated intervertebral disc-associated periradicular fibrosis and vascular abnormalities occur without inflammatory cell infiltration. Spine. 20 (5): 591-8. Summary: STUDY DESIGN. Prospective histologic comparison of perineural tissues from patients requiring decompression surgery for herniated intervertebral disc with those from cadaveric controls. OBJECTIVES. To examine the significance of herniated intervertebral-disc-associated perineural vascular and fibrotic abnormalities with respect to back pain symptom generation. SUMMARY OF BACKGROUND DATA. Previous cadaveric studies have demonstrated perineural vascular congestion, dilatation, and thrombosis and perineural and intraneural fibrosis occurring in association with herniated intervertebral disc. It was suggested that these neural abnormalities were the result of ischemia, due to venous outflow obstruction, and also represented a possible cause of ongoing back pain symptoms. Criticisms of such a conclusion arose, however, because the possibility could not be excluded that these abnormalities were the result of postmortem artifact. METHODS. Histologic and immunohistochemical comparison of discal and peridiscal tissues removed from 11 patients with radiographically proven herniated intervertebral disc requiring decompressive surgery and from 6 fresh cadavers without history of back pain in life. RESULTS. Histology and immunohistochemistry of perineural and extraneural tissues from patients revealed vascular congestion, neovascularization, and endothelial abnormalities including luminal platelet adhesion, in association with reductions in von Willebrand factor levels, together with perivascular and perineural fibrosis. Elevated fibrogenic cytokine concentrations were also detected in patients' tissues. These changes occurred without evidence of inflammation and were absent in cadaveric control tissues. CONCLUSIONS. The vascular abnormalities detected in patients may represent an important etiopathologic factor predisposing to intraneural and perineural fibrosis, and hence to chronic pain symptoms, after disc herniation. It seems important to preserve the perineural microcirculation following disc herniation. <> Department of Rheumatology, University of Manchester, United Kingdom.

• Castro-Moure F, Kupsky W and Goshgarian HG (1997). Pathophysiological classification of human spinal cord ischemia. J Spinal Cord Med. 20 (1): 74-87. Summary: Diagnosis of myelopathies of vascular origin is difficult and they are probably underdiagnosed at this time because of the lack of diagnostic tools. A recent report of a 58 year old patient who developed ASAS after an episode of cardiac arrest pointed out the importance of MRI and somatosensory evoked potentials (SEP) to support the diagnosis. MRI with T2 weighted imaging demonstrated diffuse signal abnormalities in both gray matter and surrounding white matter below T7. Furthermore, SEP latencies showed a delay between T6 and T7. Therefore, new technologies including MRI and SEP may improve the diagnosis of spinal cord ischemic injuries. A brief discussion of the normal blood supply of the human spinal cord is presented in this review followed by new, pathophysiologically based classifications of the clinical syndromes of vascular myelopathies. A complete description of the clinical syndromes related to vascular myelopathies is included. Vascular myelopathies were divided into acute and chronic syndromes depending on the time at which the pathophysiological events take place. Subsequently, the two major groups of vascular myelopathies were divided depending on the type of vascular damage, e.g., arterial, venous and/or mixed origin. Posttraumatic spinal cord ischemia is included in the present classification because it is generally considered to be a significant factor contributing to secondary damage following blunt trauma. Since several new diagnostic techniques are now available to characterize the pathology of spinal cord injury, physicians involved in the diagnosis and treatment of vascular myelopathies may find the new classification useful in correlating clinical presentation with subjacent pathology. Identification of the correct pathology should result in more accurate treatment approaches. <> Dept of Anatomy and Cell Biology, Wayne State University School of Medicine, Detroit, MI 48201, USA.

• Cosan TE, Tel E, Durmaz R, Gulec S and Baycu C (2000). Non-hindbrain-related syringomyelia. Obstruction of the subarachnoid space and the central canal in rats. An experimental study. J Neurosurg Sci. 44 (3): 123-7. Summary: BACKGROUND: The aim is to determine the mechanism of non-hindbrain-related syringomyelia in experimental models. The effects of obstruction of central canal and subarachnoid space on occurrence of cavities were discussed. METHODS: 31 Sprague-Dawley rats were used with eight (Group D) as a control. In 10 rats (Group A) 1.5 microl kaolin was microinjected into the dorsal columns and central gray matter of the spinal cord at the level of Th6-10. In 10 rats (Group B) 0.1 cc kaolin was injected into the subarachnoid space at the same level. In 3 rats (Group C), 1.5 microl kaolin was administered into both dorsal midline of the spinal cord and the subarachnoid space. RESULTS: In Group A, histological examination revealed cystic cavity and dilatation of the central canal in five rats; denuded ependymal line and multicystic formations in ependymal and periependymal areas in seven rats. In Group B, denuded ependymal line in three rats and microcystic formations in ependymal and periependymal areas in four rats were revealed. In Group C, there were microcystic formations in two rats and syrinx cavity in one rat. CONCLUSIONS: Developments leading to occurrence of cavities are focused on the central canal in all groups. These models indicate that the CSF-flow is from the subarachnoid space to the central canal leading to changes of cavities. In cases of obstruction of the subarachnoid space or the central canal, the occurrence of syrinx cavity initially is due to increased CSF (cerebrospinal fluid) pressure in the central canal. Flow changes in spinal cord is indicated by this study. <> Department of Neurosurgery, Osmangazi University, Ekisehir, Turkey.

• Delamarter RB, Bohlman HH, Dodge LD and Biro C (1990). Experimental lumbar spinal stenosis. Analysis of the cortical evoked potentials, microvasculature, and histopathology. J Bone Joint Surg Am. 72 (1): 110-20. Summary: An animal model of lumbar spinal stenosis was developed in which the pathophysiology of this condition could be examined. Four experimental groups, each containing six dogs, were studied. One group had a laminectomy of the sixth and seventh lumbar vertebrae only; these animals served as controls. In the three other groups, a laminectomy was performed and the cauda equina was constricted by 25, 50, or 75 per cent to produce chronic compression. Cortical evoked potentials were recorded preoperatively, immediately after constriction, and at one, two, and three months postoperatively. Daily neurological examinations were carried out, and the neurological deficits were graded using the Tarlov system. After three months of constriction, the cauda equina of three dogs in each group was examined histologically, and the vascular circulation was examined by latex and India-ink injection with a modification of the Spalteholz technique. The animals in the control group showed no neurological abnormalities, no changes in cortical evoked potentials, normal microvascularity, and no histopathological changes in the nerve roots or the spinal cord. The dogs in which the cauda equina had been constricted 25 per cent had no neurological deficits, mild changes in cortical evoked potentials, slight histological changes, and venous congestion of the root and dorsal root ganglion of the seventh lumbar nerve. The dogs in which the cauda equina had been constricted 50 per cent had mild initial motor weakness, major changes in cortical evoked potentials, edema and loss of myelin in the root of the seventh lumbar nerve, and moderate or severe venous congestion of the root and dorsal root ganglion of the seventh lumbar nerve. The dogs in which the cauda equina had been constricted 75 per cent had significant weakness, paralysis of the tail, and urinary incontinence; two dogs recovered by the third month, but all had neurogenic claudication for three months. All six dogs had dramatic changes in cortical evoked potentials and had complete nerve-root atrophy at the level of the constriction. There was blockage of axoplasmic flow and wallerian degeneration of the motor nerve roots distal to the constriction and of the sensory roots proximal to the constriction, as well as degeneration of the posterior column. Severe arterial narrowing at the level of the constriction and venous congestion of the roots and dorsal root ganglia of the seventh lumbar and first sacral nerves were also present. Cortical evoked potentials revealed neurological abnormalities before the appearance of neurological signs and symptoms.(ABSTRACT TRUNCATED AT 400 WORDS). <> Division of Orthopaedic Surgery, University of California School of Medicine, Los Angeles 90024-1795.

• Duggal N and Lach B (2002). Selective vulnerability of the lumbosacral spinal cord after cardiac arrest and hypotension. Stroke. 33 (1): 116-21. Summary: Background and Purpose- It is generally accepted that the gray matter in the watershed area of the midthoracic level of the spinal cord is the ischemic watershed zone of the spinal cord. We performed a retrospective study to reevaluate the frequency and distribution of spinal cord injury after a global ischemic event. METHODS: Clinical files and neuropathology specimens of all adult patients with either a well-documented cardiac arrest or a severe hypotensive episode, as well as pathologically confirmed ischemic encephalopathy and/or myelopathy, were reviewed by an independent reviewer. RESULTS: Among 145 cases satisfying selection criteria, ischemic myelopathy was found in 46% of patients dying after either a cardiac arrest or a severe hypotensive episode. Among the patients with myelopathy, predominant involvement of the lumbosacral level with relative sparing of thoracic levels was observed in >95% of cardiac arrest and hypotensive patients. None of the examined patients developed neuronal necrosis limited to the thoracic level only. CONCLUSIONS: Our findings indicate a greater vulnerability of neurons in the lumbar or lumbosacral spinal cord to ischemia than other levels of the spinal cord. <> Department of Clinical Neurological Sciences (Neurosurgery), University of Western Ontario, London, Ontario (N.D.), and Department of Laboratory Medicine and Pathology, University of Ottawa and Ottawa Hospital, Ottawa (B.L.), Canada.

• Emery E, Redondo A and Rey A (1997). Syringomyelia and Arnold Chiari in scoliosis initially classified as idiopathic: experience with 25 patients. Eur Spine J. 6 (3): 158-62. Summary: The authors analysed the clinical and radiological findings and the surgical management of 25 patients admitted for scoliosis classified as idiopathic at first presentation, but in fact associated with spinal cord and/or brain stem anomalies. Twenty patients had syringomyelia, 19 had Chiari malformation. Scoliosis was the only presenting symptom when all these patients were referred to the orthopaedic surgeon. On examination, five patients had normal neurological findings, while the others showed very mild neurological deficits. The diagnosis of syringomyelia and Chiari malformation was established by MRI, which is the best form of neuroradiological examination for discovering spinal abnormalities. Neurosurgical treatment is strongly recommended as the first step in the management of "pseudo" idiopathic scoliosis. <> Department of Neurosurgery, Beaujon Hospital, Clichy, France.

• Estrera AL, Miller CC, 3rd, Huynh TT, Porat E and Safi HJ (2001). Neurologic outcome after thoracic and thoracoabdominal aortic aneurysm repair. Ann Thorac Surg. 72 (4): 1225-30; discussion 1230-1. Summary: BACKGROUND: Neurologic deficit (paraparesis and paraplegia) after repair of the thoracic and thoracoabdominal aorta remains a devastating complication. The purpose of this study was to determine the effect of cerebrospinal fluid drainage and distal aortic perfusion upon neurologic outcome during repair of thoracic and thoracoabdominal aortic aneurysm (TAAA) repair. METHODS: Between February 1991 and March 2000, we performed 654 repairs of the thoracic and thoracoabdominal aorta. The median age was 67 years and 420 (64%) patients were male. Forty-five cases (6.9%) were performed emergently. Distribution of TAAA was the following: extent I, 164 (25%); extent II, 165 (25%); extent III, 61 (9%); extent IV, 95 (15%); extent V, 23 (3.5%); and descending thoracic, 147 (22%). The adjuncts cerebrospinal fluid drainage and distal aortic perfusion were used in 428 cases (65%). RESULTS: Thirty-day mortality was 14% (94 of 654). The in-hospital mortality was 16% (106 of 654). Early neurologic deficits occurred in 33 patients (5.0%). Overall, 14 of 428 (3.3%) neurologic deficits were observed in the adjunct group, and 19 of 226 (8.4%) in the nonadjunct group (p = 0.004). When the adjuncts were used during extent II repair, the incidence was 10 of 129 (7.8%) compared with 11 of 36 (30.6%) in the nonadjunct group (p < 0.001). Multivariate analysis demonstrated that risk factors for neurologic deficit were cerebrovascular disease and extent of TAAA (II and III) (p < 0.05). CONCLUSIONS: The combined adjuncts of distal aortic perfusion and cerebrospinal fluid drainage demonstrated improved neurologic outcome with repair of thoracic and TAAAs. In extent II aneurysms, adjuncts continue to make a considerable difference in the outcome and to provide significant protection against spinal cord morbidity. Future research should focus on spinal cord protection in patients with high-risk extent II aneurysms. <> Department of Cardiothoracic and Vascular Surgery, The University of Texas at Houston Medical School, Memorial Hermann Hospital, 77030, USA.

• Fischbein NJ, Dillon WP, Cobbs C and Weinstein PR (1999). The "presyrinx" state: a reversible myelopathic condition that may precede syringomyelia. AJNR Am J Neuroradiol. 20 (1): 7-20. Summary: BACKGROUND AND PURPOSE: Alteration of CSF flow has been proposed to be an important mechanism leading to the development of syringomyelia. We hypothesize that a "presyrinx" condition attributable to a potentially reversible alteration in normal CSF flow exists and that its appearance may be caused by variations in the competence of the central canal of the spinal cord. METHODS: Five patients with clinical evidence of myelopathy, no history of spinal cord trauma, enlargement of the cervical spinal cord with T1 and T2 prolongation but no cavitation, evidence of altered or obstructed CSF flow, and no evidence of intramedullary tumor or a spinal vascular event underwent MR imaging before and after intervention that alleviated obstruction to CSF flow. RESULTS: Preoperatively, all patients had enlarged spinal cords and parenchymal T1 and T2 prolongation without cavitation. Results of MR examinations after intervention showed resolution of cord enlargement and normalization or improvement of cord signal abnormalities. In one patient with severe arachnoid adhesions who initially improved after decompression, late evolution into syringomyelia occurred in association with continued CSF obstruction. CONCLUSION: Nontraumatic obstruction of the CSF pathways in the spine may result in spinal cord parenchymal T2 prolongation that is reversible after restoration of patency of CSF pathways. We refer to this MR appearance as the "presyrinx" state and stress the importance of timely intervention to limit progression to syringomyelia. <> Department of Radiology, University of California, San Francisco 94143, USA.

• Griepp RB, Ergin MA, Galla JD, Lansman S, Khan N, Quintana C, McCollough J and Bodian C (1996). Looking for the artery of Adamkiewicz: a quest to minimize paraplegia after operations for aneurysms of the descending thoracic and thoracoabdominal aorta. J Thorac Cardiovasc Surg. 112 (5): 1202-13; discussion 1213-5. Summary: All patients undergoing resection of thoracic or thoracoabdominal aneurysms at Mount Sinai Hospital since November 1993 had spinal cord function monitored with somatosensory-evoked potentials as part of a multimodality approach to reducing spinal cord injury. In the segment to be resected, each pair of intersegmental vessels was sequentially clamped, and they were subsequently sacrificed only if no change in somatosensory evoked potentials occurred within 8 to 10 minutes after occlusion. Adjunctive protective measures included mild hypothermia (31 degrees to 33 degrees C), distal perfusion, corticosteroids, maintenance of high normal blood pressures, avoidance of nitroprusside, and cerebrospinal fluid drainage. Ninety-five consecutive patients operated on since 1993 (group II) were compared with 138 earlier patients (group I). Preoperative characteristics such as age, sex, etiology of aneurysm, emergency operation, and reoperation did not differ between groups, nor did operative variables such as incidence of rupture and extent of resection. Group I had slightly more smokers and slightly fewer hypertensive individuals. Group II patients had a significantly better outcome with respect to in-hospital mortality (10.5% vs 18%, p = 0.045) and paraplegia (2% vs 8%, p = 0.008). By multivariate analysis, rupture and diabetes were associated with significantly higher in-hospital mortality, and smoking greatly increased the incidence of paraplegia. The extent of the aneurysm was a major determinant of mortality and paraplegia. The low paraplegia rate in group II was achieved without reattachment of a single intercostal or lumbar artery. No patient with fewer than 10 intersegmental arteries severed had paraplegia, and spinal cord ischemia was reversible in three patients after adjunctive maneuvers were performed to improve perfusion, suggesting that spinal cord blood supply is unlikely to depend on a single "artery of Adamkiewicz." <> Department of Cardiothoracic Surgery, Mount Sinai Medical Center, New York, N.Y 10029, USA.

• Gupta R, Sharma R, Vashisht S, Ghandi D, Jayaswal AK, Dave PK and Berry M (1999). Magnetic resonance evaluation of idiopathic scoliosis: a prospective study. Australas Radiol. 43 (4): 461-5. Summary: The objective of the present study was to determine the incidence of unsuspected intraspinal pathology and to assess the value of atypical clinical features as predictors of these intraspinal pathologies, in patients with idiopathic scoliosis. Twenty-five consecutive patients (13 boys, 12 girls) with idiopathic scoliosis were prospectively evaluated with MRI. Magnetic resonance imaging detected intraspinal pathology in seven patients (28%), which included syringohydromyelia with Chiari I malformation (n = 5), and syringomyelia and dumb-bell neurofibromas in one patient each, respectively. Dural ectasia was also present in five patients. Atypical features, described in the literature as pointers to intraspinal pathologies such as the age < 11 years at presentation, presence of pain, hyperkyphosis, severe curves and the presence of the left thoracic or thoracolumbar curves, were seen to be equally distributed between the two groups (those with and without intraspinal pathologies), thus raising doubts about the importance of these features. <> Department of Radio Diagnosis, All India Institute of Medical Sciences, New Delhi, India.

• Hanieh A, Sutherland A, Foster B and Cundy P (2000). Syringomyelia in children with primary scoliosis. Childs Nerv Syst. 16 (4): 200-2. Summary: The clinical notes of 35 children presenting with scoliosis were reviewed; all 35 had been investigated with MRI. Seven were found to have syringomyelia, and six of these had Chiari malformation. Correction of the syrinx resulted in improvement or stabilisation of the spinal curvature. We recommend that all cases presenting with primary scoliosis should have MRI and should be treated if a syrinx is found. <> Department of Neurosurgery, Women's and Children's Hospital, North Adelaide, South Australia, Australia.

• Hayashi K, Kitagawa N, Takahata H, Yoshioka T, Matsuo Y, Morikawa M, Ochi M, Kaminogo M and Shibata S (2001). [A case of spinal dural AVF treated by endovascular embolization]. No To Shinkei. 53 (4): 381-5. Summary: Here we report a case of spinal dural arteriovenous fistula(AVF) treated by endovascular embolization. A 58-year-old female presented with progressive intermittent claudication and numbness of the lower extremities. MRI showed swelling of the spinal cord with intramedullary high signal intensity on T2-weighted image and intramedullary enhancement, suggested spinal cord myelopathy. Myelography demonstrated the dilated serpentine vessels in the subarachnoid space and focal filling defect. Angiography showed spinal dural AVF fed by bilateral lateral sacral artery. The draining vein was posterior spinal vein. Endovascular embolization using liquid material was performed under general anesthesia. The injection of glue included the distal feeding artery, the shunt itself and the initial part of draining vein. A complete cure was achieved, with a normal postoperative angiogram. MRI returned to normal with complete disappearance of T2 high signal, cord enlargement and enhancement by contrast medium. It was suggested that venous congestion induced the transient spinal ischemia, manifested as intermittent claudication. Endovascular embolization using liquid material was safe and quite effective for spinal dural AVF. <> Department of Neurosurgery and Radiology, Nagasaki University School of Medicine, 1-7-1 Sakamoto, Nagasaki 852-8501, Japan.

• Heros RC, Debrun GM, Ojemann RG, Lasjaunias PL and Naessens PJ (1986). Direct spinal arteriovenous fistula: a new type of spinal AVM. Case report. J Neurosurg. 64 (1): 134-9. Summary: A patient presenting with progressive paraparesis was found to have a spinal arteriovenous fistula at the T3-4 vertebral level. The lesion consisted of a direct communication of the anterior spinal artery with a very distended venous varix that drained mostly superiorly to the posterior fossa and simulated a posterior fossa arteriovenous malformation (AVM) on vertebral angiography. The patient was treated by surgical ligation of the fistula through an anterior transthoracic approach. He deteriorated abruptly on the 4th postoperative day, probably because of retrograde thrombosis of the enlarged anterior spinal artery. Over the next few months, he improved to the point of being able to walk with crutches. He has also regained sphincter control. The different types of spinal AVM's are reviewed. Our case does not fit into any of these groups. A new category, Type IV, is proposed to designate direct arteriovenous fistulas involving the intrinsic arterial supply of the spinal cord. <>

• Hill CE, Beattie MS and Bresnahan JC (2001). Degeneration and sprouting of identified descending supraspinal axons after contusive spinal cord injury in the rat. Exp Neurol. 171 (1): 153-69. Summary: Contusive spinal cord injury (SCI) results in the formation of a chronic lesion cavity surrounded by a rim of spared fibers. Tissue bridges containing axons extend from the spared rim into the cavity dividing it into chambers. Whether descending axons can grow into these trabeculae or whether fibers within the trabeculae are spared fibers remains unclear. The purposes of the present study were (1) to describe the initial axonal response to contusion injury in an identified axonal population, (2) to determine whether and when sprouts grow in the face of the expanding contusion cavity, and (3) in the long term, to see whether any of these sprouts might contribute to the axonal bundles that have been seen within the chronic contusion lesion cavity. The design of the experiment also allowed us to further characterize the development of the lesion cavity after injury. The corticospinal tract (CST) underwent extensive dieback after contusive SCI, with retraction bulbs present from 1 day to 8 months postinjury. CST sprouting occurred between 3 weeks and 3 months, with penetration of CST axons into the lesion matrix occurring over an even longer time course. Collateralization and penetration of reticulospinal fibers were observed at 3 months and were more extensive at later time points. This suggests that these two descending systems show a delayed regenerative response and do extend axons into the lesion cavity and that the endogenous repair can continue for a very long time after SCI. <> Department of Neuroscience, The Ohio State University, Columbus, Ohio 43210, USA.

• Hoyland JA, Freemont AJ and Jayson MI (1989). Intervertebral foramen venous obstruction. A cause of periradicular fibrosis? Spine. 14 (6): 558-68. Summary: Disc herniation into the intervertebral foramen (IVF) or osteophytic outgrowths from the margins of the apophyseal joints that project into the IVF may compress the neural structures, but in this cadaveric study of 160 lumbar foramens (age range, 35-91 years), the authors have found that they were much more commonly associated with compression and distortion of the large venous plexus within the IVF. In the absence of direct nerve compression (seen in only eight specimens), the most severe neural changes were associated with compression, congestion, and resultant dilatation of foraminal veins. Pathologic changes within and around the nerve root complex included peri- and intraneural fibrosis, edema of nerve roots, and focal demyelination. Inflammatory cells were notably absent. Vascular changes within the thickened fibrous sheath about damaged nerves, namely, basement membrane thickening, suggestive of endothelial cell injury also were observed. The association between vascular compression, tissue fibrosis, and endothelial injury distant from the compression may be causal--probably due to ischemia as a result of reduced venous outflow. Such observations have led the authors to propose that venous obstruction may be an important pathogenic mechanism in the development of perineural and intraneural fibrosis. <> Department of Rheumatology, University of Manchester, England.

• Hui F, Trosselo MP, Meisel HJ, Alvarez H, Sequeira E and Lasjaunias P (1994). Paraspinal arteriovenous shunts in children. Neuroradiology. 36 (1): 69-73. Summary: Arteriovenous shunts within the spinal canal and in the paraspinal region are unusual. Spinal cord and dural arteriovenous communications have been the subject of numerous reports but paraspinal shunts causing venous congestion in the spinal canal are rarer and may present special problems in diagnosis and management. We describe three children with paraspinal arteriovenous malformations, associated with overt or potential venous congestion in the spinal canal. In each case, the lesion was successfully obliterated by endovascular therapy. Embolisation with permanent occlusive agents is an effective treatment for these rare but potentially debilitating lesions. <> Neuroradiologie Vasculaire, Hopital Bicetre, Le Kremlin-Bicetre, France.

• Jellinger K (1986). Vascular malformations of the central nervous system: a morphological overview. Neurosurg Rev. 9 (3): 177-216. Summary: Vascular malformations of the central nervous system (C.N.S.) are classified by size, location, and morphologic type, distinguishing capillary telangiectasias, cavernous malformations, venous angiomas, arteriovenous malformations (AVMs) including varix of the great vein of Galen, and other vascular malformations (e.g. Sturge-Weber syndrome). The morphology and predominant location pattern of the different types of vascular malformations in the brain and spinal cord, and their embryology are reviewed. In the brain and its coverings, all types mainly AVMs and venous angiomas do occur, representing 5-9% of all intracranial space-occupying lesions and 20-40% of the sources of surgically treated intracranial hemorrhages. 50-80% of the angiomas are located in the cerebral hemispheres, 10-18% in central brain areas (basal ganglia, internal capsule, choroid plexus), and 10-30% in the posterior fossa. The major types of cerebral vascular malformations are described with reference to their anatomical features, location, chief arterial and venous supply, and prominent complications. Spinal vascular malformations, accounting for 3 to 12% of spinal space-occupying lesions, include vertebral, extradural, dural, subpial and intramedullary angiomas which occur as isolated or complex vascular anomalies and may involve various covering layers at the same level. The preferential occurrence of angiomas on the dorsal surface of the cord and in the caudal regions is related to the embryologic development of spinal vasculature. Frequent association of spinal angiomas (20-25%) with other vascular anomalies and dysplasias emphasizes their hamartomatous nature and developmental origin. Spinal angiomas include capillary telangiectasias with extra- or intradural and, rarely, intramedullary location, cavernomas, mainly arising in vertebral bodies, venous angiomas, mainly located in vertebral bodies and in the extradural space, and AVMs constituting the commonest type, that may affect both the pial and radicular vessels and can penetrate into the cord. They present as simple AV fistulas, cirsoid angiomas with localized vascular plexuses and large complex convolutions ("juvenile" type). The complications of spinal angiomas include subarachnoid hemorrhage, rare epidural hematoma, hematomyelia, compression lesions of the cord and roots, and ischemic changes causing chronic progressive radiculomyelopathy, previously referred to as Foix-Alajouanine syndrome. Chronic damage to the cord and spinal roots results from pressure effects, thrombosis of the abnormal vessels, disorders of venous drainage, and "steal" phenomena related to the vascular anomalies. <>

• Kataoka H, Miyamoto S, Nagata I, Ueba T and Hashimoto N (2001). Venous congestion is a major cause of neurological deterioration in spinal arteriovenous malformations. Neurosurgery. 48 (6): 1224-9; discussion 1229-30. Summary: OBJECTIVE: Although venous congestion is considered to be a major cause of progressive myelopathy in patients with spinal dural arteriovenous fistulae (DAVFs), the neurological deterioration in patients with spinal intradural arteriovenous malformations (AVMs) has been attributed to hemorrhage or to vascular steal. To reexamine this theory, we analyzed our own cases of spinal vascular diseases. METHODS: In 24 patients with spinal vascular diseases, those who demonstrated progressive myelopathy with T2 hyperintensity in the spinal cord on magnetic resonance imaging (MRI) were diagnosed as patients with congestive myelopathy. We further examined the clinical courses, MRI findings, and reversibility of these cases. RESULTS: Venous congestion was judged to be a cause of neurological deterioration in 13 patients (7 DAVFs, 6 intradural AVMs). The T2 signals on these patients' MRI scans were located in the center and extended over several levels not corresponding to distribution of ischemia due to arterial steal. Of the patients who were diagnosed with congestive myelopathy, no differences between those with DAVFs and those with intradural AVMs were apparent in terms of clinical manifestations and reversibility. Eight (four DAVFs, four intradural AVMs) of 13 patients experienced neurological improvement after treatment. All patients with poor outcomes had intervals from onset of more than 3 years and showed contrast enhancement of the spinal cord on MRI studies. CONCLUSION: Spinal intradural AVMs as well as spinal DAVFs can be a cause of venous congestive myelopathy. Regardless of its etiology, congestive myelopathy is potentially reversible if properly diagnosed and treated. <> Department of Neurosurgery, Graduate School of Medicine, Kyoto University, Japan.

• Kim RC, Smith HR, Henbest ML and Choi BH (1984). Nonhemorrhagic venous infarction of the spinal cord. Ann Neurol. 15 (4): 379-85. Summary: A 71-year-old man experienced gradually progressive leg weakness, urinary retention, and mild loss of sensation in dermatomes T8 through T12 bilaterally. After 5 to 6 weeks of illness, he developed flaccid paraplegia and sensory loss below T8. He died 16 weeks after onset of neurological symptoms. Neuropathologically, there was widespread, subtotal necrosis of the spinal cord, largely of nonhemorrhagic character, from T8 downward. Dorsal and anterior median spinal veins were occluded by a partially organized thrombus. Comparison of this case with 19 previously recorded examples of venous infarction of the spinal cord (8 hemorrhagic, 7 nonhemorrhagic, and 4 embolic) suggests major differences in clinical presentation, rate of progression, and length of survival among the three groups. <>

• Klekamp J, Volkel K, Bartels CJ and Samii M (2001). Disturbances of cerebrospinal fluid flow attributable to arachnoid scarring cause interstitial edema of the cat spinal cord. Neurosurgery. 48 (1): 174-85; discussion 185-6. Summary: OBJECTIVE: Spinal arachnoid scarring may be caused by trauma, inflammation, surgery, spinal instability, degenerative diseases, or malformations and may lead to progressive neurological deficits and syringomyelia. We wanted to investigate the effects of focal arachnoid scarring in the cervical spinal canal of cats on pressures in the subarachnoid space and spinal cord tissue, as well as on spinal cord histological features. METHODS: Twenty-nine adult cats were used for this study. Nine animals served as control animals, whereas 20 animals received a focal arachnoid scar at C1-C2, which was produced by placement of a kaolin-soaked fibrin sponge on the posterior surface of the spinal cord. After 4 months, pressure recordings above and below the scar, in the subarachnoid space and spinal cord, were performed. Elasticity measurements were performed with small bolus injections. Morphometric analyses of brain and ventricle volumes, sizes of the central canal, and sizes of the perivascular spaces in gray and white matter were also performed. RESULTS: No animal developed clinical or neurophysiological evidence of neurological symptoms at any time. In the kaolin-treated group, pressure recordings revealed a significant increase in the subarachnoid pressure at C1, because of the cerebrospinal fluid flow obstruction. Pressure gradients tended to increase at all measuring points. A significant difference was detected between the spinal cord and subarachnoid space at C2, where the intramedullary pressure exceeded the subarachnoid pressure. Elasticity was significantly increased in the spinal cord at C2. Intracranially, no evidence of hydrocephalus was observed. In the spinal cord, perivascular spaces were significantly enlarged in the posterior white matter above the arachnoid scar and in the central gray matter below the area of scarring in the cervical cord. CONCLUSION: Arachnoid scarring at C1-C2 produces an interstitial type of edema in the central gray matter below the area of scarring in the cat cervical cord, because of altered cerebrospinal fluid and extracellular fluid flow dynamics. These changes may be interpreted as the initial stage in the development of syringomyelic cavities. <> Department of Neurosurgery, Nordstadt Krankenhaus, Medizinische Hochschule, Hannover, Germany.

• Koch C, Hansen HC, Westphal M, Kucinski T and Zeumer H (1998). [Congestive myelopathy caused by spinal dural arteriovenous fistulas. Anamnesis, clinical aspects, diagnosis, therapy and prognosis]. Nervenarzt. 69 (4): 279-86. Summary: Congestive myelopathy, formerly referred to as varicosis spinalis or Foix-Alajouanine syndrome, is caused by a spinal dural arteriovenous fistula (SDAVF). So far, the blood supply from the meningeal arteries draining through the fistula into the medullary venous system can only be verified by spinal angiography. Patients predominantly male and over the age of 60 are afflicted. Initially reversible functional disorders caused by the congestion of the spinal cord veins eventually become irreversible, the most common symptom being an increasingly paretic gait disorder, the signs of which generally begin symmetrically and progress from distal to proximal signs. Simultaneously, predominantly transverse sensory dysfunctions develop, as well as bladder and bowel dysfunctions, most often leading to incontinence. MRI typically shows a central medullary signal enhancement with slight swelling of the afflicted region, initially indicative of a reversible congestive edema and later of an irreversible infarction, and extended perimedullar vessels. Thus, if the clinical course and the characteristic MRI findings suggest the possibility of disease related to congestive myelopathy, spinal angiography becomes indispensable. Since ensuing the success of therapy and prognosis depends on rapid determination of the extent of the illness, a speedy diagnostic reaction is mandatory to institute the treatment necessary to prevent paraplegia. <> Abteilung fur Neuroradiologie, Universitats-Krankenhaus Eppendorf, Hamburg.

• Kothbauer K and Seiler RW (1997). [Tethered spinal cord syndrome in adults]. Nervenarzt. 68 (4): 285-91. Summary: The tethered spinal cord syndrome is more often encountered in children, but does also occur in adults. Its clinical spectrum comprises low back pain, neurological deficits such as distal motor weakness and trophic and sensory disturbances in the legs, urological symptoms and such musculoskeletal signs as scoliosis or foot deformities. In addition, cutaneous lesions or subcutaneous lipomas in the lumbosacral region may be indirect signs of an intraspinal pathology. This consists in a tight, thickened and sometimes shortened filum terminale, an intraspinal lipoma, intradural scar formation or other lesions that lead to conus fixation. The common mechanism of injury of these types of pathologies is an impairment of longitudinal movement of the spinal cord, especially the conus medullaris, which subsequently leads to chronic local ischemia. Diagnosis is most readily achieved by magnetic resonance imaging. Treatment is aimed at the restoration of cord mobility by means of microsurgical release of the conus, the cauda equina and the filum terminale with the aid of cauda equina neuromonitoring. Further progression can be effectively halted; in fact almost half of the patient actually improve. Therefore, every patients presenting with the clinical diagnosis of tethered cord syndrome should be offered specialized surgical treatment. <> Neurochirurgische Klinik, Inselspital, Universitat Bern.

• Lee RR, Abraham RA and Quinn CB (2001). Dynamic physiologic changes in lumbar CSF volume quantitatively measured by three-dimensional fast spin-echo MRI. Spine. 26 (10): 1172-8. Summary: STUDY DESIGN: Lumbar MRI of normal adults. Image analysis to measure lumbar CSF volumes at rest and during physiologic maneuvers. OBJECTIVES: 1) Validate an MRI technique to measure CSF volumes, 2) use this technique to measure the resting volume of lumbar CSF, 3) measure changes in CSF volume with physiologic maneuvers, and 4) demonstrate the anatomic basis for these volume changes. SUMMARY OF BACKGROUND DATA: Studies using radiograph and radionuclide myelography in dogs and humans in the 1960s-1980s qualitatively showed decreases in spinal CSF volume with physiologic maneuvers. Theories were proposed to explain these changes, but they could not be confirmed because only the contrast-laden CSF was visualized using these techniques. METHODS: Four adult volunteers had lumbar MRI using a fat-saturated T2-weighted three-dimensional fast spin-echo sequence. Quantitative analysis of images was used to measure lumbar CSF volume; the technique was validated using a water phantom. Lumbar CSF volume was measured 1) at rest, 2) with hyperventilation, 3) with abdominal compression, and 4) with both hyperventilation and abdominal compression. RESULTS: Resting lumbar CSF volume ranged from 28 to 42 mL. Reversible changes in lumbar CSF volume resulting from physiologic maneuvers are visualized by MR myelography and measured. The volume change (10% reduction in volume with hyperventilation, 28% with compression, and 41% with combined hyperventilation and abdominal compression) is directly visualized to be caused by engorgement of the epidural venous plexus, compressing the thecal sac. CONCLUSIONS: MRI provides a noninvasive means to measure spinal CSF volume and demonstrates the anatomic basis of physiologic volume changes. This has important implications for spinal anesthesia. <> Department of Radiology, Johns Hopkins Hospital, Baltimore, Maryland, USA.

• Lee TT, Alameda GJ, Gromelski EB and Green BA (2000). Outcome after surgical treatment of progressive posttraumatic cystic myelopathy. J Neurosurg. 92 (2 Suppl): 149-54. Summary: OBJECT: Progressive posttraumatic cystic myelopathy (PPCM) can occur after an injury to the spinal cord. Traditional treatment of PPCM consists of inserting a shunt into the cyst. However, some authors have advocated a more pathophysiological approach to this problem. The authors of the present study describe their surgical treatment protocol and outcome in a series of patients with syringomyelia. METHODS: Medical records of 34 patients undergoing surgical treatment for PPCM were reviewed. Laminectomies and intraoperative ultrasonography were performed. In patients without focal tethering of the spinal cord and in whom only a confluent cyst had been revealed on ultrasonography, a syringosubarachnoid shunt was inserted; in those with both tethering and a confluent cord cyst, an untethering procedure was performed first. When a significant reduction (>50%) in the size of the cyst was shown after the untethering procedure, no shunt was inserted. When no changes in cyst size were demonstrated on ultrasonography, a short syringosubarachnoid shunt was used. The mean follow-up period was 28.7 months (range 12-102 months). The interval between the mechanism of injury and the operation ranged from 5 months to 37 years (mean 11 years). Pain was the most frequent symptom, which was followed by motor deterioration and spasticity. Postoperative improvement was noted in 55% of patients who experienced motor function deterioration and in 53% of those who demonstrated worsening spasticity. In 14 of 18 patients with an associated tethered spinal cord, tethering alone caused significant collapse of the cyst. Postoperative magnetic resonance imaging demonstrated cyst collapse in 92% of patients who had undergone untethering alone and in 93% of those who underwent syringosubarachnoid shunt placement. Treatment failure was observed in 7% of the former group and in 13% of the latter. CONCLUSIONS: Posttraumatic cystic myelopathy can occur with or without the presence of tethered cord syndrome. Intraoperative ultrasonography can readily demonstrate this distinction to aid in surgical decision making. Untethering alone in patients with tethered cord syndrome and cyst formation can reduce the cyst size and alleviate symptoms and signs of posttraumatic cystic myelopathy in the majority of these cases. Untethering procedures in which duraplasty is performed to expand the subarachnoid space may be a more physiologically effective way of treating tethered cord with associated syringomyelia. <> Department of Neurological Surgery, University of Miami School of Medicine, Florida, USA.

• Mifsud V and Pullicino P (2000). Spinal cord MRI hyperintensities in cervical spondylosis: an ischemic pathogenesis? J Neuroimaging. 10 (2): 96-100. Summary: The pathophysiology of focal spinal cord MRI T2 hyperintensity (SCHI) in patients with cervical spondylosis is uncertain. This study was undertaken to determine the frequency and cause of SCHI. The authors reviewed serial cervical spine magnetic resonance imaging (MRI) reports and reviewed scans with spondylosis and cord compression or SCHI. The authors noted the location, shape, and extent of SCHI, and severity of spondylosis (expressed as a spondylosis score [SS]). The authors recorded the age and vascular risk factors for each patient. Nineteen of 273 scans (7%) with cervical spondylosis and 19 of 36 scans (53%) with cord compression had SCHI. The SCHI extended for one vertebral level from the compression in 12 patients and for three vertebral levels in 5 patients, and were distant from the compression in 2 patients. The SCHI had a focal, symmetrical, anterior spinal artery terminal zone location in 16 of 19 scans (84%). A rim isointense with normal cord separated all SCHI from the pial surface. Patients with SCHI were older (58.3 years +/- 12.8 years versus 46.8 +/- 8.1 years) (p = 0.007) and had a higher SS (5.7 +/- 2.4 versus 3.9 +/- 1.4) (p = 0.02) than patients without SCHI. The SCHI relates to the severity of cervical spondylosis. The anterior spinal artery territory location, the normal cord between SCHI and the compressive lesion, and the presence of SCHI at a distance from the compressive level all suggest an ischemic basis for SCHI. <> Department of Neurology, State University of New York, Buffalo General Hospital 14203, USA.

• Muraszko KM and Oldfield EH (1990). Vascular malformations of the spinal cord and dura. Neurosurg Clin N Am. 1 (3): 631-52. Summary: Current techniques of diagnosis and treatment allow for earlier detection, precise delineation of the vascular anatomy, and, most important, successful treatment of most patients with spinal AVMs. Magnetic resonance imaging is useful in the initial assessment of patients with progressive myelopathy but cannot replace myelography or arteriography in screening patients who may have a spinal AVM. The most common variety of spinal AVM is a dural arteriovenous fistula. Dural arteriovenous fistulas cause cord injury by producing venous congestion, and symptoms can be reversed by elimination of venous congestion of the spinal cord. Dural arteriovenous fistulas can be treated successfully by interrupting the arteriovenous fistula either in the dura or by disconnecting the dural fistula from the coronal venous plexus in the subarachnoid space. This can be done by interrupting the medullary vein that drains the arterial blood from the dural fistula into the coronal venous plexus of the spinal cord. Stripping of the engorged venous network on the surface of the spinal cord is unwarranted and may cause further cord injury. In dural arteriovenous fistulas embolization is often beneficial in patients with acute neurologic deterioration, to permit time for stabilization and improvement in spinal cord hemodynamics and in cord function before neurosurgical intervention is undertaken. Embolization also may be indicated in patients with intradural spinal AVMs in which surgery cannot be performed safely. Although embolic occlusion does not permanently occlude most intradural AVMs, it often permits stabilization of neurologic function and may be repeated later if neurologic dysfunction returns or progresses. Although the outcome after treatment is dependent on the type and location of the spinal AVM, as in most treatable neurologic disorders the functional outcome of patients with spinal AVMs is directly related to their neurologic condition at the time of treatment. Patients with minimal dysfunction, and with easily accessible AVMs, such as dural arteriovenous fistulas, have the greatest chance for useful recovery or stabilization. Since these patients represent the largest number of patients with spinal AVMs, they must be diagnosed and treated early to achieve the best possible outcome. <> Section of Neurosurgery, University of Michigan Medical Center, Ann Arbor.

• Nielsen OA, Biering-Sorensen F, Botel U, Gardner BP, Little J, Ohta H, Shrosbree R and Melwill R (1999). Post-traumatic syringomyelia. Spinal Cord. 37 (10): 680-4. Summary: Post-traumatic syringomyelia is estimated to develop in more than 20% of individuals with traumatic spinal cord injury (SCI). The development can give rise to clinical symptoms 6 months to 26 years after the injury, and presentation 40 years post-injury has been seen by one of the authors.1234 We present an unusual case for comments and discussion. <> Department of Neurosurgery, The Neuroscience Centre, Rigshospitalet, Copenhagen University Hospital, Denmark.

• Nolli M, Crispino M, Nicosia F, Borghi B and Montone N (2001). Diagnosis and therapy of intrathecal bleeding. Minerva Anestesiol. 67 (9 Suppl 1): 82-91. Summary: AIM: Define pathophysiology, epidemiology, diagnosis and therapy in case of spinal bleeding after central neural blockade (CNB). METHODS: Spinal epidural hematoma (SEH) following CNB may occur due to vascular trauma from needle/catheter placement and can occur in subdural and epidural spaces. Epidural artery bleeding seems the source of SEH: the damage mechanism depends on compression and neural vascular ischemia of cord, nerve roots, ganglion and toxicity from blood cell lysis products. Incidence varies (1:150.000 - 1:500.000) but SEH may be asymptomatic. CONCLUSIONS: SEH starts with acute severe low back and/or radicular pain and neurologic signs that may progress to paraparesis, sensory loss and sphincter disturbances. After CNB, the only sign of SEH may be an unusually prolonged motor and sensory block. Symptoms may start even 96 hours after CNB and/or removal of the epidural catheter. Neurological recovery is related to severity and speed of preoperative deficits development and surgical decompression. MR imaging features (diagnostic tool of choice), including degree of cord compression, are useful to establish or confirm the diagnosis of SEH but do not influence the management or predict outcome. Hematoma resolution and severity of neurologic impairment has the greatest impact on management and outcome. Preoperative MRI information and intraoperative evidence of subarachnoid hemorrage (SAH) and CSF leakage is important: SAH worsens outcome for its negative effect on spinal cord and cauda equina. Conservative therapy may be successful in cases with minimal neurologic deficits, despite cord compression. <> Anaesthesia and Intensive Care Department, S. Andrea Hospital, La Spezia, Italy.

• Ohtakara K, Kuga Y, Murao K, Kojima T, Taki W and Waga S (2000). Preoperative embolization of upper cervical cord hemangioblastoma concomitant with venous congestion--case report. Neurol Med Chir (Tokyo). 40 (11): 589-93. Summary: A 16-year-old male presented with a large, solid hemangioblastoma located in the upper cervical cord manifesting as hyperactive reflexes, subtle weakness, and diminished position sense in all extremities. Neuroimaging studies indicated venous congestion due to arteriovenous shunt through the tumor. Preoperative embolization was accomplished without morbidity, and resulted in marked devascularization of the tumor and elimination of an early filling vein. Four days after embolization, the tumor was totally excised without excessive intraoperative bleeding. His neurological deficits gradually improved after surgery. Preoperative embolization is a valuable adjunct to surgical excision of large intramedullary hemangioblastomas, especially those associated with arteriovenous shunt, as cord dysfunction related to venous congestion and the risk of torrential intraoperative bleeding are reduced. <> Department of Neurosurgery, Mie University School of Medicine, Tsu, Japan.

• Paramore CG (2000). Dorsal arachnoid web with spinal cord compression: variant of an arachnoid cyst? Report of two cases. J Neurosurg. 93 (2 Suppl): 287-90. Summary: Spinal arachnoid cysts are diverticula of the subarachnoid space that may compress the spinal cord; these lesions are most commonly found in the thoracic spine. Two patients who presented with thoracic myelopathy were noted on magnetic resonance imaging to have focal indentation of the dorsal thoracic cord, with syringomyelia inferior to the site of compression. Both patients were found at operation to have discrete arachnoid "webs" tenaciously attached to the dura mater and pia mater. These webs were not true arachnoid cysts, yet they blocked the flow of cerebrospinal fluid (CSF) and caused focal compression of the spinal cord. The mass effect appeared to be the result of a pressure gradient created by the obstruction of CSF flow in the dorsal aspect of the subarachnoid space. Both patients responded well to resection of the arachnoid web. Arachnoid webs appear to be rare variants of arachnoid cysts and should be suspected in patients with focal compression of the thoracic spinal cord. <> Department of Surgery (Neurosurgery), University of Alabama at Birmingham, USA.

• Partington MD, Rufenacht DA, Marsh WR and Piepgras DG (1992). Cranial and sacral dural arteriovenous fistulas as a cause of myelopathy. J Neurosurg. 76 (4): 615-22. Summary: The authors report a series of seven patients with myelopathy who were found to have spinal dural arteriovenous (AV) fistulas in which the nidus was located at some distance from the spinal cord. The nidus was intracranial in three cases and involved a sacral nerve root sheath in the other four; in each case, the arterialized draining vein led into the coronal plexus of medullary veins. A lack of normal draining radicular veins was noted in all cases. Magnetic resonance images were obtained in four patients and demonstrated spinal cord tissue changes only in the lower thoracic cord in three cases and in the cervical cord in one, all consistent with an ischemic process secondary to venous hypertension. Five patients were managed surgically by division of the draining vein, with improvement of the neurological deficit in all. One patient was treated by embolization alone and had stabilization of her deficit. The remaining patient in the series died of unrelated systemic disease before the spinal dural AV fistula could be treated. These cases support the theory that venous hypertension is the dominant pathophysiological mechanism involved in spinal dural AV fistulas independent of their location. In patients with a suspected spinal dural AV fistula, lumbar and thoracic spinal angiography will reveal the site of the fistula in the majority of cases (88% in this series). In the remaining patients, the possibility of a remote fistula must be considered. The lack of normal venous drainage of the cord following injection in the artery of Adamkiewicz is the most reliable indicator of venous hypertension in the cord and can be helpful in making the decision to proceed with a search for a cranial or sacral arterial supply. <> Department of Neurological Surgery, Mayo Graduate School, Rochester, Minnesota.

• Rosenblum B, Oldfield EH, Doppman JL and Di Chiro G (1987). Spinal arteriovenous malformations: a comparison of dural arteriovenous fistulas and intradural AVM's in 81 patients. J Neurosurg. 67 (6): 795-802. Summary: The medical records and arteriograms of 81 patients with spinal arteriovenous malformations (AVM's) were reviewed, and the vascular lesions were classified as dural arteriovenous (AV) fistulas or intradural AVM's. Intradural AVM's were further classified as intramedullary AVM's (juvenile and glomus types) and direct AV fistulas, which were extramedullary or intramedullary in location. Dural AV fistulas were defined as being supplied by a dural artery and draining into spinal veins via an AV shunt in the intervertebral foramen. Intramedullary AVM's were defined as having the AV shunt contained at least partially within the cord or pia and receiving arterial supply by medullary arteries. Of the 81 patients, 27 (33%) had dural AV fistulas and 54 (67%) had intradural AVM's. Several dissimilarities in clinical and radiographic findings of the two subgroups were evident. The patients with intramedullary AVM's were younger; the age at onset of symptoms averaged 27 years compared to 49 years for dural AV fistulas. The most common initial symptom associated with dural AV fistulas was steadily progressive paresis, whereas hemorrhage was the most common presenting symptom in cases of intramedullary lesions. No patients with dural AV fistulas had subarachnoid hemorrhage. Activity exacerbated symptoms more frequently in patients with dural lesions. Associated vascular anomalies occurred only in cases of intradural AVM's. In 96% of the dural lesions the AV nidus was in the low thoracic or lumbar region; in only 15% did the intercostal or lumbar arteries supplying the AVM also provide a medullary artery which supplied the spinal cord. In contrast, most intradural AVM's (84%) were in the cervical or thoracic segments of the spinal cord and all of them were supplied by medullary arteries. Transit of contrast medium through the intradural AVM's was rapid in 80% of cases, suggesting high-flow lesions. Forty-four percent of the patients with AVM's of the spinal cord had associated saccular arterial or venous spinal aneurysms. No dural AV fistulas displayed these characteristics. A good outcome occurred in 88% of patients with dural AV fistulas after nidus obliteration, while 49% of patients with intramedullary AVM's did well after surgery or embolization. These findings suggest that dural and intradural AVM's differ in etiology (acquired vs. congenital) and that they have different pathophysiology, radiographic findings, clinical presentation, and response to treatment. <> Clinical Neurosurgery Section, National Institute of Neurological and Communicative Disorders and Stroke, Bethesda, Maryland.

•• Salvador de la Barrera S, Barca-Buyo A, Montoto-Marques A, Ferreiro-Velasco ME, Cidoncha-Dans M and Rodriguez-Sotillo A (2001). Spinal cord infarction: prognosis and recovery in a series of 36 patients. Spinal Cord. 39 (10): 520-5. Summary: OBJECTIVE: To study the clinical evolution and the functional outcome of patients suffering from spinal cord infarction who were treated at the Spinal Cord Injuries Unit. To try to determine the factors that could have influence in their functional outcome. SETTING: In a Spinal Cord Injuries Unit, regionally-based, and which forms part of a general hospital with a high level of specialization. METHOD: Retrospective study of the medical records of patients suffering from vascular spinal cord ischemia, as acute anterior spinal artery syndrome or associated with aortic surgery or rupture. Cases that were due to compressive, tumoral or inflammatory pathologies were excluded. Assessment of the neurological syndrome followed the ASIA/IMSOP criteria. Age, sex, history and magnetic resonance imaging (MRI) findings were analyzed. Assessment of functional outcome was made regarding ambulatory ability or wheelchair use, and bladder/sphincter control. RESULTS: Thirty-six cases were selected, the commonest group being spinal cord ischemia due to idiopathic causes (36.1%). Following these, there were cases associated with aortic surgery (25%), systemic arteriosclerosis (19.4%) and acute deficit of perfusion (11.1%). The average age of the patients was 59.3 years, with a mortality of 22.2% during the hospital stay. Regarding the functional outcomes at the moment of discharge, it must be pointed out that 57.1% of the patients were wheelchair users, 25% were ambulatory, using technical aids, and 17.9% were fully ambulatory. The group who could perform some kind of walking was significantly younger than the group of wheelchair users (48.17 vs 61.38 years). Additionally, it became evident that those patients who did not show voluntary muscle contraction at the time of admission (ASIA groups A and B) presented a higher risk of being wheelchair users. CONCLUSION: Acute spinal cord ischemia syndrome has a severe prognosis with permanent and disabling sequelae. Initial neurological assessment following ASIA/IMSOP classification proves to be the best predictor of prognosis, and the patient's advanced age constitutes a negative factor for functional recovery. <> Spinal Cord Injuries Unit, Hospital Juan Canalejo, A Coruna, Spain.

• Schaan M and Jaksche H (2001). Comparison of different operative modalities in post-traumatic syringomyelia: preliminary report. Eur Spine J. 10 (2): 135-40. Summary: Post-traumatic syringomyelia (PTS) is a relatively rare, but potentially disastrous, complication of spinal cord injury. Operative treatment by shunting procedures often shows only a short-term improvement, and the rate of recurrence of syringomyelia is high, so different treatment modalities have been used in the last years. The various results are discussed in this analysis. A prospective clinical study was conducted of 30 patients with PTS treated by shunting procedures or with pseudomeningocele over a period of 9 years, and followed with regular clinical and magnetic resonance imaging examinations. Shunting procedures like syringosubarachnoid and syringopleural or -peritoneal shunting showed good results only at the first follow-ups. In our department, we perform an artificial liquor reservoir at the level of the lesion after opening the spinal pathways and arachnoid adhesions at that level. This procedure was performed in 12 patients. Five of these had been previously operated by shunting procedures; all of them had suffered a recurrence of syringomyelia because of internal occlusion. In the group of patients treated by shunting procedures, a neurological improvement was be recorded in five, and a steady state in eight. Five patients showed a further deterioration. The performance of an artificial liquor reservoir to guarantee a free flow of cerebrospinal fluid around the lesion resulted in a neurological improvement in ten patients, with two maintaining a steady state. Our experience is that shunting procedures often show a neurological improvement only in the short term; the rate of recurrence of typical shunting complications is high. The performance of a pseudomeningocele is an encouraging new step in the treatment of PTS. Further long-term follow-up studies are necessary to assess the benefits of this new method. <> Department of Neurosurgery, BG-Unfallklinik Murnau, Prof. Kuntscher-Strasse 8, 82418 Murnau, Germany.

• Stoodley MA, Gutschmidt B and Jones NR (1999). Cerebrospinal fluid flow in an animal model of noncommunicating syringomyelia. Neurosurgery. 44 (5): 1065-75; discussion 1075-6. Summary: OBJECTIVE: The source of fluid and the mechanism of cyst enlargement in syringomyelia are unknown. It has been demonstrated that cerebrospinal fluid (CSF) normally flows from the subarachnoid space through perivascular spaces and into the spinal cord central canal. The aim of this study was to investigate whether this flow continues during cyst formation in an animal model of syringomyelia and to determine the role of subarachnoid CSF flow in this model. METHODS: The intraparenchymal kaolin model of noncommunicating syringomyelia was established in 78 Sprague-Dawley rats. Horseradish peroxidase was used as a tracer to study CSF flow at 1 day, 3 days, 1 week, and 6 weeks after kaolin injection. CSF flow was studied at 0, 10, and 30 minutes after horseradish peroxidase injection into the cisterna magna or thoracic subarachnoid space. RESULTS: The central canal became occluded at the level of the kaolin injection and at one or more rostral levels. Segments of the central canal isolated between occlusions gradually dilated, and axonal retraction balls were detected in the surrounding white matter. There was a partial blockage of subarachnoid CSF flow at the site of the kaolin injection, both in a rostral-caudal direction and in a caudal-rostral direction. Horseradish peroxidase was detected at all time points, in a distinctive pattern, in perivascular spaces and the central canal. This pattern was seen even where segments of the central canal were isolated and dilated. CONCLUSION: In this animal model, noncommunicating syringes continue to enlarge even when there is evidence that they are under high pressure. There may be an increase in pulse pressure rostral to the block of subarachnoid CSF flow, causing an increase in perivascular flow and contributing to syrinx formation. The source of fluid in noncommunicating syringomyelia may be arterial pulsation-dependent CSF flow from perivascular spaces into the central canal. <> Department of Surgery (Neurosurgery), University of Adelaide, Royal Adelaide Hospital, Australia.

• Tartaglino LM, Croul SE, Flanders AE, Sweeney JD, Schwartzman RJ, Liem M and Amer A (1996). Idiopathic acute transverse myelitis: MR imaging findings. Radiology. 201 (3): 661-9. Summary: PURPOSE: To analyze the magnetic resonance (MR) imaging findings in idiopathic acute transverse myelitis (IATM) in relation to pathologic findings and MR findings in Guillain-Barre syndrome and ischemia. MATERIALS AND METHODS: The cases of 19 patients with IATM seen over a 4-year period were retrospectively reviewed. Clinical parameters and laboratory test findings were recorded for each patient independently of the MR findings. RESULTS: Ten (53%) patients experienced upper respiratory infection or vaccination within 4 weeks of symptom onset. The majority (82%) of cases occurred between December and May each year. In seven of 12 patients who underwent electromyography and nerve conduction examinations, evidence of peripheral nerve injury was seen. On T2-weighted axial images, 13 of 18 lesions were depicted with holocord abnormal signal intensity, seven (39%) had gray matter involvement similar to that seen in spinal cord ischemia, and three (16%) had isolated white matter involvement. Enhancement patterns varied. In three (17%) of the 18 lesions, enhancement in the cauda equina was similar to that seen in Guillain-Barre syndrome. CONCLUSION: IATM may be caused by a small vessel vasculopathy. MR findings in IATM also occasionally are similar to those described in Guillain-Barre syndrome and suggest a possible relationship. <> Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, PA 19107, USA.

• Tokunaga M, Minami S, Isobe K, Moriya H, Kitahara H and Nakata Y (2001). Natural history of scoliosis in children with syringomyelia. J Bone Joint Surg Br. 83 (3): 371-6. Summary: We performed a retrospective review of 27 scoliotic patients with syringomyelia using MRI. Their mean age at the first MRI examination was 10.9 years, and at the final review 15.8 years. The mean ratio of the diameter of the syrinx to the cord on the midsagittal MRI (S/C ratio) decreased from 0.49 to 0.24; 14 patients showed a decrease of 50% or more (reduction group). In this reduction group, the cerebellar tonsillar herniation decreased from a mean of 11.3 mm to 6.0 mm, and some improvement in dissociated sensory disturbance was seen in nine of 13 patients. The scoliosis improved by 5 degrees or more in six patients in the reduction group. Our results indicate that spontaneous shrinkage of syringomyelia in children is not unusual and is associated with improvement in the tonsillar herniation, the scoliosis and the neurological deficit. <> Department of Orthopaedic Surgery, Chiba University and Chiba Higashi Hospital, Japan.

• Tsuladze, II (1999). [The selective phlebography of the large tributaries of the vena cava system in the diagnosis of venous circulatory disorders in the spinal complex]. Zh Vopr Neirokhir Im N N Burdenko. (2): 8-13; discussion 14. Summary: As any other organ, the spinal cord also suffers in chronic congestion. Since the epidural venous system drains into the vena cava system and participates in collateral circulation, there is increased inflow with impaired blood flow along its large tributaries in the vertebral canal along with poor outflow, resulting in intracanal hypertension and chronic congestion. Venous hemodynamic disorders are found beyond the vertebral canal and detected by selective phlebography of the large tributaries of the vena cava system. The technique was used to examine 46 patients with spastic paraparesis or tetraparesis of unclear etiology, which provides evidence for the fact that vena cava stenoses, compressions, atresia, and thromboses can be responsible for impaired venous hemodynamics in the vertebral apparatus and its surgical correction is possible. <>

• Voskuhl RR and Hinton RC (1990). Sensory impairment in the hands secondary to spondylotic compression of the cervical spinal cord. Arch Neurol. 47 (3): 309-11. Summary: In a 5-year period, 11 patients with spondylotic compression of the cervical spinal cord presented with a clinical picture dominated by glove-distribution sensory loss in the hands. Compressive lesions in each case were documented by myelography. The hand sensory loss was often global, and in some patients the involvement extended proximally as far as the elbows. Motor findings in the hands were no more than mild to moderate, as were motor and sensory findings in the legs. Nine patients improved with surgical decompression. The syndrome may result from ischemia to the intrinsic border areas of collateralization between the superficial pial network and the central arterial supply to the cervical cord, although venous stagnation may also play a role. This clinical presentation should always raise the suspicion of a cervical myelopathy, which is potentially treatable. <> Department of Neurology, University of Texas Southwestern Medical School, Dallas 75235-9036.

• Willinsky R, Lasjaunias P, Terbrugge K and Hurth M (1990). Angiography in the investigation of spinal dural arteriovenous fistula. A protocol with application of the venous phase. Neuroradiology. 32 (2): 114-6. Summary: The authors present their protocol for spinal angiography in their investigation of dural arteriovenous fistula (DAVF). The protocol has been used in approximately 120 patients from 1983 to the present at Bicetre Hospital. The approach is based on the fact that venous congestion is responsible for the myelopathy of DAVF. If the venous phase of the spinal circulation is normal, this alone rules out DAVF as the cause of the patient's symptoms. If there is stasis in the spinal circulation, this is consistent with DAVF, and thus complete spinal angiography is necessary. Complete angiography includes the selective intercostal arteries, including the lateral sacrals, as well as the supply to the cervical cord and posterior fossa. <> Department of Radiology, Hopital De Bicetre, Le Kremlin-Bicetre, France.

• Zhang Z, Nonaka H, Nagayama T, Hatori T, Ihara F, Zhang L and Akima M (2001). Circulatory disturbance of rat spinal cord induced by occluding ligation of the dorsal spinal vein. Acta Neuropathol (Berl). 102 (4): 335-8. Summary: Spinal cord infarction can be caused by venous disturbances due to trauma or cancer invasion. However, the precise mechanism of venous infarction is not fully understood. To characterize disorders associated with spinal venous occlusion, we performed time-kinetic pathological analyses of rat spinal cord infarction induced by transdural ligation of the dorsal spinal vein at the levels of the T10-T13 vertebrae. One day after ligation congestion, edema and hemorrhage were observed mainly in the dorsal funiculus. Axons were well preserved, but on the 3rd day axonal degeneration became evident. On the 7th day, the necrotic lesion was confined to the dorsal funiculus and was round in shape with foamy macrophage infiltration and astrocytic gliosis. On the 14th day, the involved cord became atrophic, and infiltration of foamy macrophages and astrocytosis became more prominent. After 21-28 days, the infarction focus decreased in size due to gliosis, and residual macrophages were observed. The main lesion was confined to the dorsal funiculus at all times. However, the severity of the softening varied among rats. Thus, we conclude that the disturbance of venous drainage actually results in spinal cord softening. The variability in the lesions is probably due to the presence of unexpected anastomoses of the spinal venous system. <> First Department of Pathology, Toho University, School of Medicine, Tokyo, Japan.