SPINE Volume 28, Number 20, pp E413–E419 ©2003, Lippincott Williams & Wilkins, Inc. Altered Subarachnoid Space Compliance and Fluid Flow in an Animal Model of Posttraumatic Syringomyelia Andrew R. Brodbelt MBChB, FRCS(Ed),* Marcus A. Stoodley, PhD, FRACS,* Amy M. Watling, BSc,* Jian Tu, MB BS, PhD,* Sea´n Burke, DCR(R),† and Nigel R. Jones, DPhil, FRACS‡ Study Design. A histologic study of cerebrospinal fluid tracers in Sprague-Dawley rats undergoing lumboperito- neal shunt insertion in the excitotoxic animal model of posttraumatic syringomyelia (PTS). Objectives. To determine the effects of cerebrospinal fluid (CSF) diversion from the subarachnoid space on perivascular flow (PVS) and syrinx formation in posttrau- matic syringomyelia. Summary of Background Data. In an animal model of PTS, fluid enters syringes from the subarachnoid space via perivascular spaces. Preferential PVS flow occurs at the level of the syrinx. It has been suggested that arach- noiditis predisposes to posttraumatic syringomyelia for- mation by obstructing subarachnoid cerebrospinal fluid flow and enhancing perivascular flow. Materials and Methods. Thirty-two male Sprague- Dawley rats were investigated using the CSF tracer horse- radish peroxidase (HRP), the excitotoxic model of PTS, and lumboperitoneal shunt insertion. Five experimental groups consisted of normal controls, syrinx only and shunt only controls, and shunt insertion before or after syrinx formation. In all groups except normal controls, CSF flow studies were performed 6 weeks after the final intervention. Grading scales were used to quantify HRP staining. Results. All excitotoxic model animals formed sy- ringes. Perivascular flow was greatest at the level of the syrinx. Cerebral cortex perivascular flow was significantly reduced after shunt insertion in animals with a syrinx (P  0.05). Shunt insertion did not alter syrinx length or size. There were no significant differences between shunt and syrinx first groups. Conclusions. Increasing caudal subarachnoid space compliance with a shunt does not affect local CSF flow into the spinal cord and syrinx. These results suggest that localized alterations in compliance, as opposed to ob- struction from traumatic arachnoiditis, may act as an important factor in syrinx pathogenesis. [Key words: syringomyelia, animal model, posttraumatic, fluid flow, perivascular space, shunt, treatment] Spine 2003;28: E413–E419 Posttraumatic syringomyelia develops in up to 28% of patients after spinal cord injury and reported case series demonstrate that only half will improve after treat- ment. 1–9 Effective treatment is unlikely without an un- derstanding of the underlying etiology. The hydrody- namics involved in the etiology of posttraumatic (extracanalicular) syringomyelia are not readily ex- plained by previous theories that have concentrated on Chiari malformation associated (canalicular) syrinxes. 10 –15 Previous work using the CSF tracer, horseradish per- oxidase (HRP), has demonstrated that fluid flows along perivascular spaces into the spinal cord and central ca- nal. 16,17 An animal model of posttraumatic syringomye- lia has been characterized that combines excitotoxic amino acid injury with localized adhesive arachnoid- itis. 18,19 In this excitotoxic model of posttraumatic syrin- gomyelia, CSF flows mainly along central branches of the anterior spinal artery and flow is preferential at the level of the syrinx and arachnoiditis. 20 The properties of adhesive arachnoiditis that potenti- ate syrinx formation have been postulated as CSF flow obstruction, cord tethering, or a change in the subarach- noid space compliance. 20–22 Little experimental evi- dence exists investigating these important etiological pa- rameters. This study was performed to explore the effects of shunting CSF from the lumbar cistern to alter total spinal subarachnoid space compliance while maintain- ing any obstructive or tethering effects of localized arachnoiditis. Materials and Methods After ethical approval from the Animal Care and Ethics Com- mittee of the University of New South Wales, 26 male Sprague- Dawley rats weighing 399  63 g (mean  SD) were investi- gated using the CSF tracer HRP, the excitotoxic and arachnoiditis model of PTS, and lumboperitoneal shunt inser- tion. Five experimental groups consisted of normal controls, syrinx only and shunt only controls, and shunt insertion before or after syrinx formation (Table 1). Syrinx first animals were used in a previous study. 20 One animal in the shunt control group had cisternal fluorescein instead of HRP injected in an attempt to confirm shunt patency. Syrinx induction and shunt insertion were performed in a From *Prince of Wales Medical Research Institute, †Department of Medical Imaging, Prince of Wales Hospital, ‡Department of Surgery (Neurosurgery), University of Adelaide. Acknowledgment date: March 3, 2003. First revision date: April 13, 2003. Acceptance date: May 29, 2003. The device(s)/drug(s) that is/are the subject of the manuscript is/are not intended for human use. Federal/Foundation funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. This Research Project was supported by a grant from the National Health and Medical Research Council of Australia (Project No 157063). A. R. Brodbelt is supported by a grant from the Madeline Foundation for Neurosurgical Research. Address correspondence and reprint requests to Dr. Andrew Brodbelt, Prince of Wales Medical Research Institute, Barker Street, Randwick, New South Wales 2031 Australia; e-mail: abrodbelt@doctors.org.uk E413