Intrathecal pressure monitoring and cerebrospinal fluid drainage in acute spinal cord injury: a prospective randomized trial

Abstract

Object Ischemia is an important factor in the pathophysiology of secondary damage after traumatic spinal cord injury (SCI) and, in the setting of thoracoabdominal aortic aneurysm repair, can be the primary cause of paralysis. Lowering the intrathecal pressure (ITP) by draining CSF is routinely done in thoracoabdominal aortic aneurysm surgery but has not been evaluated in the setting of acute traumatic SCI. Additionally, while much attention is directed toward maintaining an adequate mean arterial blood pressure (MABP) in the acute postinjury phase, little is known about what is happening to the ITP during this period when spinal cord perfusion pressure (MABP − ITP) is important. The objectives of this study were to: 1) evaluate the safety and feasibility of draining CSF to lower ITP after acute traumatic SCI; 2) evaluate changes in ITP before and after surgical decompression; and 3) measure neurological recovery in relation to the drainage of CSF. Methods Twenty-two patients seen within 48 hours of injury were prospectively randomized to a drainage or no-drainage treatment group. In all cases a lumbar intrathecal catheter was inserted for 72 hours. Acute complications of headache/nausea/vomiting, meningitis, or neurological deterioration were carefully monitored. Acute Spinal Cord Injury motor scores were documented at baseline and at 6 months postinjury. Results On insertion of the catheter, mean ITP was 13.8 ± 1.3 mm Hg (± SD), and it increased to a mean peak of 21.7 ± 1.5 mm Hg intraoperatively. The difference between the starting ITP on catheter insertion and the observed peak intrathecal pressure after decompression was, on average, an increase of 7.9 ± 1.6 mm Hg (p < 0.0001, paired t-test). During the postoperative period, the peak recorded ITP in the patients randomized to the no-drainage group was 30.6 ± 2.3 mm Hg, which was significantly higher than the peak intraoperative ITP (p = 0.0098). During the same period, the peak recorded ITP in patients randomized to receive drainage was 28.1 ± 2.8 mm Hg, which was not statistically higher than the peak intraoperative ITP (p = 0.15). Conclusions The insertion of lumbar intrathecal catheters and the drainage of CSF were not associated with significant adverse events, although the cohort was small and only a limited amount of CSF was drained. Intraoperative decompression of the spinal cord results in an increase in the ITP measured caudal to the injury site. Increases in intrathecal pressure are additionally observed in the postoperative period. These increases in intrathecal pressure result in reduced spinal cord perfusion that will otherwise go undetected when measuring only the MABP. Characteristic changes in the observed intrathecal pressure waveform occur after surgical decompression, reflecting the restoration of CSF flow across the SCI site. As such, the waveform pattern may be used intraoperatively to determine if adequate decompression of the thecal sac has been accomplished.