Symptom predictors of cerebrospinal fluid leaks

Can J Neurol Sci. 2008 Sep;35(4):452-7. doi: 10.1017/s0317167100009112.

Abstract

Background: Spinal cerebrospinal fluid (CSF) leaks, which are considered a cause of intracranial hypotension, generally do not cause any local symptoms. Although symptoms are key elements for further evaluation, few studies have examined symptom predictors of intracranial hypotension. The aim of this study was to determine what symptoms are predictors of CSF leaks in patients suspected of intracranial hypotension.

Methods: We performed radionuclide cisternography in 207 consecutive patients suspected of intracranial hypotension. Intracranial hypotension was suspected when a patient had a history of minor trauma and complained about uncontrolled headache, cranial nerve dysfunction, autonomic dysfunction, or higher brain dysfunction. The leakage of CSF was defined as direct signs of tracer leak into the spinal epidural space or early accumulation of the tracer in the urinary bladder. We obtained information on 16 symptoms commonly reported in previous studies.

Results: CSF leaks were observed in 154 cases (74%). Back pain, limb pain, and limb numbness were inversely associated with CSF leaks (p = 0.042, p = 0.045, and p = 0.006, respectively). In logistic regression analysis, diplopia was a positive predictor of CSF leaks (odds ratio [OR], 6.53; 95% confidence interval [CI], 1.49 to 28.51), whereas limb numbness was a negative predictor (OR, 0.38; 95% CI, 0.17 to 0.84). Of the 21 patients in whom diplopia was present and limb numbness was absent, 20 had CSF leaks (specificity, 98%; positive predictive value, 95%).

Conclusion: Some symptoms may be helpful in the diagnosis of CSF leaks in patients suspected of intracranial hypotension.

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Biomarkers / metabolism
  • Cerebrospinal Fluid*
  • Child
  • Female
  • Humans
  • Intracranial Hypotension* / cerebrospinal fluid
  • Intracranial Hypotension* / etiology
  • Male
  • Middle Aged
  • Regression Analysis
  • Wounds and Injuries* / cerebrospinal fluid
  • Wounds and Injuries* / complications

Substances

  • Biomarkers