Role of imaging studies in the evaluation and localization of cerebrospinal fluid leaks

Introduction

The radiographic approach to the assessment of skull base cerebrospinal fluid (CSF) leaks, whether traumatic, iatrogenic, congenital, secondary to underlying increased intracranial pressure, or otherwise spontaneous, can be aimed at attempting to initially answer the question of whether a true CSF leak can be detected and then to both localize the site of the leak and to assess the surrounding anatomy for potential preoperative planning purposes. The choice of modality and use of contrast is highly variable across practices. Although algorithms have been proposed, what follows is less of a prescriptive pathway to follow and more of an exposure to what can be used and what the variations and limitations can be to allow for adaptation to your practice. Finally, this chapter reviews some lesser known techniques and their potential role in the direction imaging might take in the future.

An important consideration at the outset is whether or not the CSF leak is active at the time of imaging because this may modify the approach to imaging. In the absence of an active CSF leak or one that can be induced by provocative maneuvers, the sensitivity of computed tomography (CT) cisternography has been shown to be only 48%. Although the term cisternography is variably used in the literature, it will be taken here to mean an examination utilizing the administration of an intrathecal agent for any imaging modality. Although it is not absolutely necessary for there to be active rhinorrhea at the time of the scan, its presence is usually associated with a higher likelihood of successfully identifying the leak site. ^,

In general, nuclear medicine examinations are reserved for select cases to assess if a leak is present, as not all clear rhinorrhea is necessarily CSF. The primary modality for localization is currently CT. Magnetic resonance imaging (MRI) cisternography has been shown to be valuable for leak localization in some small studies as discussed later in the chapter and may become more of a standard in the future. More commonly, MRI serves a more supplementary role because it is clearly superior for characterizing relevant intracranial parenchymal abnormalities at or adjacent to the leak site or otherwise relevant to the operative approach in addition to evaluating for potential secondary intracranial complications. The complication rate before dural repair can be quite high, the primary consideration being meningitis, and can range from 10% to 37% in the short term with a similar recurrence rate and up to 85% over 10 years. Although the rate of infection is lower in patients being given antibiotic prophylaxis, up to 10% still develop meningitis. MRI with intravenous (IV) contrast may play an essential role as part of the evaluation if there is clinical concern for intracranial infection in the setting of CSF rhinorrhea.

Intrathecal gadolinium for MRI cisternography is currently used infrequently for leak localization in the skull base, primarily because although it is technically feasible, the off-label use of intrathecal gadolinium and potential safety issues are valid concerns. Still, more recent studies are showing the promise of this technique for localization in addition to the aforementioned benefits of all in one imaging examination as discussed later in the chapter. Fluoroscopy, radiography, and ultrasonography are generally of no value in the evaluation of CSF rhinorrhea.

Imaging protocols