Images below require Macromedia's Flash Player to view

Neuroradiology Case of the Week

Case 116

Clinical Presentation: The patient is a 28-year-old male who suffered a gun shot wound to the vertex a few years ago. He sustained multiple fractures, including those of the left maxilla, orbital floor, lamina papyracea and cribriform plate. The patient underwent placement of a surgical plate on the right aspect of the cribriform plate. His clinical course was complicated by bacterial meningitis, persistent left lower extremity weakness and seizures. He now presents with continued rhinorrhea for which he was referred for a CT cisternogram to locate the area of CSF leakage.

Radiological Findings: Figure 1 is a coronal slice from a CT scan of the face performed before injection of contrast material. An osseous defect is noted to the left in the cribriform plate with opacification of ethmoid air cells underneath.
     A lumbar puncture was performed at the L3-L4 level, and 10cc of Omnipaque-300 was injected into the thecal sac. The patient was placed in a prone Trendelenburg position for a few minutes after which he was transferred to CT. During the CT scan, he was placed prone with the neck extended. Direct coronal images of the face and the anterior fossa of the skull base were obtained. The bony defect to the left in the cribriform plate shows leakage of contrast material into ethmoid air cells (Figure 2). There is a surgical plate seen along the right aspect of the cribriform plate, but no leakage of contrast medium in this region.

Discussion: Cerebrospinal fluid (CSF) leaks arise from a number of etiologies. The most common cause of CSF leaks is accidental trauma resulting in a closed head injury (CHI). Approximately 1-3% of CHI patients develop CSF leaks. These leaks often begin within 48 hours, and 95% are manifest within 3 months of injury. More than 70% will close with conservative management such as bed rest or lumbar drains However, patients assume an increased risk of ascending meningitis (30-40% increase) during the period of closure. Other etiologies of CSF leaks include surgical trauma, tumors, congenital and spontaneous leaks.
     Diagnosing CSF leaks have continued to be challenging to the clinician and radiologist. Rhinorrhea can be sent for beta-2-transferrin, a protein specific to CSF, glucose and protein concentrations. However, preoperative assessment requires localization of the osseous defect. Radionuclide cisternography and contrast-enhanced CT cisternography have been the preferred methods of examination. More recently, high resolution CT and MR cisternography have become alternate means of evaluating CSF leaks.
     Radionuclide cisternography involves endoscopic placement of intranasal pledgets, injection of 99mTc-DTPA into thecal sac through a lumbar puncture, and recording of images at several points in time over a course of 36 hours. This technique has become less frequently used due to the length of study time, relatively higher false-positives and -negatives, and poor localization.
     CT cisternography, which was used in this study, allows for more accurate anatomic localization, though findings are often subtle and difficult to detect, especially if there is not a reasonable contrast medium concentration in the subarachnoid space. This method entails administration of intrathecal contrast material through a lumbar puncture, placing the patient in a prone Trendelenburg position for a few minutes, preferably until CSF is dripping from the nose, followed by CT examination, most often using direct coronal but sometimes axial images. Extracranial contrast, areas of pooled contrast at the fracture site, and deformity of the adjacent sulcus suggest location for CSF leakage. This exam is most valuable in the evaluation of frontal and sphenoid sinus leaks. Intermittent leaks may, however, be missed.
     Although HRCT does not demonstrate actual CSF leakage, it has proved valuable and perhaps, more sensitive, in detecting CSF leaks with provided clinical knowledge of the leak. 1-3 mm axial and coronal cuts are typically obtained, providing anatomic localization of osseous defects and information helpful in determining the surgical approach. MR cisternography is noninvasive with a high sensitivity (85-92%) and specificity (100%), though evaluation of osseous detail is suboptimal compared to CT.
     This patient demonstrated a large defect in the left cribriform plate, which required surgical management.

References:

1. Schlosser RJ, Bolger WE. Nasal cerebrospinal fluid leaks: critical review and surgical considerations. Laryngoscope 2004 Feb 114: 255-265. [Medline]
2. Stone JA, Catillo M, Neelon B, Mukherji SK. Evaluation of CSF leaks: high-resolution CT compared with contrast-enhanced CT and radionuclide cisternography. AJNR 1999; 20:706-712. [Medline]
3. Briggs RD, Ryan M, Quinn FB. Cerebrospinal fluid rhinorrhea and otorrhea. Grand Rounds Presentation UTMB, Dept. of Otolaryngology, 2 Oct 2002. (Text: PDF) (Slides: PDF)
4. Jinkins JR, Rudwan M, Krumina G, Tali ET. Intrathecal gadolinium-enhanced MR cisternography in the evaluation of clinically suspected cerebrospinal fluid rhinorrhea in humans: early experience. Radiology 2002; 222:555-559. [Medline]