Intrathecal Gadolinium-enhanced MR Cisternography in the Evaluation of Clinically Suspected Cerebrospinal Fluid Rhinorrhea in Humans: Early Experience1

¹ From the Department of Radiological Sciences, Medical College of Pennsylvania-Hahnemann University, 245 N 15th St, Mailstop 206, Philadelphia, PA 19102-1192 (J.R.J.); Department of Radiology, Ibn Sina Hospital, Kuwait City, Kuwait (M.R.); Radiology Clinic, Education and Research Center, Riga, Latvia (G.K.); and Department of Radiology, Gazi University School of Medicine, Ankara, Turkey (E.T.T.). Received January 9, 2001; revision requested February 28; revision received April 20; accepted May 21. 

	ABSTRACT

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
In this prospective multicenter study, the authors evaluated the utility of magnetic resonance (MR) cisternography after intrathecal administration of gadopentetate dimeglumine in 15 patients clinically suspected of having cerebrospinal fluid (CSF) rhinorrhea. By means of lumbar puncture, a single dose of 0.5 mL of gadopentetate dimeglumine was injected into the lumbar subarachnoid space. Thirteen patients showed leakage of contrast material through the cribriform plate into the ethmoid or sphenoid air cells. No leakage was observed in two patients. The study results show the relative safety and feasibility of low-dose gadolinium-enhanced MR cisternography in confirming the presence and determining the focus of active CSF leaks.

　

Index terms: Cerebrospinal fluid, leakage, 121.423, 167.423 • Cisternography, MR, 121.12143, 167.12175 • Fistula, cerebrospinal, 121.24, 167.24 • Magnetic resonance (MR), contrast media, 167.12143

	INTRODUCTION

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
Cerebrospinal fluid (CSF) rhinorrhea implies an abnormal communication between the subarachnoid space and the nasal cavity, with subsequent leakage of CSF through the anterior nasal apertures. CSF rhinorrhea is generally classified as traumatic, nontraumatic (ie, spontaneous), or postsurgical in origin (1). Most cases are traumatic, and the most common site of traumatic CSF rhinorrhea is the anterior cranial fossa, where the dura mater is particularly adherent to the thin overlying bone (2–4). CSF rhinorrhea occurs in 2%–3% of all cases of head injury and becomes clinically apparent within 48–72 hours after the traumatic incident. Approximately 70% of traumatic CSF fistulas close spontaneously without surgical intervention within 1 week after the injury (2,3). Despite early spontaneous closure, and even in the absence of gross CSF rhinorrhea, patients remain at risk for recurrent CSF leakage, pneumocephalus, and infectious meningitis (2). Precise identification of the location of the CSF fistula properly focuses surgical planning, optimizes the chance of a successful repair, and can ultimately lead to the prevention of subsequent infectious complications (5,6).

Despite advances in imaging and the availability of several different and potentially useful diagnostic methodologies, accurate demonstration of the site of the CSF leakage remains a challenge for radiologists and clinicians. Imaging techniques have evolved from conventional cranial radiography to polytomography, radionuclide cisternography, thin-section computed tomography (CT), and currently, intrathecal water-soluble iodinated contrast agent–enhanced CT cisternography (2–13). Currently, the most common method for evaluating a patient suspected of having CSF rhinorrhea is a combination of thin-section CT and subsequent CT cisternography. However, contrast-enhanced CT cisternography is not without risk of side effects, including headache, nausea, vomiting, seizures, allergic reaction, and rarely, intracerebral hemorrhage (14,15). Although thin-section CT is highly sensitive for detection of a fracture at the skull base, the actual site of the dural tear, and therefore the active CSF leak, is impossible to confirm with use of this technique alone. To be accurate, however, the less severe reactions observed in patients undergoing cisternographic studies (eg, headache, nausea, and vomiting) are due to the lumbar puncture, not the cisternographic contrast medium (16).

Results of recent studies support the application of magnetic resonance (MR) imaging with T2-weighted sequences without the need for intrathecal contrast agent to prove the presence of a CSF fistula. This support is based on the following criteria: (a) demonstration of an area of high signal intensity (ie, fluid) on T2-weighted images that extends directly into the paranasal sinuses "apparently" from the CSF cisterns at the skull base, (b) an "apparent" cortical bone defect involving the cribriform plate, and (c) brain herniation directly through the cribriform plate into the subjacent paranasal sinuses (2,4,17–25). However, some or all of these findings can sometimes be seen in the absence of fistula formation on MR images obtained for reasons other than CSF leak. In one imaging study, Hegarty and Millar (26) found that up to 42% of patients who were not clinically suspected of having a CSF fistula fulfilled one or more of the aforementioned MR criteria for this diagnosis. Other authors also have encountered false-positive so-called MR cisternographic studies (27).

The purpose of our study was to evaluate and report our experience in analyzing clinically suspected cranial CSF fistulas by using MR imaging combined with intrathecal administration of a gadolinium-based contrast agent—that is, gadolinium-enhanced MR cisternography.

	Materials and Methods

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion EFERENCES

 
Fifteen consecutive patients (eight male and seven female patients aged 9–68 years) clinically suspected of having active CSF rhinorrhea were included in this multicenter study. The protocol for this study was reviewed and approved at each of the three study sites by the respective institutional ethics committees or the equivalent institutional medical officers. Informed consent was obtained and placed in the patient’s chart in all cases. Nine patients experienced cranial trauma, and three patients had spontaneous CSF rhinorrhea (one case of pituitary adenoma, one case of congenital ethmoidal meningoencephalocele, and one case of benign intracranial hypertension). Three patients had previous anterior skull base surgery 1–8 years before the MR examination. Preceding thin-section CT studies showed evidence of anteromedial skull base fractures, defects, or erosions in each patient.

T1-weighted imaging (500/15 [repetition time msec/echo time msec], two signals acquired) and T2-weighted fast spin-echo imaging (3,500–4,000/40; echo train length of eight to 10) were performed in three orthogonal planes by using a 1.5-T MR unit (GE Medical Systems, Milwaukee, Wis). In all patients, an area of hyperintensity was seen in the ethmoid and/or sphenoid air cells on these preliminary T2-weighted images. By means of a lumbar puncture, 3–5 mL of CSF was withdrawn and mixed with a single volume of 0.5 mL of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) (469.01 mg gadopentetate dimeglumine per milliliter). This solution was then injected into the subarachnoid space, and the needle was removed.

Because the diluted gadolinium-based contrast agent could not be visualized with fluoroscopy during the injection procedure, the patients were positioned prone in the 30°–40° Trendelenburg position for 10–20 minutes after withdrawal of the needle to maximize the potential for contrast agent accumulation in the intracranial basal subarachnoid cisterns. This prolonged Trendelenburg positioning maneuver presumably resulted in an elevated hydrostatic pressure differential at the skull base and thereby improved the potential for leakage of the basal subarachnoid cistern contrast agent across a putative dural tear and cribriform dehiscence. Immediate T1-weighted images were then obtained with the patient prone, using parameters that were identical to those used for precontrast MR imaging. For analysis, gadolinium enhancement was sought in the areas of the ethmoid or sphenoid air cells and the nasal passageways by the coordinating member of the study group (J.R.J.).

All patients were hospitalized for an observation period of 24 hours after the examination. After they returned to the ward, hourly checks were made for gross behavioral alterations, neurologic impairment, changes in mental clarity, subjective complaints, and vital signs, as well as for more serious potential effects such as seizure activity and anaphylactoid reaction. These assessments were undertaken in light of baseline determinations before MR cisternography. In addition, a monthly clinical neurologic follow-up was performed for 6 or 12 months, depending on how long the patient had been tracked when the study was closed, for purposes of data analysis (12-month group, nine patients; 6-month group, six patients). This serial follow-up was undertaken by either the neuroradiologist (M.R., G.K., or E.T.T.) in charge of the project or alternatively by a consulting neurology staff officer.

	Results

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
No patient manifested gross behavioral changes, neurologic impairment, alterations in mental clarity, vital sign deviation from baseline, anaphylactic or other allergic reaction (eg, "hives"), or seizure activity at any time during the initial 24-hour observation period following MR cisternography or at monthly reevaluation through 6 or 12 months. Three patients experienced transient headache (<36 hours in duration) that resolved spontaneously by means of lying flat in bed.

In all patients, the gadopentetate dimeglumine entered the subarachnoid spaces at the base of the cranial cavity (ie, basal subarachnoid cisterns). Eleven patients showed leakage of gadolinium-enhanced fluid through the cribriform plate into the region of the ethmoid air cells (Figs 1, 2). Two patients had leakage into the sphenoid sinus. No leakage was observed in two patients: one with pituitary adenoma that had not been operated on previously and one following cranial trauma.

fig.ommitted Figure 1a. CSF rhinorrhea following head trauma in a 9-year-old girl. (a) Coronal thin-section CT scan shows defect in the right side of the cribriform plate (arrow) and opacification of the right ethmoid air cells. (b) Coronal T2-weighted fat-suppressed fast spin-echo MR image (4,000/100) shows inferior bifrontal contusions with encephalomalacia (open arrows) and hyperintensity (solid arrow) in the right ethmoid air cells. (c) Coronal T1-weighted MR cisternogram (500/15) obtained after intrathecal administration of gadopentetate dimeglumine shows downward herniation of the gyrus rectus (asterisk) and contrast enhancement (arrow) extending from the cranial subarachnoid space into the ethmoid air cell region, both on the right side. (d) Transverse T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows focal contrast material leakage (arrow) into the right midethmoid region. (e) Right parasagittal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (100/90) also shows focal ethmoid contrast material leakage (arrow).

 

fig.ommitted Figure 1b. CSF rhinorrhea following head trauma in a 9-year-old girl. (a) Coronal thin-section CT scan shows defect in the right side of the cribriform plate (arrow) and opacification of the right ethmoid air cells. (b) Coronal T2-weighted fat-suppressed fast spin-echo MR image (4,000/100) shows inferior bifrontal contusions with encephalomalacia (open arrows) and hyperintensity (solid arrow) in the right ethmoid air cells. (c) Coronal T1-weighted MR cisternogram (500/15) obtained after intrathecal administration of gadopentetate dimeglumine shows downward herniation of the gyrus rectus (asterisk) and contrast enhancement (arrow) extending from the cranial subarachnoid space into the ethmoid air cell region, both on the right side. (d) Transverse T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows focal contrast material leakage (arrow) into the right midethmoid region. (e) Right parasagittal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (100/90) also shows focal ethmoid contrast material leakage (arrow).

 

fig.ommitted Figure 1c. CSF rhinorrhea following head trauma in a 9-year-old girl. (a) Coronal thin-section CT scan shows defect in the right side of the cribriform plate (arrow) and opacification of the right ethmoid air cells. (b) Coronal T2-weighted fat-suppressed fast spin-echo MR image (4,000/100) shows inferior bifrontal contusions with encephalomalacia (open arrows) and hyperintensity (solid arrow) in the right ethmoid air cells. (c) Coronal T1-weighted MR cisternogram (500/15) obtained after intrathecal administration of gadopentetate dimeglumine shows downward herniation of the gyrus rectus (asterisk) and contrast enhancement (arrow) extending from the cranial subarachnoid space into the ethmoid air cell region, both on the right side. (d) Transverse T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows focal contrast material leakage (arrow) into the right midethmoid region. (e) Right parasagittal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (100/90) also shows focal ethmoid contrast material leakage (arrow).

 

fig.ommitted Figure 1d. CSF rhinorrhea following head trauma in a 9-year-old girl. (a) Coronal thin-section CT scan shows defect in the right side of the cribriform plate (arrow) and opacification of the right ethmoid air cells. (b) Coronal T2-weighted fat-suppressed fast spin-echo MR image (4,000/100) shows inferior bifrontal contusions with encephalomalacia (open arrows) and hyperintensity (solid arrow) in the right ethmoid air cells. (c) Coronal T1-weighted MR cisternogram (500/15) obtained after intrathecal administration of gadopentetate dimeglumine shows downward herniation of the gyrus rectus (asterisk) and contrast enhancement (arrow) extending from the cranial subarachnoid space into the ethmoid air cell region, both on the right side. (d) Transverse T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows focal contrast material leakage (arrow) into the right midethmoid region. (e) Right parasagittal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (100/90) also shows focal ethmoid contrast material leakage (arrow).

 

fig.ommitted Figure 1e. CSF rhinorrhea following head trauma in a 9-year-old girl. (a) Coronal thin-section CT scan shows defect in the right side of the cribriform plate (arrow) and opacification of the right ethmoid air cells. (b) Coronal T2-weighted fat-suppressed fast spin-echo MR image (4,000/100) shows inferior bifrontal contusions with encephalomalacia (open arrows) and hyperintensity (solid arrow) in the right ethmoid air cells. (c) Coronal T1-weighted MR cisternogram (500/15) obtained after intrathecal administration of gadopentetate dimeglumine shows downward herniation of the gyrus rectus (asterisk) and contrast enhancement (arrow) extending from the cranial subarachnoid space into the ethmoid air cell region, both on the right side. (d) Transverse T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows focal contrast material leakage (arrow) into the right midethmoid region. (e) Right parasagittal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (100/90) also shows focal ethmoid contrast material leakage (arrow).

 

fig.ommitted Figure 2a. Spontaneous CSF rhinorrhea from congenital ethmoidal meningoencephalocele in a 39-year-old woman. (a) Coronal T1-weighted MR image (90/10) shows downward ethmoid herniation of the left medial gyrus rectus (arrowhead) and a midline CSF signal intensity nasoethmoid lesion (asterisk). (b) Coronal T2-weighted fast spin-echo MR image (4,000/90) shows the generally hyperintense ethmoid lesion (black asterisk) and the downwardly displaced medial gyrus rectus (white asterisk) on the left side. (c) Coronal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows contrast material enhancement extending into the ethmoid meningocele (*) from the cranial subarachnoid space.

 

fig.ommitted   Figure 2b. Spontaneous CSF rhinorrhea from congenital ethmoidal meningoencephalocele in a 39-year-old woman. (a) Coronal T1-weighted MR image (90/10) shows downward ethmoid herniation of the left medial gyrus rectus (arrowhead) and a midline CSF signal intensity nasoethmoid lesion (asterisk). (b) Coronal T2-weighted fast spin-echo MR image (4,000/90) shows the generally hyperintense ethmoid lesion (black asterisk) and the downwardly displaced medial gyrus rectus (white asterisk) on the left side. (c) Coronal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows contrast material enhancement extending into the ethmoid meningocele (*) from the cranial subarachnoid space.

 

fig.ommitted Figure 2c. Spontaneous CSF rhinorrhea from congenital ethmoidal meningoencephalocele in a 39-year-old woman. (a) Coronal T1-weighted MR image (90/10) shows downward ethmoid herniation of the left medial gyrus rectus (arrowhead) and a midline CSF signal intensity nasoethmoid lesion (asterisk). (b) Coronal T2-weighted fast spin-echo MR image (4,000/90) shows the generally hyperintense ethmoid lesion (black asterisk) and the downwardly displaced medial gyrus rectus (white asterisk) on the left side. (c) Coronal T1-weighted intrathecal gadolinium-enhanced MR cisternogram (90/10) shows contrast material enhancement extending into the ethmoid meningocele (*) from the cranial subarachnoid space.

 
On the basis of the gadolinium-enhanced MR cisterogram findings in seven patients, surgery was performed, during which a dural patch was placed over the surgically visualized dehiscence of dura in the area of suspected CSF leakage. The CSF rhinorrhea halted after the surgery in each of the seven patients operated on and did not recur. The remaining patients continue to be followed conservatively, with surgery being contemplated if an eventual spontaneous cessation of rhinorrhea is not encountered or if infection intervenes.

	Discussion

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
In this pilot study, intrathecal gadopentetate dimeglumine administration was used to enhance CSF leakage through a putative dural defect in patients clinically suspected of having CSF rhinorrhea. Although the results of previously published studies have demonstrated that nonenhanced MR imaging has some utility in demonstrating CSF fistulas (28,29), there is a relatively high frequency (42%) of false-positive findings, especially in the presence of intercurrent, nonspecific paranasal sinus disease (26). False-negative findings have also been reported on nonenhanced MR images that subsequently revealed skull base fractures and frank CSF leaks at surgery (26). In support of gadolinium-enhanced MR cisternography, prior animal studies have shown the practicability of this technique in the visualization of the site of surgically induced nasoethmoidal CSF fistulas (30,31).

In the present study involving 15 patients with preceding positive T2-weighted MR imaging findings (ie, an area of hyperintensity in the ethmoid or sphenoid air cells) and positive thin-section CT findings (ie, anterior skull base fracture), gadolinium-enhanced MR cisternography was performed in all 15 patients and demonstrated positive findings in 13 patients and negative findings in two. Seven patients had surgical confirmation and repair of the dural tear. There were no false-positive gadolinium-enhanced MR cisternographic studies—that is, MR cisternograms did not reveal any case that was positive when in fact there was no tear in the dura mater in the cribriform plate or planum sphenoidale region at surgery. However, six patients with positive and two with negative MR cisternograms did not have surgical confirmation. Because these patients did not have operative evaluation, these eight cases at present cannot be proved to be either true or false with regard to gadolinium-enhanced MR cisternography, or with regard to T2-weighted MR imaging or thin-section CT, for that matter. Nevertheless, the accuracy rate of gadolinium-enhanced MR cisternography within the surgical group was 100%.

Many studies have documented the safety of intravenously administered gadopentetate dimeglumine (32), and several published animal experiments have demonstrated the seemingly safe use of intrathecal gadopentetate dimeglumine, especially when used at low doses (33–36). Recent experiments have also proved the safety of this technique in animals by means of various analyses (eg, electroencephalography, vascular physiologic response, and histology) (37). Furthermore, the results of initial human studies (38,39) have shown that the low doses of intrathecal gadopentetate dimeglumine that are adequate for diagnostic enhancement of the subarachnoid space of humans at MR imaging do not manifest clinical evidence of gross physical or neurologic abnormalities, CSF changes, or electroencephalographic alterations after gadolinium-enhanced MR cisternography. In addition, these patients did not show chronic MR imaging alterations of the brain at 1 year after gadolinium-enhanced MR cisternography. From a dose standpoint, the dose of gadopentetate dimeglumine used in this human study was almost 30 times smaller than that of the lowest dose that caused visible toxic changes in the underlying central nervous system parenchyma in laboratory animals: If the volume of gadopentetate dimeglumine containing 500 µmol of gadolinium per milliliter (the volume of gadopentetate dimeglumine as supplied) x 0.5 mL of gadopentetate dimeglumine (the volume of gadopentetate dimeglumine used in this study) = 250 µmol per 1,400 g (the average weight of the normal human brain), and 250 µmol per 1,400 g = 0.17 µmol per gram of brain tissue (the dose used in current study), then 5 µmol of gadopentetate dimeglumine (the lowest dose with toxic central nervous system alterations) ÷ 0.17 (the dose used in current study) = 29.4 times the lowest toxic dose (33,35,39). The doses used in this human study were in fact a human extrapolation of the doses used in several laboratory animal studies (31,33,37) that showed that the doses used for laboratory analysis were adequate for subarachnoid space visualization on MR images.

The postprocedural headaches encountered were consistent in type and frequency with those observed following either water-soluble iodinated contrast material–enhanced myelographic procedures or simple diagnostic lumbar puncture. Thus, this subjective complaint is explained by the lumbar subarachnoid space tap (ie, iatrogenic CSF space postural hypotension associated with needle puncture–related transient CSF leak) and not gadolinium-enhanced MR cisternography.

In conclusion, intrathecal gadolinium-enhanced MR cisternography is a promising technique that may permit direct, sensitive visualization of the site of spontaneous, posttraumatic, or postsurgical CSF leakage. It is also apparent that thin-section CT is complementary to gadolinium-enhanced MR cisternography and therefore should be performed in all cases. No comparison studies with CT cisternography were performed, however, because it was considered unethical to have the patients undergo a second intrathecal contrast-enhanced cisternographic study solely for the sake of data collection and comparison.

Alternative methods that may be useful in future studies to compare and/or couple with the present gadolinium-enhanced MR cisternographic technique include flow-sensitive acquisitions (eg, diffusion-weighted MR imaging) (40) and positional MR imaging (41,42).

Intrathecal administration of a gadolinium-based contrast agent is not currently approved worldwide. Furthermore, our study represents an investigation of the use of only one gadolinium preparation—gadopentetate dimeglumine. Because other gadolinium-based contrast agents exist, our study findings cannot be generalized to the routine intrathecal use of all gadolinium-based contrast agents.

　

	ACKNOWLEDGMENTS

 
The authors thank Tina Clifton for manuscript transcription.

	REFERENCES

Top ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 

1. Park J, Strelzow V, Friedman W. Current management of cerebrospinal fluid rhinorrhea. Laryngoscope 1983; 93:1924-1300.

2. Stafford-Johnson DB, Brennan P, Toland J, O’Dwyer AJ. Magnetic resonance imaging in the evaluation of the cerebrospinal fluid fistula. Clin Radiol 1996; 51:837-841.

3. Colquhoun IR. CT cisternography in the investigation of cerebrospinal rhinorrhea. Clin Radiol 1993; 47:403-408.

4. Wakhloo AK, van Velthoven V, Shumacher M, Krauss JK. Evaluation of MR imaging, digital subtraction cisternography, and CT cisternography in diagnosing CSF fistula. Acta Neurochir (Wien) 1991; 111:119-127.

5. Manelfe C, Cellerier P, Sobel D, Prevost C, Bonafe A. Cerebrospinal fluid rhinorrhea: evaluation with metrizamide cisternography. AJR Am J Roentgenol 1982; 138:471-476.

6. Nickaus P, Dutcher PO, Kido DK, Hengerer AS, Nelson CN. New imaging techniques in the diagnosis of cerebrospinal fluid fistula. Laryngoscope 1998; 98:1065-1068.

7. Curnes JT, Vicent LM, Kowalsky RJ, McCartney WH, Staab EV. CSF rhinorrhea: detection and localization using overpressure cisternography with TC-99m-DTPA. Radiology 1985; 154:795-799.

8. Lloyd MNH, Kimber PM, Burrows EH. Posttraumatic cerebrospinal fluid rhinorrhea: modern high-resolution computed tomography is all that is required for effective demonstration of the site of leakage. Clin Radiol 1994; 49:100-103.

9. Ahmadi J, Weiss MH, Segall HD, Schultz DH, Zee CS, Giannotta SL. Evaluation of cerebrospinal fluid rhinorrhea by metrizamide computed tomography. Neurosurgery 1985; 16:54-59.

10. Drayer BP, Wilkins RH, Boehnke M, Horton JA, Rosenbaum AE. Cerebrospinal fluid rhinorrhea demonstrated by metrizamide CT cisternography. AJR Am J Roentgenol 1977; 129:149-151.

11. Ozgen T, Tekkok IH, Cila A, Erzen C. CT cisternography in evaluation of cerebrospinal fluid rhinorrhea. Neuroradiology 1990; 32:481-484.

12. Naidich TP, Moran CJ. Precise anatomic localization of atraumatic sphenoethmoidal cerebrospinal fluid rhinorrhea by metrizamide CT cisternography. J Neurosurg 1980; 53:222-228.

13. Stone JA, Castillo M, Neelon B, Mukherji SK. Evaluation of CSF leaks: high-resolution CT compared with contrast-enhanced CT and radionuclide cisternography. AJNR Am J Neuroradiol 1999; 20:706-712.

14. Sand T, Myhr G, Stovner LJ, Dale LG. Side effects after lumbar iohexol myelography: relation to radiological diagnosis, sex and age. Neuroradiology 1990; 31:523-528.

15. Van de Kelft E, Bosmans J, Paziel P, Van Vyve PM, Selosse P. Intracerebral hemorrhage after lumbar myelography with iohexol: report of a case and review of the literature. Neurosurgery 1991; 28:570-574.

16. Woodcock RJ, Marx WF, Johnson RM, Lowe JM, Lipper MH, Kallmes DF. Needle diameter in outpatient myelography: rates of adverse effects and current practice trends. Neuroradiology 2000; 42:371-374.

17. Murata Y, Yamada I, Suzuki S, Isotani E. MRI in spontaneous cerebrospinal fluid rhinorrhea. Neuroradiology 1995; 37:453- 455.

18. Muftah S, Eljamel M, Christopher N, et al. MRI cisternography, and the localization of CSF fistula. Br J Neurosurg 1994; 8:433-437.

19. Eberhardt KEW, Hollenbach HP, Deimling M, Tomandl BF, Huk WJ. MR cisternography: a new method for the diagnosis of CSF fistulae. Eur Radiol 1997; 7:1485-1491.

20. El Gammal T, Brooks BS. MR cisternography: initial experience in 41 cases. AJNR Am J Neuroradiol 1994; 15:1647-1656.

21. Eljamel MS, Pidgeon CN, Toland JB, Phillips J, O’Dwyer AJ. MRI cisternography and the localization of CSF fistulae. Br J Neurosurg 1994; 8:433-437.

22. Zlab MK, Moore GF, Daly DT, Yonkers AJ. Cerebrospinal fluid rhinorrhea: a review of the literature. Ear Nose Throat J 1992; 71:314-317.

23. Johnson DBS, Brennan P, Toland J, O’Dwyer AJ. Magnetic resonance imaging in the evaluation of cerebrospinal fluid fistulae. Clin Radiol 1996; 51:837-841.

24. Eberhardt KEW, Deimling M, Hollenbach HP, Huk WJ. MR-diagnostik bei liquorfisteln mit einer 3D-CISS-sequenz. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1997; 167:605-611.

25. Gupta V, Goyal M, Mishra N, Gaikwad S, Sharma A. MR evaluation of CSF fistulae. Acta Radiol 1997; 38:603-603.

26. Hegarty SE, Millar JS. MRI in the localization of CSF fistulae: is it of any value?. Clin Radiol 1997; 52:768-770.

27. Chinoy S, Jinkins JR, Shroff MM. MR vs CT cisternography in the evaluation of sphenoethmoidal cerebrospinal fluid leak. Asian Oceanian J Radiol 1997; 2:77-84.

28. El Gammal T, Sobol W, Wadlington VR, et al. Cerebrospinal fluid fistula: detection with MR cisternography. AJNR Am J Neuroradiol 1998; 19:627-631.

29. Shetty PG, Shroff MM, Sahani DV, Kirtane MV. Evaluation of high-resolution CT and MR cisternography in the diagnosis of cerebrospinal fluid fistula. AJNR Am J Neuroradiol 1998; 19:633-639.

30. Di Chiro G, Girton ME, Frank JA, et al. Cerebrospinal fluid rhinorrhea: depiction with MR cisternography in dogs. Radiology 1986; 160:221-222.

31. Ibarra R, Jinkins JR, Korvick D, Xiong L, Gao JH. Evaluation of intrathecal gadolinium-enhanced MR cisternography in a rabbit model of traumatic nasoethmoidal CSF fistula. J Magn Reson Imaging 2000; 11:20-24.

32. Niendorf HP, Haustein J, Cornelius I, Alhassan A, Claub W. Safety of gadolinium-DTPA: extended clinical experience. Magn Reson Med 1991; 22:222-228.

33. Jinkins JR, Williams RF, Xiong L. Evaluation of gadopentetate dimeglumine magnetic resonance cisternography in an animal model. Invest Radiol 1999; 34:156-159.

34. Ray DE, Cavanagh JB, Nolan CC, Williams SCR. Neurotoxic effects of gadopentetate dimeglumine: behavioral disturbance and morphology after intracerebroventricular injection in rats. AJNR Am J Neuroradiol 1996; 17:365-373.

35. Skalpe IO, Tang GJ. Magnetic resonance imaging contrast media in the subarachnoid space: a comparison between gadodiamide injection and gadopentetate dimeglumine in an experimental study in pigs. Invest Radiol 1997; 32:140-148.

36. Skalpe IO. Is it dangerous to inject MR contrast media into the subarachnoid space? (letter). Acta Radiol 1998; 39:100.

37. Toney GM, Chavez HA, Ibarra R, Jinkins JR. Acute and subacute physiological and histological studies of the central nervous system after intrathecal gadolinium injection in the anesthetized rat. Invest Radiol 2001; 36:33-40.

38. Zeng QY, Xiong L, Jinkins JR, Liu Z, Fan Z. Intrathecal gadolinium-enhanced MR myelography: a pilot study in human patients (abstr). Radiology 1997; 205(P):486.

39. Zeng QY, Xiong L, Jinkins JR, Fan Z, Liu Z. Intrathecal gadolinium-enhanced MR myelography and cisternography: a pilot study in human patients. AJR Am J Roentgenol 1999; 173:1109-1115.

40. Levy LM, Gulya AJ, Davis SW, LeBihan D, Rajan SS, Schellinger D. Flow sensitive magnetic resonance imaging in the evaluation of cerebrospinal fluid leaks. Am J Otol 1995; 16:591-596.

41. Gupta V, Goyal M, Mishra NK, Sharma A, Gaikwad SB. Positional MRI: a technique for confirming the site of leakage in cerebrospinal fluid rhinorrhea. Neuroradiology 1997; 39:818-820.

42. Jinkins JR, Green C, Damadian R. Upright, weight-bearing, dynamic-kinetic MRI of the spine: pMRI/kMRI (abstr). Riv Neuroradiol 2001; 14:135.