The Dural Tail Sign in the Diagnosis of Meningiomas

Introduction

DIAGNOSIS OF specific tumor type from magnetic resonance (MR) imaging characteristics is sometimes difficult in both people and animals and definitive diagnosis may depend upon incisional or excisional biopsies. One particular MR imaging sign, the aural tail, has been reported to be specific for meningiomas in humans, occurring in approximately 50-60% of subjects.l-8 Detection of the dural tail allows the diagnosis of a meningioma to be made with reasonable confidence, although this sign has been reported to occur occasionally with other diseases.3,8-10
 
 The dural tail is a linear enhancement of thickened dura mater adjacent to an extra-axial mass seen on gadolinium-diethylenetriaminepentacetic acid (Gd-DTPA) enhanced T1 weighted images (Fig. 1). A number of criteria for diagnosing a dural tail have been described: it should be continuous with the associated extra-axial mass, it should enhance to an equal or greater degree than the associated mass and it should be seen in at least two contiguous slices or in two imaging planes to preclude confusion with vascular structures (Fig. 2).4

 The dural tail is well recognized in people but the only mention of the dural tail in animals is in a review of the MR imaging features of thirteen meningiomas.14 The author noted the presence of a dural tail in three of these tumors but this observation was not discussed. Our retrospective study was designed to determine the incidence of the dural tail with canine and feline meningiomas. Secondly we sought to determine if this sign has similar specificity for meningiomas in companion animals as in humans.
 

 Materials

 MR examinations of 18 dogs and four cats for which a definitive diagnosis was available were reviewed. The MR examinations were performed with a l.5T superconducting magnetic resonance scanner using a human knee coil. The patients were imaged over the period from February 1991 to January 1996. The protocols employed varied slightly. The spin echo sequences most often employed were repetition time (TR) 550 ms and time to echo (TE) 17 ms for T1 weighted images and TR of 6000 ms and TE of 92 ms for T2 weighted images. Proton density weighted images were not routinely obtained. Slices were usually 3 or 4 mm thick and noncontiguous with a 1 mm interslice spacing. The images were displayed in a 256 x 192 matrix. We most commonly employed an imaging sequence consisting of T2 weighted dorsal images and T1 weighted transverse images followed by T1 weighted transverse and sagittal images after bolus administration of 0.1 mmol/kg of Gd-DTPA. Additional imaging planes were sometimes used based upon clinical suspicions or lesions identified on initial images.

 Seventeen (13 dogs and 4 cats) of the 22 patients had an intracranial neoplasm. All had mass lesions which were extra-axial or intra-axial with meningeal contact. There were ten meningiomas (seven dogs and three cats) and one example each of a pituitary macrotumor, chromophobe adenocarcinoma, oligodendroglioma, astrocytoma, osteosarcoma (a cat), plasma cell tumor and choroid plexus papilloma. The pituitary tumor was diagnosed on the basis of clinical and biochemical changes consistent with pituitary dependent hyperadrenocorticism, a mass lesion in the pituitary region and resolution of clinical and biochemical abnormalities following irradiation of the mass. The other neoplasms were confirmed histologically either from excisional biopsies or necropsy material.

 The other five patients, all dogs, had inflammatory CNS disease. All had mass lesions which abutted or involved the meninges. One of these, toxoplasma meningoencephalitis, was confirmed by histologic examination of tissue obtained at necropsy. In three dogs, a presumptive diagnosis of distemper meningoencephalitis was made based on necrologic signs and the presence of IgM distemper antibodies in serum and/or the presence of IgG distemper antibodies in CSF15,16 with no history of recent vaccination. One of these dogs also tested positive for Cryptococcus antigen in its CSF. Whether the MR lesions seen were due to cryptococcosis or distemper or both was not determined. Bacterial meningoencephalitis was diagnosed in one dog which was thought to have arisen by extension from a retrobulbar abscess. This diagnosis was based upon clinical signs, cytologic and biochemical changes in the CSF and resolution of clinical signs and MR changes following antibiotic therapy.
 

Methods

 Three radiologists (SMN, AKV and GDR) reviewed the MR images without knowledge of the final diagnoses. They were specifically asked to evaluate the T1 weighted Gd-DTPA enhanced images for the presence of a dural tail associated with a mass lesion. The conclusions of the reviewers were compared to the final diagnosis for each patient.

 For the purpose of statistical analysis, each patient was assigned a numerical score which corresponded to the number of reviewers reporting a dural tail for that patient. A Wilcoxon rank sum test (exact version) was performed to test the hypothesis that there was a significant difference between the median scores of the patients with meningiomas versus the other patients.
 

Results

 There was some variation in the frequency with which individual observers reported a dural tail (Table 1). Each observer identified a dural tail in one patient having a lesion other than a meningioma, two of the three reported a dural tail associated with a chromophobe adenocarcinoma and one reported a dural tail with a Toxoplasma mass lesion. All three reported a dural tail with four of the meningiomas. Of the three feline meningiomas, all three observers reported a dural tail in two patients and two observers reported a dural tail in the third cat. All of the other patients having a dural tail reported by one or two of the reviewers had a meningioma. The average scores for the meningioma group were higher than the scores in the group consisting of all the other diseases (p = 0.0023).
 

Discussion

 Several reports describe the aural tail sign associated with meningiomas and occasionally with other lesions in humans. In MR images of eighteen humans with meningioma the aural tail sign was detected in thirteen.1 In another report thirty meningiomas, seven acoustic neuromas, seventeen superficially located gliomas and seventeen supra- and infratentorial superficial metastatic lesions which abutted the dural surface were examined.6 A dural tail was found with eighteen of thirty meningiomas but with no other tumor type and it was concluded that the dural tail sign was specific for meningiomas. Other reports have, however, described a dural tail associated with nonmeningiomatous lesions, both neoplastic and inflammatory. For example, in 16 humans with a dural tail, three did not have a meningioma (chloroma, lymphoma, sarcoidosis).3 In another review of 29 meningiomas and 24 other tumor types, a dural tail was found with fifteen of the meningiomas, an oligodendroglioma and a neurinoma.5 Additionally, MR images of over three hundred nonmeningiomatous malignant lesions were analyzed and an appearance consistent with a dural tail was found adjacent to ten (six glioblastomas, three superficial parenchymal metastatic deposits and one dural metastasis).9 Although it appeared initially that the dural tail was specific for meningioma it has now been shown to occur occasionally with other diseases. Nonetheless it occurs with 50-60% of meningiomas and it is considered a useful MR imaging sign in people.8

 There is some dispute in people about the histologic changes associated with the dural tail. In some reports histologic examination has shown that the dural tail is due to direct tumor extension or thickened dura which contains meningothelial tumor nodules.6,7 The dural tail often extends beyond the region of tumor infiltration (2-3 mm) but nodules may be present some distance from the neoplasm. In other studies correlating MR findings with histopathologic appearance it was concluded that the dural tail represented hypervascular, nonneoplastic reaction. No evidence of tumor extension was detected within the dural tail, being confined to the junction of the tumor and the dural tail.2,4

 Whether the dural tail represents neoplastic tissue is an important question. Multiple meningiomas occur relatively commonly in humans17,18 and cats19,20 and occasionally in dogs.12,20 In humans, multiple meningiomas have been described as multiple masses at various locations within the cranium (distant multicentricity) or as a cluster of mass lesions at a single location (regional multicentricity). In humans, tumor recurrence following an apparent total excision of a histologically benign meningioma is common.l7 It has been suggested that this is a manifestation of regional multicentricity with a single large dominant mass and multiple small or microscopic tumors in the same area. The dural tail seen on MR images may represent multiple small or microscopic tumors and reactive tissue outside the margins of the principal mass. The presence of such neoplastic tissue would influence treatment planning, specifically the margin of excision required to prevent recurrence and whether postoperative radiotherapy would be required. Therefore the detection of a dural tail on MR images is not only a diagnostic sign but it may have therapeutic and prognostic implications.

 The number of reports describing the MRI appearance of intracranial neoplasms in animals has increased in recent years. Meningiomas are among the most commonly reported primary CNS neoplasms both in surgical reviews,21-24 imaging reportsl3,25,26 and postmortem surveys.12,l9,27,28 Meningiomas in dogs and cats have a relatively well defined and constant MR appearance very similar to human tumors, although atypical examples may be encountered. These tumors are usually isointense or hypointense on T1 weighted images, hyperintense on T2 weighted images and will enhance markedly following administration of Gd-DTPA.13 However, the presence of a dural tail associated with meningiomas is mentioned in only one report.14 Perhaps because we actively looked for the dural tail sign, we detected it in a larger proportion of patients with meningioma, ranging from four to eight of the ten patients, depending upon the observer (Table 1). The variability in the rate of reporting a dural tail between observers probably reflects individual tendencies to underread or overread and a general unfamiliarity with this sign. Based on our small sample the dural tail appears to be relatively specific for meningiomas.

 The potential for misdiagnosing a dural tail exists. There were two instances of nonmeningiomatous disease in which at least one observer reported a dural tail. One of the observers recorded a dural tail in a dog which had toxoplasma meningoencephalitis. This animal had multifocal abnormalities involving the entire right cerebral hemisphere. An area of faint meningeal or dural enhancement adjacent to a poorly defined, apparently intra-axial, mass with patchy minimal enhancement was reported as a dural tail (Fig. 3). In retrospect similar meningeal or dural enhancement could be seen at other sites overlying the affected cerebral hemisphere and not specifically associated with any mass lesion (Fig. 4). These lesions most likely represented foci of inflamed meningeal tissue. The suspected dural tail did not fulfill all the defining criteria because of its relatively poor enhancement and the location of the associated mass. In addition it was not continuous with the mass. The presence of multifocal meningeal changes is also atypical for the dural tail and should probably have ruled out this diagnosis. In the second patient, two observers recorded a dural tail with a chromophobe adenocarcinoma. This was a large, aggressive and invasive tumor and the meningeal enhancement was seen in the region where the tumor invaded the caudal orbit from the calvarium (Fig. 5). This lesion had some of the characteristics of a dural tail, i.e. continuous with the mass and good enhancement. However most of the lesion lay between the mass and the bony calvarium rather than extending away from the mass. Extension beyond the mass was seen in one slice only while a dural tail should be visible in two adjacent slices or two imaging planes. In this patient the meningeal enhancement was most likely due to invasion and destruction of the meninges by the expanding tumor mass.

 Trauma, surgery, hemorrhage, infection and metastasis have all been reported to cause meningeal enhancement following intravenous contrast medium administration. In the two patients without a meningioma but having a dural tail reported the meningeal enhancement could be attributed to inflammation and infection or neoplastic invasion. A1though these patients were thought to have a dural tail, in retrospect the meningeal enhancement was considered atypical for this sign. These diagnoses may reflect our relative inexperience and unfamiliarity with the dural tail sign.

 In conclusion, the dural tail sign appears to occur relatively commonly with canine and feline meningioma although the rate of detection varied from 40 to 80%. In most patients, the location and appearance of a lesion will allow a presumptive diagnosis of a meningioma to be made. Detecting a dural tail may improve the confidence of the radiologist interpreting the images in making this diagnosis. In humans, the dural tail is most useful in identifying atypical meningiomas and this may eventually prove to be true regarding companion animals also. The presence of a dural tail associated with an extra-axial mass should prompt the diagnosis of meningioma. It remains to be resolved whether neoplastic tissue is present in the dural tail and if possible tissue from this area should be examined histologically. Our sample was relatively small and contained only single examples of various nonmeningiomatous tumors and no patients with metastatic disease. A larger and more varied sample population would help to evaluate the specificity of this sign.

Table 1. Dural Tail Identification by Reviewers

 

	Reviewer 1		Reviewer 2		Reviewer 3
	Positive	Negative	Positive	Negative	Positive	Negative
Meningiomas	4	6	6	4	8	2
Nonmeningiomatous	1	11	1	11	1	11

This table shows the performance of the individual reviewers in detecting a dural tail with meningiomas and other disease.

References

 1. Aoki S, Sasaki Y, Michida T, Tanioka H. Contrast enhanced MR images in patients with meningioma: importance of enhancement of the dura adjacent to the tumor. Am J of Neurorad 1990;11:935-938.

 2. Tokomaru A, O'uchi T, Eguchi T. Prominent meningeal enhancement adjacent to meningiomas on Gd-DTPA enhanced MR images: histopathologic correlation. Radiol 1990;175:431-433.

 3. Tien RD, Yang PJ, Chu PK. "Dural tail sign": a specific MR sign for meningioma? J of Comp Assist Tomo 1991;15:64-66.

 4. Nägele T, Petersen D, Grodd W, Opitz H, Voigt K. The "dural tail" adjacent to meningiomas studies by dynamic contrast-enhanced MRI: a comparison with histopathology. Neuroradiol 1994;36:303-307.

 5. Schörner W, Schubeus P, Henkes H, Lanksch W, Felix R. Meningeal sign: a characteristic finding on meningiomas on contrast enhanced MR images. Neuroradiol 1990;32:90-93.

 6. Goldsher D, Litt AW, Pinto RS, Bannon KR, Kricheff II. Dural "tail" associated with meningiomas on Gd-DTPA enhanced MR images: characteristics, differential diagnostic value, and possible implications for treatment. Radiol 1990;176:447-450.

 7. Wilms G, Lammens M, Marchal G. Thickening of the dura surrounding meningiomas: MR features. J of Comp Assist Tomo 1989;13:763-768.

 8. Hicks RJ. Supratentorial brain tumors. In: Edelman RR, Zlatkin MB, Hesselink JR. Clinical Magnetic Resonance Imaging. Philadelphia: Saunders, 1996:533-556.

 9. Wilms G, Lammens M, Marchal G. Prominent dural enhancement adjacent to nonmeningiomatous malignant lesions on contrast enhanced MR images. Am J of Neuroradiol 1991,12:761-764.

10. Paz-Fumagalli R, Daniels DL, Millen SJ, Meyer GA, Thieu TM. Dural "tail" associated with an acoustic schwannoma in MR imaging with gadopentate dimeglumine. Am J of Neuroradiol 1991;12:1206.

11. Braund KG, Ribas JL. Central nervous system meningiomas. Comp Cont Ed Pract Vet 1986;8(4):241-247.

12. Patnaik AK, Kay WJ, Hurvitz AI. Intracranial meningioma: a comparative pathologic study of 28 dogs. Vet Path 1986;23:369-373.

13. Thomas WB, Wheeler SJ, Kramer R, Kornegay JN. Magnetic resonance features of primary brain tumors in dogs. Vet Radiol & Ultrasound 1996;37(1):20-27.

14. Hathcock JT. Low field magnetic resonance imaging characteristics of cranial vault meningiomas in 13 dogs. Vet Radiol & Ultrasound 1996;37(4):257-263.

15. Greene CE, Appel MJ. Chapter 16. Canine distemper. In: Greene CE. Infectious diseases of dogs and cats. Philadelphia: Saunders, 1990:226 239.

16. Thomas WB, Sorjonen DC, Steiss JE. A retrospective evaluation of 38 cases of canine distemper encephalomyelitis. J Am Anim Hosp Assoc 1993;29(2):129-133.

17. Borovich B, Doron Y. Recurrence of intracranial meningiomas: the role played by regional multicentricity. J Neurosurg 1986;64(1):58-63.

18. Borovich B, Doron Y, Braun J. The incidence of multiple meningiomas: do solitary meningiomas exist? Acta Neurochirurgica 1988;90(1):15-22.

19. Nafe LA. Meningiomas in cats: a retrospective clinical study of 36 cases. J Am Vet Med Assoc 1979;174(11):1224 1227,

20. LeCouteur RA. Tumors of the nervous system. In: Withrow SJ, MacEwen KG. Small animal clinical oncology. Philadelphia: Saunders, 1996.

21. Lawson DC, Burk RL, Prata RG. Cerebral meningioma in the cat: diagnosis and surgical treatment of ten cases. J Am Anim Hosp Assoc 1984;20(2):333-342.

22. Gallagher JG, Berg J, Knowles KE, Williams LL, Bronson RT. Prognosis after surgical excision of cerebral meningiomas in cats: 17 cases (1986 1992). J Am Vet Med Assoc 1993;203(10):1437-1440.

23. Niebauer GW, Habil MV, Dayrell DB, Speciale J. Evaluation of craniotomy in dogs and cats. J Am Vet Med Assoc 1991;198(1):89-95.

24. Gordon LE, Thacher C, Matthiesen DT, Joseph RJ. Results of craniotomy for the treatment of cerebral meningioma in 42 cats. Vet Surg 1994;23(2):940-100.

25. Turrel JM, Fike JR, LeCouteur RA, Higgins RJ. Computed tomographic characteristics of primary brain tumors in 50 dogs. J Am Vet Med Assoc 1986;188(8):851-856.

26. Dykes NL, Warnick LD, Summers B, Wallace RJ, Kallfelz FA. Retrospective analysis of brain scintigraphy in 116 dogs and cats. Vet Radiol & Ultrasound 1994;35(1):59 65.

27. Schulman FY, Ribas JL, Carpenter JL, Sisson AF, LeCouteur RA. Intracranial meningioma with pulmonary metastasis in three dogs. Vet Path 1992;29(3):196-202.

28. Patnaik AK, Lieberman PH, Erlandson RA, Shaker E, Hurvitz AI. Paranasal meningioma in the dog: a clinicopathologic study of ten cases. Vet Path 1986;23(4):362-368.

FIG.1. Canine meningioma, Tl weighted sagittal image following Gd-DTPA administration. A well defined hyperintense mass is seen dorsal to the cerebellum. A dural tail extends caudally from the mass towards the foremen magnum (arrow).

FIG. 2. Canine meningioma, Tl weighted transverse images following Gd-DTPA administration. A well defined mass with associated dural tails (arrowheads) is seen in this transverse image (A). In the two slices immediately rostral to this (B and C), the mass is no longer visible but the dural tail is still clearly seen (arrowheads).

FIG. 3. Dog with toxoplasma meningoencephalitis, Tl weighted transverse images following Gd-DTPA administration. Patchy, ill defined, relatively poor contrast enhancement is present in the right olfactory lobe (arrows). Slight enhancement of the dura or meninges is also present (arrowheads); however, the enhancement is poor and is not continuous with the mass lesion.

FIG. 4. Same dog as in Fig. 3, transverse section through cerebrum, T weighted transverse image following Gd-DTPA administration. Faint enhancement of the meninges is noted overlying the right cerebral hemisphere (arrows).

FIG. 5. Canine chromophobe adenocarcinoma, transverse Tl weighted image following Gd-DTPA administration. A small amount of neoplastic tissue which has been enhanced is seen at the ventrolateral aspect of the olfactory lobe. Enhancement of the dura or meninges is seen between the  intracranial neoplastic tissue and the calvarium.

  Hop Back to the Neurology Journal Club Page!