Angiographic demonstration of a cerebral venous angioma Case report
RANDALL S. PREISSlG, M.D., SANDRA H. PREISSIG, M.D., JOHN A. GOREE, M.D.
AND
Departments of Radiology and Pathology, Duke University Medical Center, Durham, North Carolina ~" This patient presented with a cerebral venous angioma in which the angiographic findings were suggestive of a malignant brain tumor; the true nature of this lesion was documented at surgery. The diagnostic features of this case are discussed. KEY WORDS angiography
9 vascular malformation
9 cerebral venous angioma
9
NGIOGRAPHICdifferentiation of an arteriovenous malformation from malignant glioma will permit accurate diagnosis from the cerebral angiogram alone in most instances? We recently examined a patient with venous angioma of the brain with confusing angiographic features. A review of the literature reveals only a few instances of venous angioma demonstrated by angiography.
gram revealed some generalized paroxysmal activity suggestive of epilepsy. Radionuclide cerebral angiogram demonstrated a focus of abnormally increased activity deep within the right hemisphere during the arterial phase which cleared late in the venous phase (Fig. 1). A static scan immediately following angiography was normal. The combination of increased activity on the radionuclide cerebral angiogram with a normal static scan was interpreted as highly suggestive of aneurysm or arteriovenous malformation. Case Report A right carotid angiogram (Fig. 2) showed This 23-year-old woman had had head- a lesion in the white matter of the right fronaches for the past 5 years. Ten weeks before tal lobe immediately above the roof of the admission, the headaches increased in lateral ventricle. The abnormal area was first severity and began localizing in the right fron- identified as a diffuse blush, measuring 3.5 tal region; during that time she complained of cm in diameter. There were no hypertrophic decreased memory. arterial trunks or arteriovenous shunting Examination. Neurological examination present. The lesion drained into enlarged, showed only slightly decreased strength and parallel, medullary veins which converged at sensation on the left. An electroencephalo- the roof of the lateral ventricle; the entire le-
A
628
J. Neurosurg. / Volume 44 / May, 1976
Cerebral venous angioma
FI6. 1. Radionuclide cerebral angiogram demonstrating increased activity deep within the right hemisphere during the capillary and venous phases.
sion was in the shape of an inverted pyramid. A single large vein appeared to drain the confluence of the medullary veins and then emptied into the superior sagittal sinus. No drainage of the lesion by the deep venous system was noted and no mass effect was demonstrated. The angiographic circulation time of the lesion was normal. The angiographic diagnosis was malignant glioma based on the location of the lesion in the white matter, and an intrinsic vascular pattern characterized by clear spaces separating the individual abnormal vessels from each other. Operation. The patient underwent a right frontal craniotomy. A large red vein that drained into the superior sagittal sinus was noted to rise from within the frontal lobe between the middle and superior frontal gyri (Fig. 3). A 3 • 9-cm block resection of the involved area was performed. Numerous "thin walled vessels" and veins were noted during resection. Pathological examination revealed an angiomatous malformation consisting entirely of venous structures separated by normal neural tissue (Fig. 4). The patient has done well for 2 years on medication for control of her seizures.
Discussion
Angiography of arteriovenous malformations (AVM's) shows that they are composed of dilated arteries and veins with gross shunts between them; histologically they contain abnormal, hyalinized vessels which are "neither artery or vein, ''~ in addition to arteries and veins. In contrast, no abnormal arteries or arteriovenous shunts were seen at angiography in the present case. Microscopically, the lesion was composed of numerous thinwalled vessels separated by normal brain parenchyma that conform to McCormick's description of venous angiomas. ~ The current case represents a venous angioma and not an AVM, since it contains no arterial elements and the arteries in the brain substance surrounding the lesion are normal. This explains why no abnormal arteries were seen at angiography and why no arteriovenous shunts were demonstrated. In the absence of angiographic evidence for arteriovenous shunting, the red color noted at surgery of the thin-walled vein draining the angioma is difficult to explain. It may be that this red color reflects decreased percent oxy-
FIG. 2. Selected films of right carotid angiograms, lateral projection. After a normal early arterial phase, a diffuse "blush" in the frontal lobe is seen. Multiple parallel veins converge at the roof of the lateral ventricle during the early venous phase and a single large vein then drains this confluence into the superior sagittal sinus.
J. Neurosurg. / Volume 44 / May, 1976
629
R. S. Preissig, S. H. Preissig a n d
J.
A. G o r e e
this, we can find only a few instances in which a venous angioma was detected by angiography,la,8 Noran" mentioned that the venous angioma is characteristically pyramidal, the apex of the pyramid being at the ventricle. Courville stated that these lesions are rarely multiple and that "one sometimes finds a large vascular channel of uniform diameter passing from the subcortical lesion directly through the cerebral or cerebellar cortex to the superficial meningeal veins. Such a venous shunt must serve as a bypass to evacuate the venous blood from the cluster of anomalous veins to the exterior. ''~ This combination of descriptions is demonstrated in the current case, in FIG. 3. At surgery, the large vein demon- which the angiograms first showed a "blush" strated at angiography was noted to rise from within the frontal lobe (smaller arrow) and drain without evidence of arteriovenous fistula. Multiple parallel vessels were then seen draininto the superior sagittal sinus (larger arrow). ing this area toward the ventricular roof in an gen extraction due to the relatively increased inverted pyramid configuration; these veins blood volume within this lesion as demon- appeared to be dilated medullary veins. Then a single large vessel drained the lesion from strated by the blush at angiography. Courville2 described a series of 22 cerebral the confluence of these veins (the apex of the vascular malformations which fulfilled the pyramid at the ventricular roof) to the criteria of venous angiomas, dispelling the superior sagittal sinus (Fig. 2). It is interesting to note that the diagnosis of belief that these lesions are rare.' Despite "vascular malformation" was made at the time of the radionuclide cerebral angiogram and static brain scan (Fig. 1). The primary reason for that interpretation was that to our knowledge, no convincing case has been reported in which a malignant brain tumor has demonstrated increased flow on the radionuclide angiogram together with a normal static brain scan. At cerebral angiography, the demonstration of clear spaces separating abnormal vessels located primarily within the white matter is characteristic of brain tumor? However, the venous drainage of a tumor in this location classically should be predominantly by way of the subependymal veins of the lateral ventricle and to the deep venous system, rather than through a single vein to the superficial venous system. An enlarged draining vein as was seen in this lesion can be seen in an AVM or malignant tumor, but these lesions would not demonstrate enlarged medullary veins without concomitant arteriovenous shunting. The absence of a mass effect in a lesion 3.5 cm in diameter is atypical for a F~o. 4. Photomicrograph of the surgical specimen showing normal neural tissue separating malignant tumor, although possible. Noran the venous structures of the malformation. H & E- states that "venous angioma is not an expansile tumor; rather it replaces the tissue of the Luxol fast blue stain, • 40. 630
J. Neurosurg. / Volume 44 / May, 1976
Cerebral venous angioma involved area. ''~ Of course, the presence of a hematoma secondary to this lesion could create a mass effect, but intracranial hemorrhage secondary to this lesion is distinctly rare2 With newer advances in neuroradiology, including direct magnification and subtraction techniques, smaller venous angiomas may be picked up which have previously not been demonstrable by angiography. This lesion is not rare? Such small lesions would almost certainly not be demonstrated on the radionuclide cerebral angiogram and therefore the potential triad of increased activity on the radionuclide angiogram, normal static brain scan, and angiographic findings described above would not be useful. In a previously reported venous angioma demonstrated at angiography, S the findings were "an irregularly shaped area containing abnormal vessels of small size fairly evenly distributed in the parasagittal white matter. In the late arterial phase, filling of these small vessels had begun in the more superficial portion of the lesion. In the late arterial phase, there was filling of an enlarged vein in the white matter medial to the lesion and of a single connecting superficial vein draining to the superior sagittal sinus." Moreover, a second lesion which did not fill was demonstrated by its large mass effect in the temporal lobe. At autopsy, a large hematoma destroyed most of the temporal lobe, and several venous angiomas were found in the brain, including a large venous angioma corresponding to the lesion filled at angiography) The cataclysmic clinical presentation, multiplicity of lesions, and the presence of arteriovenous shunting separate this from our case. Courville states that most of these lesions are single and rarely bleed, 2 as in the present case. Moreover, the combination of angiographic findings described in our case should separate this lesion from an AVM or malignant tumor, whereas the presence of arteriovenous shunting could well confuse the diagnosis.
J. Neurosurg. / Volume 44 / May, 1976
The angiographic picture of a white matter lesion with the characteristics described in this case should be consistent with a venous angioma. The presence of a mass effect is dependent on hemorrhage of the lesion, which would be unusual as it apparently is rare for these lesions to bleed. Acknowledgments The authors would like to thank Drs. Guy L. Odom and Stephen F. Vogel for their help in the
preparation of this manuscript. References
1. Constans JP, Dilenge D, V~drenne C: An-
giomes veineux c6r6breux. Nenrochirnrgie 14: 641-650, 1968 2. Courville CB: Morphology of small vascular malformations of the brain. With particular reference to the mechanism of their drainage.
J Neuropathol Exp Neurol 22:274-284, 1963 3. Goree JA, Dukes HT: The angiographic differential diagnosis between the vascularized malignant glioma and the intracranial arteriovenous malformation. Am J Roentgenoi Radium Ther Nucl Med 90:512-521, 1963
4. Hamby WB: The pathology of supratentorial angiomas. J Neurosurg 15:65-75, 1958 5. McCormick WF: The pathology of vascular ("arteriovenous") malformations. J Neurosnrg 24:807-816, 1966 6. Noran HH: Intracranial vascular tumors and malformations. Arch Pathol 39:393-416, 1945 7. Scotti LN, Goldman RL, Rao GR, et al: Cerebral venous angioma. Neuroradioiogy 9:125-128, 1975 8. Wolf PA, Rosman NP, New PFJ: Multiple small cryptic venous angiomas of the brain mimicking cerebral metastases. A clinical pathological, and angiographic study. Neurology 17:491-501, 1967
Address reprint requests to." John A. Goree, M.D., Box 3808, Duke University Medical Center, Durham, North Carolina 27710.
631
RANDALL S. PREISSlG, M.D., SANDRA H. PREISSIG, M.D., JOHN A. GOREE, M.D.
AND
Departments of Radiology and Pathology, Duke University Medical Center, Durham, North Carolina ~" This patient presented with a cerebral venous angioma in which the angiographic findings were suggestive of a malignant brain tumor; the true nature of this lesion was documented at surgery. The diagnostic features of this case are discussed. KEY WORDS angiography
9 vascular malformation
9 cerebral venous angioma
9
NGIOGRAPHICdifferentiation of an arteriovenous malformation from malignant glioma will permit accurate diagnosis from the cerebral angiogram alone in most instances? We recently examined a patient with venous angioma of the brain with confusing angiographic features. A review of the literature reveals only a few instances of venous angioma demonstrated by angiography.
gram revealed some generalized paroxysmal activity suggestive of epilepsy. Radionuclide cerebral angiogram demonstrated a focus of abnormally increased activity deep within the right hemisphere during the arterial phase which cleared late in the venous phase (Fig. 1). A static scan immediately following angiography was normal. The combination of increased activity on the radionuclide cerebral angiogram with a normal static scan was interpreted as highly suggestive of aneurysm or arteriovenous malformation. Case Report A right carotid angiogram (Fig. 2) showed This 23-year-old woman had had head- a lesion in the white matter of the right fronaches for the past 5 years. Ten weeks before tal lobe immediately above the roof of the admission, the headaches increased in lateral ventricle. The abnormal area was first severity and began localizing in the right fron- identified as a diffuse blush, measuring 3.5 tal region; during that time she complained of cm in diameter. There were no hypertrophic decreased memory. arterial trunks or arteriovenous shunting Examination. Neurological examination present. The lesion drained into enlarged, showed only slightly decreased strength and parallel, medullary veins which converged at sensation on the left. An electroencephalo- the roof of the lateral ventricle; the entire le-
A
628
J. Neurosurg. / Volume 44 / May, 1976
Cerebral venous angioma
FI6. 1. Radionuclide cerebral angiogram demonstrating increased activity deep within the right hemisphere during the capillary and venous phases.
sion was in the shape of an inverted pyramid. A single large vein appeared to drain the confluence of the medullary veins and then emptied into the superior sagittal sinus. No drainage of the lesion by the deep venous system was noted and no mass effect was demonstrated. The angiographic circulation time of the lesion was normal. The angiographic diagnosis was malignant glioma based on the location of the lesion in the white matter, and an intrinsic vascular pattern characterized by clear spaces separating the individual abnormal vessels from each other. Operation. The patient underwent a right frontal craniotomy. A large red vein that drained into the superior sagittal sinus was noted to rise from within the frontal lobe between the middle and superior frontal gyri (Fig. 3). A 3 • 9-cm block resection of the involved area was performed. Numerous "thin walled vessels" and veins were noted during resection. Pathological examination revealed an angiomatous malformation consisting entirely of venous structures separated by normal neural tissue (Fig. 4). The patient has done well for 2 years on medication for control of her seizures.
Discussion
Angiography of arteriovenous malformations (AVM's) shows that they are composed of dilated arteries and veins with gross shunts between them; histologically they contain abnormal, hyalinized vessels which are "neither artery or vein, ''~ in addition to arteries and veins. In contrast, no abnormal arteries or arteriovenous shunts were seen at angiography in the present case. Microscopically, the lesion was composed of numerous thinwalled vessels separated by normal brain parenchyma that conform to McCormick's description of venous angiomas. ~ The current case represents a venous angioma and not an AVM, since it contains no arterial elements and the arteries in the brain substance surrounding the lesion are normal. This explains why no abnormal arteries were seen at angiography and why no arteriovenous shunts were demonstrated. In the absence of angiographic evidence for arteriovenous shunting, the red color noted at surgery of the thin-walled vein draining the angioma is difficult to explain. It may be that this red color reflects decreased percent oxy-
FIG. 2. Selected films of right carotid angiograms, lateral projection. After a normal early arterial phase, a diffuse "blush" in the frontal lobe is seen. Multiple parallel veins converge at the roof of the lateral ventricle during the early venous phase and a single large vein then drains this confluence into the superior sagittal sinus.
J. Neurosurg. / Volume 44 / May, 1976
629
R. S. Preissig, S. H. Preissig a n d
J.
A. G o r e e
this, we can find only a few instances in which a venous angioma was detected by angiography,la,8 Noran" mentioned that the venous angioma is characteristically pyramidal, the apex of the pyramid being at the ventricle. Courville stated that these lesions are rarely multiple and that "one sometimes finds a large vascular channel of uniform diameter passing from the subcortical lesion directly through the cerebral or cerebellar cortex to the superficial meningeal veins. Such a venous shunt must serve as a bypass to evacuate the venous blood from the cluster of anomalous veins to the exterior. ''~ This combination of descriptions is demonstrated in the current case, in FIG. 3. At surgery, the large vein demon- which the angiograms first showed a "blush" strated at angiography was noted to rise from within the frontal lobe (smaller arrow) and drain without evidence of arteriovenous fistula. Multiple parallel vessels were then seen draininto the superior sagittal sinus (larger arrow). ing this area toward the ventricular roof in an gen extraction due to the relatively increased inverted pyramid configuration; these veins blood volume within this lesion as demon- appeared to be dilated medullary veins. Then a single large vessel drained the lesion from strated by the blush at angiography. Courville2 described a series of 22 cerebral the confluence of these veins (the apex of the vascular malformations which fulfilled the pyramid at the ventricular roof) to the criteria of venous angiomas, dispelling the superior sagittal sinus (Fig. 2). It is interesting to note that the diagnosis of belief that these lesions are rare.' Despite "vascular malformation" was made at the time of the radionuclide cerebral angiogram and static brain scan (Fig. 1). The primary reason for that interpretation was that to our knowledge, no convincing case has been reported in which a malignant brain tumor has demonstrated increased flow on the radionuclide angiogram together with a normal static brain scan. At cerebral angiography, the demonstration of clear spaces separating abnormal vessels located primarily within the white matter is characteristic of brain tumor? However, the venous drainage of a tumor in this location classically should be predominantly by way of the subependymal veins of the lateral ventricle and to the deep venous system, rather than through a single vein to the superficial venous system. An enlarged draining vein as was seen in this lesion can be seen in an AVM or malignant tumor, but these lesions would not demonstrate enlarged medullary veins without concomitant arteriovenous shunting. The absence of a mass effect in a lesion 3.5 cm in diameter is atypical for a F~o. 4. Photomicrograph of the surgical specimen showing normal neural tissue separating malignant tumor, although possible. Noran the venous structures of the malformation. H & E- states that "venous angioma is not an expansile tumor; rather it replaces the tissue of the Luxol fast blue stain, • 40. 630
J. Neurosurg. / Volume 44 / May, 1976
Cerebral venous angioma involved area. ''~ Of course, the presence of a hematoma secondary to this lesion could create a mass effect, but intracranial hemorrhage secondary to this lesion is distinctly rare2 With newer advances in neuroradiology, including direct magnification and subtraction techniques, smaller venous angiomas may be picked up which have previously not been demonstrable by angiography. This lesion is not rare? Such small lesions would almost certainly not be demonstrated on the radionuclide cerebral angiogram and therefore the potential triad of increased activity on the radionuclide angiogram, normal static brain scan, and angiographic findings described above would not be useful. In a previously reported venous angioma demonstrated at angiography, S the findings were "an irregularly shaped area containing abnormal vessels of small size fairly evenly distributed in the parasagittal white matter. In the late arterial phase, filling of these small vessels had begun in the more superficial portion of the lesion. In the late arterial phase, there was filling of an enlarged vein in the white matter medial to the lesion and of a single connecting superficial vein draining to the superior sagittal sinus." Moreover, a second lesion which did not fill was demonstrated by its large mass effect in the temporal lobe. At autopsy, a large hematoma destroyed most of the temporal lobe, and several venous angiomas were found in the brain, including a large venous angioma corresponding to the lesion filled at angiography) The cataclysmic clinical presentation, multiplicity of lesions, and the presence of arteriovenous shunting separate this from our case. Courville states that most of these lesions are single and rarely bleed, 2 as in the present case. Moreover, the combination of angiographic findings described in our case should separate this lesion from an AVM or malignant tumor, whereas the presence of arteriovenous shunting could well confuse the diagnosis.
J. Neurosurg. / Volume 44 / May, 1976
The angiographic picture of a white matter lesion with the characteristics described in this case should be consistent with a venous angioma. The presence of a mass effect is dependent on hemorrhage of the lesion, which would be unusual as it apparently is rare for these lesions to bleed. Acknowledgments The authors would like to thank Drs. Guy L. Odom and Stephen F. Vogel for their help in the
preparation of this manuscript. References
1. Constans JP, Dilenge D, V~drenne C: An-
giomes veineux c6r6breux. Nenrochirnrgie 14: 641-650, 1968 2. Courville CB: Morphology of small vascular malformations of the brain. With particular reference to the mechanism of their drainage.
J Neuropathol Exp Neurol 22:274-284, 1963 3. Goree JA, Dukes HT: The angiographic differential diagnosis between the vascularized malignant glioma and the intracranial arteriovenous malformation. Am J Roentgenoi Radium Ther Nucl Med 90:512-521, 1963
4. Hamby WB: The pathology of supratentorial angiomas. J Neurosurg 15:65-75, 1958 5. McCormick WF: The pathology of vascular ("arteriovenous") malformations. J Neurosnrg 24:807-816, 1966 6. Noran HH: Intracranial vascular tumors and malformations. Arch Pathol 39:393-416, 1945 7. Scotti LN, Goldman RL, Rao GR, et al: Cerebral venous angioma. Neuroradioiogy 9:125-128, 1975 8. Wolf PA, Rosman NP, New PFJ: Multiple small cryptic venous angiomas of the brain mimicking cerebral metastases. A clinical pathological, and angiographic study. Neurology 17:491-501, 1967
Address reprint requests to." John A. Goree, M.D., Box 3808, Duke University Medical Center, Durham, North Carolina 27710.
631
