Variations in the Circle of Willis in Macaccr mulcrtta ' NEAL F. KASSELL AND THOMAS W. LANGFITT Department of Neurosurgery, Pennsylvania Hospital ABSTRACT The circle of Willis was dissected in 75 Macaca mulatta monkeys, and the anatomy is described and compared with several series of human specimens. The principal difference between monkey and man is the presence of a single distal anterior cerebral artery in the former, and the most common anomaly in the monkey is an anterior communicating artery proximal to the junction of the anterior cerebral arteries. A large vessel joining the internal carotid arteries and a n accessory branch of the anterior cerebral artery, probably communicating with the external carotid circulation, were found in a small number of our specimens. In man the posterior communicating is significantly reduced i n diameter far more frequently than the anterior communicating artery, whereas the reverse situation obtains in the monkey. In general, the intracranial distribution of blood i n the monkey is the same as in man, because the similarities in the normal anatomy and variations i n the circle of Willis outweigh their differences. However, physiological data on collateral cerebral circulation will be more meaningful when it is based on the arterial anatomy in each preparation. There have been occasional descriptions of the arterial supply of the brain in the ape and monkey, but we have been unable to find in the literature a detailed account of the circle of Willis and its variations in subhuman primates. Grunbaum and Sherrington dissected the circle of Willis and its major branches in 11 chimpanzees and one orangu-tan. Watts examined numerous primate species, including 53 Old World monkeys, and described the normal arterial anatomy at the base of the brain. He noted the characteristic single distal anterior cerebral artery but found in one Macaca mulatta a n anterior communicating and two distal anterior cerebral arteries, as i n man. In recent years the monkey has been used extensively in research in cerebrovascular physiology, primarily because the extracranial cerebral circulation in carnivora differs so much from man. Since the interpretation of results is in part dependent upon the existence of collateral channels of circulation similar to man, the present study was undertaken to describe the anatomy of the circle of Willis in a large series of monkeys and compare the anomalies with those in the human. MATERIALS AND METHODS The material consisted of the brains from 75 Macaca mulatta monkeys used ANAT. REC.,152: 257-264. primarily in a n investigation of the pathophysiology of acute and chronic increased intracranial pressure. Following sacrifice of each animal the brain was perfused with normal saline and formalin through the carotid arteries, followed by liquid latex in 30 animals. The brain was removed, variations in the arterial anatomy at the base were recorded, and the ventral surface of the brain was photographed. The circle of Willis was then removed, marked appropriately, and preserved i n formalin. RESULTS I. Normal circle of Willis The normal human circle of Willis, as defined by Padget is the familiar polygon in which the anterior communicating artery is one-half to two-thirds the diameter of the anterior cerebral artery, which is half the diameter of the internal carotid artery. Also, the posterior communicating artery is one-half the size of the posterior cerebral artery which is one-half the size of the basilar artery. Using these criteria the normal circle of Willis in the rhesus monkey is essentially the same with the following exceptions : the anterior communicating artery is absent, and the proximal anterior cerebral arteries fuse to form 1 Supported by a grant from the John A. Hartford Foundation, Inc. 257 258 NEAL F. KASSELL AND THOMAS W. LANGFITT Fig. 1 Normal circle of Willis. Single distal anterior cerebral artery and relatively small size of posterior cerebral artery distinguished this from normal human specimen. An anterior communicating artery was present in 46% of specimens, and one of these has been selected for illustration. a single vessel; the posterior communicating artery is one-third the diameter of the posterior cerebral artery. Twenty-five per cent of the circles of Willis in this study were considered to be within normal limits (fig. 1). 11. Abnormal circles of Willis A. Absent vessels. All circles of Willis examined in the monkey were complete. B. Attenuated vessels. In the monkey we have considered a component of the circle of Willis to be significantly reduced i n size when it is 50% or less the diameter of its opposite number, and this occurred i n 44% of the specimens (figs. 2 and 3 ) . C . Accessory vessels. Accessory vessels were present i n the form of duplication or triplication of one of the component vessels in 60% of monkeys. Two accessory MONKEY CIRCLE O F WILLIS 259 Fig. 2 Attenuated right posterior communicating artery. Anterior communicating artery also present in this specimen and figure 4. vessels were found in a small number of specimens which are interesting in terms of potential collateral circulation, although not a part of the circle of Willis. A rather large artery joining the internal carotids above the cavernous sinus was present in 9% (fig. 4 ) , and arteries arising from the medial portion of the anterior cerebral artery and entering foramina posterolatera1 to the optic foramen were identified in 5% of animals. An attempt was made to follow the course of these latter vessels in the latex injected preparations. Although the distribution was extracranial, we could not be certain if they perforated the base of the skull in the region of the anterior portion of the sphenoid sinus or entered the orbit. D. Anomalous origin of vessels. None of the specimens had an embryonic origin of the posterior cerebral from the internal carotid artery. 260 NEAL F. KASSELL AND THOMAS W. LANGFITT Y Fig. 3 Attenuated proximal right anterior cerebral artery. E. Fusion of the anterior cerebral arteries. In 96% of the monkeys there was fusion of the proximal anterior cerebral arteries to form a single distal anterior cerebral supplying both hemispheres. F. Multiple variations. Combinations of the above variations of the circles of Willis occurred in 24% of the specimens. In table 1 our findings are summarized and compared to a large human series. DISCUSSION The incidence of normal circles of Willis depends on the criteria employed, and in human studies it has vaned from 11% (Fawcett and Blanchford) to 73% (Hasbe). In the study by Alpers et al. 52% of the circles of Willis were normal by Padget’s definition. Hodes et al. found the circle complete in more than 1,600 consecutive autopsies in contrast to Windle who de- MONKEY CIRCLE O F WILLIS 26 1 Fig. 4 Large vessel joining internal carotid arteries at the level of the hypophyseal stalk. Black marker has been inserted between the vessel and brain for contrast. scribed absent vessels in 27 of 200 speci- present. Of the attenuated vessels 7% mens. Alpers et al. found absence of the were posterior communicating and 24% Posterior communicating artery in 0.6% anterior communicating arteries. Among and the anterior communicating artery in the human 4 to 46% contained at2% Of the specimens they examined' In tenuated vessels. Alpers found attenuation the present series the circle of Willis was of a posterior communicating artery in complete in all specimens. Attenuated vessels were observed in 22% and an anterior cerebral artery in ~ ~ a m i n were e d Patent, 44% of the monkeys, but irrespective of 396, but all v&%ds the size of the artery a lumen was always as in the monkey. 2 9 3 0.6 0.6 5 E. Fusion of anterior cerebral 2 5 5 1 0.3 0 0 0 0 29 10 2 2 16 16 F. Multiple variations 2.6 0.6 33 9.6 28 8 3 1 2 0.6 7 2 5 1. Posterior cerebral D. Anomalous origin C. Accessory vessels 1. Anterior communicating 2. Anterior cerebral a. Proximal duplication b. Distal duplication c. Distal triplication 3. Posterior communicating 4. Primitive olfactory artery 5. Artery joining internal carotids 1. Posterior communicating 2. Anterior communicating 3. Anterior cerebral B. Attenuated vessels I. Normal 11. Abnormal A. Absent vessels 1. Posterior communicating 2. Anterior communicating 4.2 0 0 13 4 13 4 0 0 14 1 0.3 1 0.3 0 0 0 0 6 6 0 0 21 21 0 0 ______ No. % No. 7'0 Bilateral 47 1 22 22 14 1 0 0 37 20 4 4 42 29 7 6 1 0 1 No' 13 0.3 7 7 4.2 0.3 0 0 11 6 1 1 13 9 2 2 0.3 0 0.3 o9' Combined variations Human Alpers et al. (350) 96 47 6 51 51 67 30 7 7 345 28 2 0 0 96 78 10 8 13 2 15 15 19 9 2 2 98 8 1 0 0 27 22 3 3 183 52 167 48 8 3 2 0.6 6 2 No' Total 0 0 % 19 0 0 0 1 5 28 20 2 2 25 0 0 0 1 7 37 27 3 3 21 28 2 3 13 17 6 8 0 0 No. 0 0 0 0 1 0 1 0 0 1 1 0 1 0 0 1 0 0 3 2 1 1 2 3 1 1 0 0 No. 3' % Unilateral Bilateral 0 71 0 0 18 24 53 0 0 18 24 96 0 0 72 0 0 53 69 35 46 6 8 3 4 3 4 0 0 1 1 4 5 7 9 53 23 31 15 20 3 4 1 1 2 3 0 0 1 1 2 3 2 3 40 0 19 25 56 75 0 0 No' 33 44 5 7 18 24 1013 0 0 % Total - 13 3 7 4 10 2 5 3 0 0 No' variations Combined Monkey Kassell and Langfitt (75) Single variation anomalies of t h e circle of Willis in m a n and m o n k e y Single variation of Unilateral Comparison TABLE 1 H r r B v) & w w E3 MONKEY CIRCLE OF WILLIS Accessory vessels occurred in 69% of the monkeys. In 46% an anterior communicating artery was present immediately proximal to the junction of the anterior cerebral arteries. Because of this high incidence it might be equally valid to consider an anterior communicating artery as part of the normal circle of Willis in the monkey. In Watts analysis of 56 Macaca mulatta monkeys an anterior communicating artery was not mentioned. Alpers et al. found duplication or triplication of component vessels in 19% of the human specimens. In 4% of our specimens there was duplication of one proximal anterior cerebral artery and in 4% duplication of the distal anterior cerebral was present, analogous to the normal human anatomy. The artery joining the internal carotids at the level of the hypophysis, observed in 9% of our specimens, has also been described by Padget in human embryological studies, and recently Hayes demonstrated communication between the internal carotids through two inferior hypophyseal and one meningeal artery in adult man. The vessels which take origin from the medial portion of the proximal anterior cerebral arteries and enter the base of the skull beneath the optic foramina may be analogous to primitive olfactory arteries described by Padget and are indicated as such in table 1. Embryonic origin of the posterior cerebral artery from the posterior communicating artery occurred in 15% of the human circles of Willis examined by Alpers et al., but this variation was not observed in the monkey. Fusion of the anterior cerebral arteries was present in 2% of the human circles of Willis examined by Alpers et al. while this occurred in 96% of the specimens examined in the present study. The normal anatomy of the circle of Willis in the monkey is quite similar to man, and though numerous anomalies occur in both species, it is unlikely that they create significant differences in cerebral 263 hemodynamics. The major difference is a single anterior cerebral artery in the monkey, but the anterior communicating artery in man permits irrigation of both frontal lobes from either side of the circle of Willis, SO physiologically the circulation is essentially the same. The incidence of attenuated posterior communicating arteries in man is three times that in the monkey which may be of some significance in terms of the contribution of the vertebral-basilar circulation to the blood supply of the cerebral hemisphere. The accessory branch of the anterior cerebra1 artery, probably communicating with the external carotid circulation, provides a collateral channel for the anterior circle of Willis which is apparently not present in man. However, this was identified in only a few specimens. A large and important accessory vessel between the internal carotid arteries was found in one of ten monkeys, but as noted, similar vessels have now been described in human material. LITERATURE CITED Alpers, J. B., R. G. Berry and R. M. Paddison 1959 Anatomical studies of the circle of Willis in normal brain. Arch. Neurol. Psychiat., 81: 409-418. Fawcett, E., and J. V. Blanchford 1905-1906 The circle of Willis: An examination of 700 specimens. J. Anat. Physiol., 40: 63-70. Grunbaum, A. S. F., and C. S. Sherrington 1902 Notes on the arterial supply of the brain in anthropoid apes. Brain, 25: 270-273. Hasbe, K. 1928 In das arteiensystem der Japaner, Kyoto, B. Adachii. Hayes, G. J. 1963 External carotid-cavernous sinus fistulas. J. Neurosurg., 20: 692-700. Hodes, P. J., F. Campoy, H. E. Riggs and P. Bly 1953 Cerebral angiography: Fundamentals in anatomy and physiology. Amer. 3. Roentgend., 70: 61-81. Padget, D. H. 1945 The circle of Willis: Its embryology and anatomy. In: Intracranial Arterial Aneurysms. Ed. W. E. Dandy, New York, Comstock Publishing Co., Chapt. 111. Watts, J. W. 1933-1934 A comparative study of the anterior cerebral artery and the circle of Willis in primates. Am. J. Anatomy, 68: 534-550. Windle, B. C. A. 1888 On the arteries forming the circle of Willis. J. Anat. Physiol., 22: 289293.