The circle of Willis (circulus arteriosus cerebri) is a polygonal arterial anastomosis in the interpeduncular and chiasmatic cisterns at the base of the brain. It joins the two internal carotid arteries (ICAs) with the vertebrobasilar system. Anteriorly it is formed by the precommunicating (A1) segments of the anterior cerebral arteries (ACAs) and the anterior communicating artery (ACoA). Laterally it is formed by the terminal ICAs and the posterior communicating arteries (PCoAs), and posteriorly by the precommunicating (P1) segments of the posterior cerebral arteries (PCAs), which arise from the basilar bifurcation [1,2]. Thomas Willis described the ring in 1664 and recognised its role as a protective collateral channel. The function of the circle is debated. It is widely regarded as a safety valve that redistributes blood when one feeding artery is narrowed or occluded. It may also equalise pressure between the carotid and vertebrobasilar territories and damp pulsatile flow [3]. In the healthy state the communicating arteries carry little flow. Their importance emerges when a major inflow vessel fails, at which point their presence and calibre determine whether the brain is adequately perfused. The textbook configuration, a complete and symmetrical ring with all segments of adequate calibre, is the exception rather than the rule. Alpers and colleagues found the classic pattern in about half of 350 normal brains [4]. Riggs and Rupp, examining 994 brains, showed that variant circles were frequent and had implications for collateral circulation [5]. Kapoor and colleagues reported a typical pattern in 45.2% of 1000 Indian autopsy brains, with variations in the remaining 54.8% [6]. Eftekhar and colleagues compared their Iranian series with the world literature and found no clear evidence that the distribution of variants differs between populations [7]. Imaging studies agree: on three-dimensional time-of-flight MR angiography, Krabbe-Hartkamp and colleagues found an entirely complete circle in only 42% of 150 adults [8]. These variations have an embryological basis. The circle forms from the primitive carotid and longitudinal neural arteries between the fourth and eighth weeks of development. In early fetal life the PCA is supplied mainly through the PCoA, and the P1 segment enlarges only later as the occipital lobe grows [9,10]. Persistence of the fetal pattern, unequal regression of vessels or incomplete fusion of the plexiform anterior network produce hypoplasia, aplasia, duplication and fenestration. The clinical stakes are considerable. An incomplete or hypoplastic circle limits collateral supply in carotid occlusion and influences infarct size and stroke outcome [11,12]. Asymmetry and variation of the circle are associated with saccular aneurysm formation, particularly at the ACoA complex, because of altered haemodynamic stress [13,14]. Knowledge of the local pattern guides aneurysm clipping, endovascular coiling, carotid endarterectomy and cardiac surgery with selective cerebral perfusion [2]. Regional cadaveric data remain valuable because they allow direct, high-resolution observation of small vessels that imaging may miss. Reports from South Asia show a wide range in the frequency of the typical configuration [6,15–17]. The present study was undertaken to document the morphological variations of the circle of Willis in adult cadaveric brains from [region], to compare them with earlier series, and to discuss their clinical significance.

Study design and setting. This was a descriptive, cross-sectional cadaveric study conducted in the Department of Anatomy, [Institution], between [month year] and [month year]. Approval was obtained from the Institutional Ethics Committee (Ref. No. [ ]). The bodies had been received through voluntary body donation and were used in accordance with the [State] Anatomy Act. Sample. Sixty adult human brains were studied: 42 from male and 18 from female cadavers, aged 25 to 80 years (mean 54.6 ± 13.2 years). Age and sex were taken from the donation records. Brains showing gross pathology at the base, such as tumour, haemorrhage, infarction, or trauma, and brains damaged during removal, were excluded. Brains with intracranial aneurysms found during dissection were retained and recorded. Specimen preparation. The cadavers had been embalmed by arterial perfusion with 10% formalin. The skull cap was removed with an oscillating saw, the dura was opened, and the cranial nerves, ICAs and vertebral arteries were divided close to the skull base so that the vessels stayed on the brain. Each brain was removed intact and further fixed by immersion in 10% formalin for at least two weeks. Dissection. The brain was placed base upward. Under a magnifying lens (×4) and an operating loupe, the arachnoid mater over the interpeduncular fossa, optic chiasma and lateral sulci was removed with fine forceps and microscissors. The following vessels were cleaned and traced on both sides: the terminal ICA, A1 segment of the ACA, ACoA, PCoA, P1 segment of the PCA, and the basilar artery with its bifurcation. Care was taken not to stretch or tear small communicating vessels. Parameters recorded. For every component the following were noted: (a) presence or absence (aplasia); (b) calibre; (c) duplication, triplication or fenestration; (d) abnormal origin or course; and (e) additional vessels such as a persistent median artery of the corpus callosum. Configuration of the PCoA–P1 junction was recorded on each side. Measurements. The external diameter of each vessel was measured at its midpoint with a digital vernier caliper (accuracy 0.01 mm). Each value was taken three times by the same observer and the mean was used. Vessels too fine for the caliper were measured against a calibrated millimetre grid under magnification. Definitions. A vessel with an external diameter of less than 1 mm was defined as hypoplastic, following earlier anatomical studies [4,6]. Aplasia was the complete absence of a segment. A fetal-type PCA was recorded when the PCoA diameter exceeded that of the ipsilateral P1 segment, so that the PCA drew its supply mainly from the ICA [18]. A typical circle was one in which all ten segments were present, of normal calibre (≥ 1 mm), single and symmetrical. A complete circle was one in which every segment was present, irrespective of calibre. An incomplete circle had at least one aplastic segment. The circle was divided into an anterior part (A1 segments and ACoA) and a posterior part (PCoAs and P1 segments) for analysis. Photography. Each specimen was photographed with a digital camera and scale bar for record and for independent verification of the findings by a second anatomist. Disagreements were resolved by joint re-examination. Statistical analysis. Data were entered in Microsoft Excel and analysed with SPSS version [ ] (IBM Corp., Armonk, NY). Categorical variables were expressed as frequency and percentage. Differences between right and left sides and between males and females were assessed with the chi-square test, or Fisher's exact test when expected cell counts were below five. A p value below 0.05 was considered significant.

Sixty brains were examined. Only about one in three showed the textbook circle; two in three carried at least one variation, most often in the posterior part. Table 1. Age and sex distribution of the specimens (n = 60) Age group (years) Male, n Female, n Total, n (%) 25–40 8 4 12 (20.0) 41–60 19 8 27 (45.0) > 60 15 6 21 (35.0) Total 42 (70.0%) 18 (30.0%) 60 (100) Males made up 70% of the sample, reflecting the donor pool. The 41–60 year group was the largest (45.0%). Table 2. Overall configuration of the circle of Willis by sex Configuration Male (n = 42), n (%) Female (n = 18), n (%) Total (n = 60), n (%) Typical (complete, normal calibre, symmetrical) 15 (35.7) 6 (33.3) 21 (35.0) Complete but variant (hypoplasia, duplication, fenestration or fetal PCA) 19 (45.2) 9 (50.0) 28 (46.7) Incomplete (aplasia of ≥ 1 segment) 8 (19.0) 3 (16.7) 11 (18.3) Total 42 18 60 χ² = 0.12, df = 2, p = 0.94. A typical circle was found in 21 brains (35.0%). Of the 39 variant brains, 28 had every segment present but at least one segment was abnormal in calibre or form; 11 lacked one or more segments. Taken together, the circle was anatomically complete in 49 brains (81.7%). Fourteen brains (23.3%) had two or more coexisting variations. Configuration did not differ significantly between males and females. Table 3. Variations of the individual components of the circle (n = 60 brains) Part Component Variation n (%) Anterior A1 segment of ACA Hypoplasia 5 (8.3) Anterior A1 segment of ACA Fenestration 1 (1.7) Anterior ACoA Hypoplasia 4 (6.7) Anterior ACoA Duplication 4 (6.7) Anterior ACoA Fenestration 2 (3.3) Anterior ACoA Absence 1 (1.7) Anterior ACoA complex Persistent median artery of corpus callosum 2 (3.3) Posterior PCoA Hypoplasia (unilateral 9, bilateral 6) 15 (25.0) Posterior PCoA Absence (unilateral 6, bilateral 2) 8 (13.3) Posterior PCA Fetal-type PCA (unilateral 7, bilateral 2) 9 (15.0) Posterior P1 segment of PCA Hypoplasia 3 (5.0) Posterior P1 segment of PCA Absence (with fetal PCA) 2 (3.3) Brains may carry more than one variation, so percentages do not sum to 100. At least one anterior variation was present in 15 brains (25.0%) and at least one posterior variation in 31 brains (51.7%). The PCoA was the most variable vessel. It was hypoplastic in a quarter of brains and absent in 13.3%; two brains lacked both PCoAs, isolating the carotid from the vertebrobasilar circulation. A fetal-type PCA was seen in nine brains, two of which had an absent P1, so that the occipital lobe on that side depended entirely on the ICA. In the anterior part, A1 hypoplasia and ACoA hypoplasia or duplication were the commonest findings. The single absent ACoA left the two anterior cerebral territories without a direct cross-connection. Table 4. Side distribution of the paired variations Variation Right only, n Left only, n Bilateral, n Total brains, n Sides affected (R / L) p PCoA hypoplasia 5 4 6 15 11 / 10 0.82 PCoA absence 4 2 2 8 6 / 4 0.51 Fetal-type PCA 4 3 2 9 6 / 5 0.75 A1 hypoplasia 3 2 0 5 3 / 2 1.00 P1 hypoplasia 2 1 0 3 2 / 1 1.00 P1 absence 1 1 0 2 1 / 1 1.00 p values compare the number of right and left sides affected (chi-square or Fisher's exact test). Variations were slightly more frequent on the right for the PCoA and fetal PCA, but no side difference reached significance. Bilateral involvement was common for PCoA hypoplasia (6 of 15 brains), supporting a developmental rather than random origin. Table 5. Comparison of the present findings with selected earlier studies Study (country) Method n Typical / complete circle Key posterior finding Kapoor et al., 2008 (India) [6] Autopsy dissection 1000 45.2% typical PCoA hypoplastic 13.2%, absent 1.0% De Silva et al., 2011 (Sri Lanka) [15] Cadaveric, calibrated grid 225 14% typical Bilateral PCoA hypoplasia 23% Krabbe-Hartkamp et al., 1998 (Netherlands) [8] 3D TOF MR angiography 150 42% complete Complete posterior part in 52% Eftekhar et al., 2006 (Iran) [7] Autopsy dissection 102 — PCoA hypoplasia commonest; both PCoAs absent in 3% Gunnal et al., 2014 (India) [16] Cadaveric dissection 150 — PCoA variations 50%, ACoA 40% Present study Cadaveric dissection 60 35.0% typical PCoA hypoplastic 25.0%, absent 13.3%

The present study confirms that the textbook circle of Willis is a minority pattern. A typical configuration was seen in 35.0% of brains, within the range reported by earlier cadaveric and imaging series (Table 5). Kapoor and colleagues found a typical pattern in 45.2% of 1000 Indian brains [6], and Krabbe-Hartkamp and colleagues an entirely complete circle in 42% of adults on MR angiography [8]. De Silva and colleagues, applying strict calibre criteria to 225 Sri Lankan brains, found a typical circle in only 14% [15]. Much of this spread reflects definitions. Studies that count any hypoplastic segment as a variation report lower typical rates than those that require only the presence of each vessel [7,19]. The posterior part of the circle was the most variable, as in almost every published series [6,7,16,19]. PCoA hypoplasia was the single commonest variant (25.0%), higher than the 13.2% of Kapoor and colleagues [6] but consistent with the predominance of PCoA variants noted by Eftekhar and colleagues [7] and in a recent meta-analysis [19]. Our higher rate of PCoA absence (13.3%) may reflect the smaller sample and the use of magnification, which separates true aplasia from a thread-like vessel less reliably than histology. The embryology explains this posterior instability. The PCoA is the original supply to the PCA; as the P1 segment enlarges, the PCoA regresses to a variable degree, leaving anything from a large fetal channel to a thread or nothing at all [9,10]. A fetal-type PCA was present in 15.0% of brains. In such hemispheres the occipital lobe depends on the ICA, so carotid stenosis or embolism can produce PCA-territory infarction, and vertebrobasilar collateral support is reduced [18]. In the two brains with absent P1, the carotid and posterior circulations were functionally separate on that side. This pattern carries particular risk during carotid surgery and endovascular procedures. The ACoA complex showed hypoplasia, duplication, fenestration and, rarely, absence, as described by Kapoor and colleagues, who reported ACoA duplication in 10% and absence in 1.8% [6]. Multiple and plexiform ACoA patterns are common in large autopsy series [20]. These details matter because the ACoA region is the commonest site of intracranial aneurysm. Kayembe and colleagues found aneurysms more often in brains with circle variations [13], and Lazzaro and colleagues linked A1 asymmetry and other anomalies to aneurysm rupture [14]. A1 hypoplasia (8.3% here) increases flow through the ACoA from the dominant side, raising wall shear stress at the junction. Clinically, an incomplete circle reduces the brain's tolerance of carotid clamping and unilateral selective cerebral perfusion during aortic arch surgery. Papantchev and colleagues, studying 500 circles, argued that such variations should be considered when planning perfusion strategy [21]. The communicating arteries are the primary collaterals in acute occlusion, and their absence predicts poorer collateral flow [11,12]. The study has limitations. The sample is modest and male-predominant, the cause of death and vascular history were often unknown, and fixation may alter vessel calibre. Diameter measurement with a caliper is less precise for the smallest vessels than morphometric software.

Only 35.0% of the brains studied showed the classic circle of Willis; 65.0% had at least one variation, and 18.3% had an incomplete ring. The posterior communicating artery was the most variable vessel, and fetal-type PCA was common. No significant sex or side differences were found. Because these variations alter collateral capacity, aneurysm risk and surgical planning, the circle should be mapped by CT, MR or catheter angiography before neurovascular, endovascular, carotid and aortic arch procedures. Larger multicentre cadaveric and imaging studies from the region would refine these estimates.

1. Standring S, editor. Gray's Anatomy: The Anatomical Basis of Clinical Practice. 42nd ed. London: Elsevier; 2020. 2. Rhoton AL Jr. The supratentorial arteries. Neurosurgery. 2002;51(4 Suppl):S53-120. 3. Vrselja Z, Brkic H, Mrdenovic S, Radic R, Curic G. Function of circle of Willis. J Cereb Blood Flow Metab. 2014;34(4):578-84. 4. Alpers BJ, Berry RG, Paddison RM. Anatomical studies of the circle of Willis in normal brain. AMA Arch Neurol Psychiatry. 1959;81(4):409-18. 5. Riggs HE, Rupp C. Variation in form of circle of Willis: the relation of the variations to collateral circulation: anatomic analysis. Arch Neurol. 1963;8:8-14. 6. Kapoor K, Singh B, Dewan LI. Variations in the configuration of the circle of Willis. Anat Sci Int. 2008;83(2):96-106. 7. Eftekhar B, Dadmehr M, Ansari S, Ghodsi M, Nazparvar B, Ketabchi E. Are the distributions of variations of circle of Willis different in different populations? Results of an anatomical study and review of literature. BMC Neurol. 2006;6:22. 8. Krabbe-Hartkamp MJ, van der Grond J, de Leeuw FE, de Groot JC, Algra A, Hillen B, et al. Circle of Willis: morphologic variation on three-dimensional time-of-flight MR angiograms. Radiology. 1998;207(1):103-11. 9. Padget DH. The development of the cranial arteries in the human embryo. Contrib Embryol. 1948;32:205-61. 10. Menshawi K, Mohr JP, Gutierrez J. A functional perspective on the embryology and anatomy of the cerebral blood supply. J Stroke. 2015;17(2):144-58. 11. Liebeskind DS. Collateral circulation. Stroke. 2003;34(9):2279-84. 12. Hartkamp MJ, van der Grond J, van Everdingen KJ, Hillen B, Mali WP. Circle of Willis collateral flow investigated by magnetic resonance angiography. Stroke. 1999;30(12):2671-8. 13. Kayembe KN, Sasahara M, Hazama F. Cerebral aneurysms and variations in the circle of Willis. Stroke. 1984;15(5):846-50. 14. Lazzaro MA, Ouyang B, Chen M. The role of circle of Willis anomalies in cerebral aneurysm rupture. J Neurointerv Surg. 2012;4(1):22-6. 15. De Silva KR, Silva R, Amaratunga D, Gunasekera WS, Jayesekera RW. Types of the cerebral arterial circle (circle of Willis) in a Sri Lankan population. BMC Neurol. 2011;11:5. 16. Gunnal SA, Farooqui MS, Wabale RN. Anatomical variations of the circulus arteriosus in cadaveric human brains. Neurol Res Int. 2014;2014:687281. 17. Iqbal S. A comprehensive study of the anatomical variations of the circle of Willis in adult human brains. J Clin Diagn Res. 2013;7(11):2423-7. 18. van Raamt AF, Mali WP, van Laar PJ, van der Graaf Y. The fetal variant of the circle of Willis and its influence on the cerebral collateral circulation. Cerebrovasc Dis. 2006;22(4):217-24. 19. Jones JD, Castanho P, Bazira P, Sanders K. Anatomical variations of the circle of Willis and their prevalence, with a focus on the posterior communicating artery: a literature review and meta-analysis. Clin Anat. 2021;34(7):978-90. 20. Klimek-Piotrowska W, Rybicka M, Wojnarska A, Wójtowicz A, Koziej M, Hołda MK. A multitude of variations in the configuration of the circle of Willis: an autopsy study. Anat Sci Int. 2016;91(4):325-33. 21. Papantchev V, Stoinova V, Aleksandrov A, Todorova-Papantcheva D, Hristov S, Petkov D, et al. The role of Willis circle variations during unilateral selective cerebral perfusion: a study of 500 circles. Eur J Cardiothorac Surg. 2013;44(4):743-53.