Vascular Anatomy in the Lumbar Spine Investigated by Three-Dimensional ComputedTomography Angiography: The Concept of Vascular Window
Cédric Barrey1,4, Bogdan Ene1, Guy Louis-Tisserand2, Pietro Montagna3, Gilles Perrin1, Emile Simon1
taatdtawf
dwnvit3tlp
sisttaawa
�Tacdt
f
�sarLbpv
INTRODUCTION
The anterior approach to treat lumbar spi-nal disorders was first described by Burns(2) in 1933 using a transperitoneal ap-proach for a case of L5-S1 spondylolisthe-sis. Lane and Moore (20) reported the firstseries of anterior lumbar interbody fusion(ALIF) procedures performed for degenera-tive changes in 1948. Southwick and Robin-son (28) were the first to describe the retro-peritoneal approach to the lumbar spine in1957. More recently, to meet the criteria ofmini-invasive surgery, Onimus et al. (24)reported the mini-open extraperitoneal ap-proach with video assistance.
With more recent progress in anterior in-strumentation and lumbar total disk re-placements, anterior exposure has becomemore popular for the management of lum-bar degenerative diseases (4, 8, 10, 12, 16,18, 27, 30). Although the anterior approach
From the Departments of 1Neurosurgery andSpine Surgery and 2Neuroradiology, Claude
Bernard University of Lyon 1, and Pierre Wertheimer Hospital,Hospices Civils de Lyon, Lyon; 3Department of Cardio-vascular
urgery, Claude Bernard University of Lyon 1, and Louis Pradelospital, Hospices Civils de Lyon, Lyon; and 4Laboratory ofiomechanics, Arts et Metiers PARISTECH, Paris, France
o whom correspondence should be addressed:édric Barrey, M.D. [E-mail: [email protected]]
itation: World Neurosurg. (2013) 79, 5/6:784-791.ttp://dx.doi.org/10.1016/j.wneu.2012.03.019
erior approach, such as less muscle dam-ge, absence of nerve root manipulation,nd better biomechanical conditions for in-erbody bone graft, it is a demanding proce-ure involving different anatomic struc-
ures. Many spine surgeons, orthopedists,nd neurosurgeons work in collaborationith general or vascular surgeons to per-
orm these procedures (1, 8, 16).Numerouscomplicationsassociatedwiththe
� OBJECTIVE: To report an in vivo anin the lumbar spine using three-dimangiography and to develop the conceto L4-5 and L5-S1 disks.
� METHODS: In 146 patients who wurgery, 3D CT angiography was perncluded degenerative disk diseaseis (n � 26). 3D reconstructions weechnique. Level of aortic bifurcationhe ascending iliolumbar vein, presnatomic variations were analyzed. As the “free vascular” area for the anindow at L4-L5 was defined as
nterolateral part of the L4-5 disk.
RESULTS: The level of aortic bifurcahe iliocavum confluence occurred mscending vein and central sacral veases. Five (3.5%) anatomic variatioraining into the left common iliac v
hree cases. A vascular window was23 mm � 8 at L4-L5. The vascular windour patients at L5-S1 and in approxim
CONCLUSIONS: This study confirmpine is characterized by a great variabssessment. 3D CT angiography alloelationships between the prevertebra4-L5 and L5-S1. Although adherencee predicted by this technique, the creoperatively by CT angiography coessel mobilization during anterior lum
nteriorapproachinthelumbarspinehavebeen t
WORLD NEUROSURGERY, http://d
escribed in the literature, including abdominalall complications, hematoma, visceral anderve injuries, retroperitoneal lymphocele, andascularcomplications(1,7,13,14,26).Vascularnjuries are rare but are considered the most po-entially dramatic complications (1, 8, 11, 19, 21,1). Most vascular injuries consist of vein lacera-ion during disk exposure and may result inife-threatening bleeding. Other reported com-lications include deep venous thrombosis, ar-
ic evaluation of prevertebral vesselsnal (3D) computed tomography (CT)vascular window for surgical access
scheduled for anterior lumbar spineed preoperatively. Spinal disorders120) and low-grade spondylolisthe-
btained using the volume-renderingiliocavum confluence, presence ofof the central sacral vessels, and
scular window at L5-S1 was definedor part of the L5-S1 disk. A vascular“free vascular” area for the left
was most often observed at L4 (64%).requently at L5 (44%). The iliolumbars were identified in 84% and 72% of
ere noted: right internal iliac veinn two cases and tortuous vessels inured to 34.5 mm � 12 at L5-S1 and to
was <25 mm in approximately one iny two in three patients at L4-L5.
at vascular anatomy in the lumbarthat has significance for preoperativefor an effective evaluation of the
ssels and the intervertebral disks atssels to the anterior ligament cannotpt of vascular windows investigatede helpful in predicting the need forspine surgery.
atomensiopt of
ereform
(n �re o
andence
vaterithe
tionost fsselns wein imeasowatel
s thilitywedl ve
of veonceuld b
bar
erial thrombosis, and rarely arterial injury (19).
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
Preoperative assessment of morphologyand location of prevertebral vascular struc-tures could be helpful to reduce periopera-tive vascular complications of anterior sur-gical procedures in the lumbar spine. Thispreoperative anatomic evaluation has beendone using computed tomography (CT) (5,15) or magnetic resonance angiography (3,22). These studies concluded that preoper-ative evaluation of the anatomic configura-tion of abdominal large vessels influencedsurgical strategy in a consistent percentageof patients, helped to reduce unnecessaryvascular retraction, and could reduce theassociated vascular complications.
To our knowledge, the study by Datta etal. (5) is the only one that investigated vas-cular anatomy of abdominal vessels with CTangiography in a population of patientswith lumbar spinal disorders. However, theanalysis was mainly descriptive, and precisemeasurements of vascular windows werenot performed. The purpose of our study isto report in vivo three-dimensional (3D)quantitative anatomic evaluation of prever-tebral lumbar great vessels using CT an-giography and to develop the concept ofvascular window for anterior access toL5-S1 and L4-5 disks.
METHODS
From January 2008 to January 2011, 146 pa-tients who were scheduled to undergo ananterior surgical procedure in the L4-S1lumbar spine were consecutively includedin the study. All patients underwent preop-erative CT angiography with the following
Figure 1. Vascular window (VW) at L4-L5 (A) andunder the vessels, and the L4-5 disk is exposedright iliac vein; LIV, left iliac vein.
protocol:
WORLD NEUROSURGERY 79 [5/6]: 784-7
y Helical CT scans were acquired from L2to S2 using the Brilliance 40 CT scanner(Philips Healthcare, Andover, Massa-chusetts, USA) with a thickness of 1.4mm and an overlap of 0.6 mm.
y An iodine contrast agent (2.5 mL/kg[150 mL minimum]; Xenetix 350 [Guer-bet, Roissy, France]) was administratedintravenously at 2 mL/s and was fol-lowed by injection of 60 mL of salineserum.
y Settings of the CT scanner were as fol-lows: 120 kV and 270 mAs per slice,standard resolution, collimation 40 �0.625, pitch 0.924, 0.5 second for rota-tion time.
(B). In most cases, the L5-S1 disk is exposedl to the left iliac vessels. IC, iliocavum; RIV,
Figure 2. Voluminous iliolumbar ascending vein (aand on native axial cuts (B). Before medial mobipresence of the iliolumbar ascending vein becau
massive intraoperative bleeding. LIV, left iliac vein.
91, MAY/JUNE 2013 ww
y The acquisition was performed at 120seconds.
y 3D reconstructions were generated withthe imaging software (EBW, CT viewer,AVA software; Philips, Suresnes, France)using the volume-rendering technique(VRT) to obtain the optimal spatial resolu-tion of the vascular and bone structures.
The following anatomic parameters wereetermined:
y Level of aortoiliac bifurcation in refer-ence to intervertebral disks and vertebralbody from L3 to L5-S1
y Level of iliocavum confluence in refer-ence to intervertebral disks and vertebralbody from L3 to L5-S1
y Presence of the ascending iliolumbarvein
y Presence of central sacral vessels
y Presence of anatomic variations and ath-erosclerosis
y Nature of vascular window at L5-S1(three types were defined—type 1, lim-ited by right iliac artery–left iliac vein;type 2, limited by right iliac vein–left il-iac vein; type 3, limited by right iliac ar-tery–left iliac artery)
The following geometric parametersere measured:
y Vascular window at L5-S1 in millimetersdefined as the distance between the
visible on three-dimensional reconstruction (A)of the left iliac vein, one must verify the
ulsion from the iliac vessel could result in
L5-S1latera
rrow)lizationse av
w.WORLDNEUROSURGERY.org 785
wapp
R
TmyIdds
i
1tw(L
i8do
oasttl
i1tnaaiw1f�umffc
if
t iliac
PEER-REVIEW REPORTS
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
most medial iliac vascular structure (ar-tery or vein) on the right side and themost medial iliac vascular structure (ar-tery or vein) on the left side, correspond-ing to the “free vascular” area at the an-terior part of the L5-S1 disk (Figure 1)
y Vascular window at L4-L5 in millimetersdefined as the distance between the lat-eral border of iliocavum vein and lateralborder of L4-L5 disk, corresponding tothe “free vascular” area at the left antero-lateral part of the L4-L5 disk (Figure 1).
For the L5-S1 level, the vascular windowas calculated on 3D reconstructions (VRT)
nd on native axial cuts. Independent-sam-les Mann-Whitney test was used to com-are the two methods.
ESULTS
his study included 146 patients with aean age of 46 years � 12 (range, 24 –72
ears); there were 82 women and 64 men.ndications for the anterior surgical proce-ure were L4-5 or L5-S1 degenerative diskisease in 120 cases and low-grade isthmicpondylolisthesis in 26 cases.
46), and L4-L5 in 14.4% (21 of 146) of pa-ients. The level of iliocavum confluenceas L4 in 13.7% (20 of 146), L4-L5 in 38.4%
56 of 146), L5 in 43.8% (64 of 146), and5-S1 in 4.1% (6 of 146) of patients. The
evere (B) atherosclerotic disease of aortoiliacartery.
Figure 4. Two illustrative cases of anatomic variathe left common iliac vein (arrow). Knowledge opermits the surgeon to determine the preferable
avoid vein avulsion. AO, aorta; LIV, left iliac vein; RintI
WORLD NEUROSURGERY, http://d
liolumbar ascending vein was seen in4.2% of cases (123 of 146) (Figure 2). Me-ian sacral vessels were identified in 71.9%f cases (105 of 146).
Atherosclerosis was found in 22.6% (33f 146) of patients (Figure 3). Concerningnatomic variations, we found tortuous ves-els in three cases and abnormal configura-ion of venous anatomy in two cases withhe right internal iliac vein draining into theeft common iliac vein (Figure 4).
The vascular window at L5-S1 was type 1n 65.1% (95 of 146), type 2 in 30.1% (44 of46), and type 3 in 4.8% (7 of 146) of pa-ients (Figure 5). The vascular window wasever limited by the right iliac vein–left iliacrtery in our series. Measurements of L4-L5nd L5-S1 vascular windows are presentedn Table 1. At L5-S1, the vascular window
as measured to �15 mm in 5.5% (8 of46), from 15–25 mm in 19.2% (28 of 146),rom 25– 45 mm in 48.6% (71 of 146), and
45 mm in 26.7% (39 of 146) of cases (Fig-re 6). At L4-L5, the vascular window waseasured to �15 mm in 17.8% (26 of 146),
rom 15–25 mm in 48.6% (71 of 146), androm 25– 45 mm in 33.6% (49 of 146) ofases and was never �45 mm (Figure 6).
Variability of the vascular window at L5-S1s illustrated in Figure 7. The mean differenceor L5-S1 vascular window measurement on
ith the right internal iliac vein draining intovariations before the surgical procedure
tion for mobilization of the abnormal vein to
and s
tions wf such
direc
V, right internal iliac vein; RIA, right iliac artery.
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
axial native slices and on VRT was 1.2 mm �1.8 (range, �2.8 to 3.1 mm; P � .05).
DISCUSSION
Vascular complications during anteriorlumbar spine surgery are regularly re-ported in the literature, and the incidenceis estimated to be 2%–12% (1, 7, 8, 11, 13,14, 19, 21, 26, 31). These complicationsinclude vein or artery injury, vein throm-bosis, and arterial occlusion. Vascular in-juries during the surgical procedure pre-dominantly involve veins (12 of 13vascular injuries in the series reported byGarg et al. (8)) and may lead to massive,potentially life-threatening intraoperative
Table 1. Measurement of L4-L5 and L5-S
L5-S1 window
Axial native slices
VRT
L4-L5 window Axial native slices
Figure 5. The typical vascular window at L5-S1 wvein and the right iliac artery (A). In approximatevascular window was limited by the right and le
VRT, volume-rendered technique.
WORLD NEUROSURGERY 79 [5/6]: 784-7
leeding. In this context, the preoperativevaluation of the vascular anatomy may beseful to limit the complications related
o vascular structures.To our knowledge, our study is the larg-
st series investigating in vivo vascular anat-my of prevertebral great vessels. Our studyuggests that there is a great variability ofrevertebral vascular anatomy in the lum-ar spine, especially concerning the rela-
ionships between the L4-5 and L5-S1 disksnd the great vessels. In our experience, the5-S1 disk is typically exposed under theessels, whereas the L4-5 disk is exposedateral to the vessels (Figure 8). The accessi-ility to the L4-5 and L5-S1 disks was evalu-ted by the concept of vascular window. As
cular Windows (in mm)
an SD Minimum Maximum
.5 12 8 66
.7 12 8 67
8 4.5 39
ited by the left iliac-third of patients, theveins (B). More
rarely, the vascular win(C). IA, iliac artery; IV,
91, MAY/JUNE 2013 ww
xpected, the vascular window was smallert L4-L5 than at L5-S1. However, even at5-S1, the vascular window could be veryarrow (minimum 8 mm). It was �25 mm
n approximately one in four patients. At4-L5, the vascular window was �25 mm inne in three patients. The larger the win-ow, the less mobilization of the vessels isecessary during the procedure. The need
was limited by the right and left iliac arteriesein.
Figure 6. Distribution of L4-L5 (A) and L5-S1
1 Vas
Me
34
35
23
as limly one
dowiliac v
(B) vascular windows.
w.WORLDNEUROSURGERY.org 787
tivd(tswwoCvppetbi
gtTtvmicpv3cSrt
raaTstmoeAodceilvbccos
ctsimf3old
C, sm
PEER-REVIEW REPORTS
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
for mobilization of iliac vessels is clearlyassociated with a risk of vein injury. Thisrisk can be due to weakness of the vein wallbut also to the presence of venovertebralveins as described by Maeng et al. (23).These veins drained into posterior marginof the left common iliac vein from the L5vertebral body. They are too small to beidentified with preoperative imaging. In aseries of 50 patients, Maeng et al. (23) iden-
Figure 7. Illustrative cases showing the great varvascular anatomy. Very narrow (A), narrow (B), nL5-S1 vascular windows (star) are presented. In
Figure 8. Intraoperative view shows the idealrelationships between L4-5 and L5-S1 disksand left iliac vessels. The course of the leftiliac vessels offers large working windows atthe two levels. This patient underwent atriple anterior lumbar interbody fusionprocedure from L3-L4 to L5-S1 for multileveldegenerative disk disease. The procedurecould be performed without the need of anaccess surgeon. LIA, left iliac artery; LIV, left
tiliac vein.
788 www.SCIENCEDIRECT.com
ified the presence of the venovertebral veinntraoperatively in 74% of patients. Mostenous injuries during the surgical proce-ure resulted from branch vessel avulsionsvenovertebral veins) or lateral wall lacera-ion during mobilization (8). This findinguggests that the vascular risk is correlatedith the need for vessel mobilization,hich can be evaluated preoperatively. Inur study, vascular windows measured onT angiography could predict the need for
essel mobilization and evaluate, at leastartially, the vascular risk of the surgicalrocedure. One limitation is that the adher-nce of the vessels, especially the veins, tohe anterior longitudinal ligament cannote reliably predicted by preoperative imag-
ng.Otherwise, data obtained from CT an-
iography probably influence the surgicalechnique making the procedure safer.he access to the L4-5 disk often necessi-
ates mobilizing medially the left iliacein (15,17,18). During mobilization, oneust take care not to avulse the ascending
liolumbar vein, which drains into the leftommon iliac vein (Figure 2). In mostatients (84%), the size and course of thisein could be identified on preoperativeD CT angiography. This information fa-ilitated the approach to the L4-5 disk.imilarly, in the presence of atheroscle-otic arteries, retraction should be limitedo avoid migration of calcifications or ar-
of the L5-S1l (C), and large (D)all vessels to the left
of the star correspondaorta; VC, vena cava.
erial thrombosis. a
WORLD NEUROSURGERY, http://d
Compared with native axial slices, 3Deconstructions provide the surgeon with
better understanding of the vascularnatomy in the lumbar spine (Figure 9).he surgeon can visualize the relation-hips between the vascular structures andhe lumbar spine in all directions. As
entioned earlier, a clear understandingf the vascular anatomy may avoid unnec-ssary exposure in the lumbosacral area.lso, potential difficulties for exposurewing to vascular structures may be pre-icted. In four personal cases (not in-luded in this series), we decided preop-ratively to convert the anterior approachnto posterior approach alone (posteriorumbar interbody fusion) because of theascular anatomy (Figure 10). It has alsoeen reported in the literature that ALIFould not be achieved in a significant per-entage of cases because of configurationf the great vessels (4 of 212 cases in theeries by Garg et al. (8)).
Although the difference was not signifi-ant, we observed a slight difference be-ween dimensions calculated on axiallices and dimensions calculated on 3Dmages. This difference, which could be 3
m, was thought to be related to severalactors. First, manual reconstructions ofD images may lead to underestimationf the size of the vessels. Second, a vascu-
ar window was measured at the L5-S1isk midplane, which could be not exactly
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
3D images. Third, there is a parallax effecton 3D images making measurements lessreliable than measurements made on ax-ial cuts. For these reasons, we advise ver-ification of measurements from 3D im-ages on native axial cuts.
Our results were consistent with resultsreported in the literature (5, 15). Inamasu etal. (15) reported a 3D CT anatomic study in apopulation of 100 subjects with nonspinaldiseases. As observed in our study, theseinvestigators found that the aortic bifurca-tion was mostly observed at L4 and at L5(upper part) for iliocavum confluence.
Figure 9. Different views of the prevertebral greawith volume-rendered technique (for the same p(3D) computed tomography (CT) angiography pevascular anatomy in all directions. 3D CT angiog
Datta et al. (5) analyzed 76 CT angiography L
WORLD NEUROSURGERY 79 [5/6]: 784-7
tudies from patients who underwent min-mally invasive anterior surgery of the lum-ar spine. They concluded that data fromreoperative CT angiography influencedurgical decisions in 21% of patients andorrelated well with intraoperative observa-ions in all patients. There were no mea-urements of vascular windows in these twotudies.
Results from cadaver studies have alsoeen reported in the literature (6, 25, 29).ribus and Belanger (29) found 33.5 mmidth between the left common iliac vein
nd the right common iliac artery at the
els reconstructedt). Three-dimensionalanalysis of thegives the surgeon a
complete understandinthe lumbar spine and pmobilization. Ao, aorta
5-S1 level, which is particularly consistent w
91, MAY/JUNE 2013 ww
ith our findings on CT angiography. How-ver, in vivo studies are more pertinent tonalyze the vascular anatomy because ofein collapse in cadaver specimens.
CT angiography results in significantadiation exposure for patients. In a seriesf 28 patients, Gstöttner et al. (9) con-luded that CT angiography should bevoided to prevent radiation-related com-lications because data provided by thexamination did not change the surgicaltrategy. In contrast to some protocolseported in the literature (5), to reduce theose of radiation received by the patient,
elationships between the great vessels ands the surgeon to predict the need of vessel
vena cava.
t vessatienrmits
g of rermit
; VC,
e decided to obtain only one acquisition
w.WORLDNEUROSURGERY.org 789
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
PEER-REVIEW REPORTS
CÉDRIC BARREY ET AL. VASCULAR WINDOW FOR SURGICAL ACCESS TO DISKS
SPIN
ES
PIN
E
at 120 seconds instead of two acquisitions(typically at 20 seconds for arterial flowand at 120 seconds for venous flow). Us-ing this protocol, the quality of imageswas sufficient in all patients to evaluatethe vascular anatomy and permitted sig-nificant reduction of the dose of radiationeven if optimized CT venography shouldallow for a better identification of venousstructures. The other major risks of CTangiography are contrast agent allergyand renal insufficiency.
CONCLUSIONS
We found that preoperative assessment ofvascular anatomy can be achieved using 3DCT angiography and may contribute to limit-ing perioperative complications of anteriorlumbar spine surgery involving vascularstructures. By predicting the need for vesselmobilization, the assessment of a vascularwindow may help the surgeon to choose thesafest and most appropriate approach.
REFERENCES
1. Brau SA, Delamarter RB, Schiffman ML, WilliamsLA, Watkins RG: Vascular injury during anteriorlumbar surgery. Spine J 4:409-412, 2004.
2. Burns BH: An operation for spondylolisthesis. Lan-cet 10:1233-1234, 1933.
3. Capellades J, Pellisé F, Rovira A, Grivé E, Pedraza
Figure 10. No vascular window at L4-L5 investigatomography (CT) angiography (A) and axial CT scauthors’ decision to avoid the anterior approachvein; RIV, right iliac vein; RIA, right iliac artery.
S, Villanueva C: Magnetic resonance anatomicstudy of iliocava junction and left iliac vein posi-
790 www.SCIENCEDIRECT.com
tions related to L5-S1 disc. Spine 25:1695-1700,2000.
4. D’Aliberti G, Talamonti G, Debernardi A, SansaloneCV, LaMaida A, Torre M, Collice M: Anterior ap-proach to thoracic and lumbar spine lesions: resultsin 145 consecutive cases. J Neurosurg Spine 9:466-482, 2008.
5. Datta JC, Janssen ME, Beckham R, Ponce C: The useof computed tomography angiography to define theprevertebral vascular anatomy prior to anterior lum-bar procedures. Spine 32:113-119, 2007.
6. Ebraheim NA, Xu R, Farooq A, Yeasting RA: Thequantitative anatomy of the iliac vessels and theirrelations to anterior lumbo-sacral approach. J Spi-nal Disord 9:414-417, 1996.
9. Gstöttner M, Glodny B, Petersen J, Thaler M, BachCM: CT angiography for anterior lumbar spineaccess: high radiation exposure and low clinicalrelevance. Clin Orthop Relat Res 469:819-824,2011.
0. Gumbs AA, Bloom ND, Bitan FD, Hanan SH: Openanterior approaches for lumbar spine procedures.Am J Surg 194:98-102, 2007.
3. Ikard RW: Methods and complications of anteriorexposure of the thoracic and lumbar spine. ArchSurg 141:1025-1034, 2006.
4. Inamasu J, Guiot BH: Vascular injury and complica-tion in neurosurgical spine surgery. Acta Neurochir148:375-387, 2006.
5. Inamasu J, Kim DH, Logan L: Three-dimensionalcomputed tomographic anatomy of the abdominalgreat vessels pertinent to L4-L5 anterior lumbar in-terbody fusion. Minim Invasive Neurosurg 48:127-131, 2005.
6. Jarrett CD, Heller JG, Tsai L: Anterior exposure ofthe lumbar spine with and without an “access sur-geon”: morbidity analysis of 265 consecutive cases. JSpinal Disord Tech 22:559-564, 2009.
7. Kang BU, Lee SH, Jeon SH, Park JD, Maeng DH,Choi YG, Tsang YS: An evaluation of vascular anat-omy for minilaparotomic anterior L4-5 procedures.J Neurosurg Spine 5:508-513, 2006.
8. Kleeman TJ, Michael Ahn U, Clutterbuck WB,Campbell CJ, Talbot-Kleeman A: Laparoscopic ante-rior lumbar interbody fusion at L4-L5: an anatomicevaluation and approach classification. Spine 27:1390-1395, 2002.
9. Kulkarni SS, Lowery L, Ross RE, Ravi Sankar K,Lykomitros V: Arterial complications following an-terior lumbar interbody fusion: report of eightcases. Eur Spine J 12:48-54, 2003.
0. Lane JD, Moore ES: Transperitoneal approach to theintervertebral disc in the lumbar area. Ann Surg 127:537-551, 1948.
Iulius Casserius and the First Anatomically Correct Depiction of the Circulus ArteriosusCerebri (of Willis)
INTRODUCTION
The circulus arteriosus cerebri is the arterialanastomotic circle at the base of the brainthat connects the bilateral internal carotidartery (ICA) systems with each other via theanterior communicating artery and alsowith the vertebrobasilar system via the pos-terior communicating arteries (PCoAs). Al-though this structure is common knowl-edge today, its discovery spanned morethan 14 centuries and was the result of along line of painstaking anatomic research,which started with the description of the rete
Key words� Anatomy� Casserius� Circle of Willis� Circulus arteriosus� Padua� Thomas Willis
Abbreviations and AcronymsICA: Internal carotid arteryPCA: Posterior cerebral arteryPCoA: Posterior communicating arterySCA: Superior cerebellar artery
Department of Neurosurgery, The JohnsHopkins University School of Medicine,
Baltimore, Maryland, USA
To whom correspondence should be addressed:Rafael J. Tamargo, M.D. [E-mail: [email protected]]
Citation: World Neurosurg. (2013) 79, 5/6:791-797.DOI: 10.1016/j.wneu.2011.10.044
Galenus of Pergamum (Galen) (A.D. 129 –ca. 202) in his major work entitled De usupartium (On the Usefulness of the Parts), whichhe wrote in Greek and finished betweenA.D. 169 and 175 (9), and culminated in theelucidation of the physiological importanceof the circulus arteriosus by Thomas Willis(1621–1675), the English anatomist andphysician.
Willis and his collaborators describedand illustrated this structure in his book
The circulus arteriosus cerebri is thethe brain, now better known as thephysician and anatomist Thomas Willithe physiologic function and observe t
as been overlooked, however, that thas provided by the Paduan anatomis
1552–1616) in two engravings publincluding the Tabulae anatomicae LXXVesalian anatomists” at the Universitost important discoveries in anatom
eview the life of Casserius, his’Acquapendente (Fabricius) (1533–161f Padua, his research in comparativerteriosus cerebri. Although previousortrayal of the circulus arteriosus inone have discussed Figure 2 of Tablenterior communicating artery complccurate course of the posterior commogether, Tables 9 and 10 represent a seserves recognition as the first accur
he base of the brain.
et usus (Anatomy of the Brain and the Descriptionand Use of the Nerves), published in 1664 (29).Accordingly, the circulus arteriosus is nowknown as the circle or polygon of Willis.Willis initiated his neuroanatomic studiesin the new era of anatomic physiology ush-ered in by William Harvey’s (1578 –1657)1628 publication of his landmark treatiseon the circulation titled Exercitatio anatomicade motu cordis et sanguinis in animalibus (AnAnatomic Exercise on the Motion of the Heart and
rial anastomotic circle at the base ofle or polygon of Willis. The British21-1675) was the first to demonstratelinical significance of the circulus. Itst accurate depiction of the circulus
ulio Cesare Casseri (Iulius Casserius)posthumously in multiple formats,
627). Casserius was the fifth of the sixadua, Italy, which was the site of the
the 16th and 17th centuries. Here wefrom servant to Girolamo FabrizioProfessor of Surgery at the University
omy, and his depiction of the circulusors have commented on Casserius’sle 10 of Tabulae anatomicae LXXIIX,his is important because whereas thes depicted clearly in one table, thecating arteries is shown in the other.sticated, sequential dissection, whichortrayal of the arterial anastomosis at
artecircs (16he ce firt GishedIIX (1y of Py inrise9) toanatauthTab9. T
