Fausto Passariello
Centro Diagnostico , Napoli, Italy

The evaluation of active collateral circulation in cerebrovascular pathology must always integrate each exam of supra-aortic trunks.
Visibility  (Table 1) of brachiocefalic trunk, subclavian artery and vertebral ostium by Echo (Pulsed Wave or P.W.) or ColorDoppler (Color) is often inadequate on several commercial echo devices, depending on the vascular probe morphology, which does not fit well to neck base conformation. Furthermore, an emergency surgical indication as the pre-occlusive stenosis of internal carotid artery, with poor residual flow but very high velocity, is unrecognised at all by P.W. and Color because of aliasing action, while in experts hands it is clearly recognised by Continous Wave Doppler (C.W.). The operator experience is clearly a limiting factor in the use of C.W.
Ophthalmic artery cannot be explored by P.W. or Color, i.e. by these devices it is not possible to perform an ophthalmic test.
At last, clinic and instrumental manoeuvres to explore collateral circulation are more easily performed by C.W.
For these reasons C.W. Doppler should always support also today the exam performed by more advanced devices and we all should be quite diffident about pure morphological examinations of the district of supra-aortic trunks.
 
 

Cerebral Arterial net
Omitting only one point (brachio-cefalic trunk), it can be regarded as an essentially symmetric structure, which can be usefully represented in a scheme  (Fig. 1).
If an occlusive pathology or a haemodynamically significant stenosis with intravasal downstream pressure fall is present, a remodelling of flow direction occurs in other vessels, depending on mechanic energy gradients.
From a macroscopic point of view compensation is a sustaining or substitutive action, exerted by vessels in the mechanism aimed at restoring circulation towards a district.
Compensatory circulation recognition is useful  Table 2  for reasons of pure classification of cerebro-vascular pathology, but also for prognostic reasons, for indication, programming and following control of the surgical and/or loco-regional thrombolytic therapy.
Evaluation of compensation must take into account the normal structure of the net and recognise the sites where each patient is far from normality.
To make all this the Operator must query the net, making on it changes and clinical and instrumental manoeuvres, which correspond to temporary changes of its structure and function.
The functional manoeuvre is essentially the upper limb hyperaemia, while there are 8 kinds of compressions in each side .  (Table 3)
Echoing past polemics on compressive manoeuvres, this can be summarised saying that they are essentially harmless, if performed following an exact rule:
l do not compress in zones where a stenosis has been detected
l always compress softly and only after having localised the pulse
l compress at maximum for one or two cardiac cycles
However it must be asserted that some compressions are difficult to be exerted, while there are combined ones which are not outlined in the table.
A classification of compensatory paths is reported in  Table 4 , where they are classified into compensatory circles and derivations, following Franceschi classification. (1)

Conpensation evaluation technique
Only notable cases will be reported. For a comprehensive treatise look at References.
For each manoeuvre consider an acquisition point, a compression point and a velocimetric effect.

Exploration of Willis Circle
If carotids and vertebrals are patent, consult  Scheme A .
After compression (i) to before compression (h) flow ratio i/h is the Compensatory Potential of the anterior or posterior communicant artery.
If internal carotid is occluded proximal to ophthalmic:
-  if ophthalmic is not inverted, consult  Scheme B .
-  if ophthalmic is inverted, consult  Scheme C .

Exploration of supra-thyroid circle
The action of supra-thyroid circle, often in the occlusion of external carotid artery, is assessed acquiring the signal on superficial temporal and facial arteries and compressing contra-lateral common carotid artery, getting a flow reduction if compensation is active. In the case that brachio-cephalic trunk is occluded, compensation comes from superior thyroid arteries if the subclavian compression reduces the flow, which instead is increased by reactive hyperaemia.
If there is an hyper-vascularized goitre, then the manoeuvre is not reliable.

Exploration of sub-thyroid circle
The action of sub-thyroid circle is assessed acquiring the signal on inferior thyroid artery and compressing the subclavian. A flow reduction is got if conpensation is active, while the homo-lateral upper limb hyperaemia manoeuvre causes instead an increase.
If there is an hyper-vascularized goitre, then the manoeuvre is not reliable.

Exploration of sub-occipital circle
These anastomoses are not generally visible in normal conditions.
Vertebral occlusion at ostium:
l the vertebro-occipital anastomosis is active if flow at Tillaux is zeroed or inverted by homo-lateral common carotid compression. Vertebral spontaneous flow can be an intermittent steal, completed by homo-lateral common carotid compression.
l revascularization trough thyreo-bi-cervico-scapular trunks (TTBCS) is active if the homo-lateral common carotid compression increases or leaves unchanged Tillaux flow, while sub-occipital compression zeroes or inverts vertebral flow.

Exploration of vertebro-subclavian derivation
It is constituted by the subclavian and vertebral arteries on the left, while on the right the brachio-cephalic trunk must be added.
Pre-vertebral subclavian stenosis causes vertebral compensation with the subclavian steal syndrome or vertebro-subclavian derivation, with pre-steal, intermittent steal and completed steal variants.
Subclavian compression
(or homo-lateral common carotid, if posterior communicants are patent) reduces the steal, while hyperaemia instead
increases it.
The occlusion of brachio-cephalic trunk can be compensated by an inverted carotid, refilled by Willis circle trough the anterior communicant artery. Right common carotid flow is reduced by right subclavian compression, while it zeroes by compression of left common carotid artery. Vertebral and right subclavian flow is zeroed by compression of right carotid artery.
If compensation comes trough right vertebral, it is inverted or reduced by subclavian or right carotid compression. Right carotid flow is towards directed and is increased by subclavian compression, while is reduced or inverted by hyperaemia.
 
 

The ophthalmic artery originates in correspondence of anterior clinoid process, as a collateral branch of internal carotid artery, at the exit of its intra-petrous path. Having penetrated the optical foramen lateral to homonymous nerve, emits the lachrymal and the retina central arteries. Crossing the optic nerve in lateral to medial and posterior to anterior direction, gives origin to supra-orbital artery, to short and long posterior cilia arteries, to superior and inferior muscular arteries.
Taking place medial to optical nerve, emits the posterior and anterior ethmoid arteries and the inferior and superior medial palpebral arteries.
In correspondence with the reflection troclea of superior oblique muscle, it divides at last into two terminal branches: the supra-troclear a., which directs itself upwards and medially towards forehead, and the dorsal of nose a., a more voluminous one, which directs itself downwards and medially, making an anastomosis with the angular a., branch of the facial artery (8)
From the haemodynamic and clinic point of view the supra-orbital, supra-troclear and dorsal of nose branches are very important, for their superficial position which allows an easy identification and for their anastomoses with the face superficial circle, fed by branches of external carotid artery.
The ophthalmic circle constitutes so a physiologic anastomotic path between the endocranic arterial circulation and the muscle-skeletal flow. In particular, the internal carotid a. has an essential role in the feeding of cerebral parenchyma, while the external carotid artery sends its blood mainly to muscle and cutaneous structures of face.
Perfectly in line with these data are the lower peripheral resistance, the higher diastolic velocity and the greater pressure regimen of the district of the internal carotid, as compared to that of the external carotid artery. I.e., the pressure gradient is directed towards the extern and the ophthalmic circle is fed by an out of encephalon flow.
In the sclerotic obstructive pathology of cerebral afferent vessels, this pre-constituted anastomotic path can have an important role of compensation, in order to derive blood over the obstacles of the main conduction paths.
When the percent stenosis of the internal carotid artery is over a critic value, the pressure inside the muscle-cutaneous district dominates and the anastomoses of the ophthalmic district invert their flow, sending the blood into the encephalon.
The described physiopathologic event is undoubtely the ideal case, which is at the base of the ophthalmic test. In clinical practise however the circulatory compensation to a sclerotic obstructive pathology of the internal carotid artery can choose ways, depending directly on possible anatomical variations and on integrity status of the other cerebral afferent vessels.
The correct interpretation of test result requires a careful knowledge of ocular vascularization and a comprehensive view of the possible compensatory circles, in case of stenosis/occlusion of the internal carotid artery.  (Table 5)
Observations made on corrosion specimens in differently aged subjects show that the orbital vascular structures are joined by a close arteriolar and capillary net into a vascular unit (orbital plexus), which allows many entry points to blood flow. (9) From anterior side, the plexus has the classical anastomoses facial/dorsal of nose and superficial temporal/supra-orbital, which connect the endocranic and esocranic districts. From medial side, instead, trough the anterior and posterior ethmoid a., communicates with branches of no moment of the medial region of orbit. From lateral side, then, there are two main paths of input flow, the ophthalmic accessory a., branch of the middle meningeal a., and an anastomotic branch coming from the internal maxillary artery. Both these last paths carry flow coming from deep branches of the external carotid artery. The activation of the lateral anastomoses of the orbital plexus can at the level of the Doppler acquisition points cause the meeting of two flows, superficial and deep, contrasting in directions, but both fed by the external carotid artery. In this case, no information can be extracted from ophthalmic test results, as the orbital plexus does not receive blood from the internal carotid artery.
Furthermore, considering phylogenesis of encephalic vascular system, note that the internal carotid artery has not the same role in man as in the inferior mammals. In these last, indeed it develops quite early during the embryo life, but degenerates as soon as well functioning anastomoses originate between its distal part and the deep branches of the external carotid a., generally the ascendant pharyngeal a. (9)
These anastomoses take place trough a vessel knot, named rete mirabile caroticum.
Present in the sheep, ox, pig and cat, the rete mirabile caroticum does not exists normally in man, but its presence was signalled in case of moya-moya disease, caused by spontaneous degeneration of proximal part of the internal carotid artery.
The human analogue of the rete mirabile caroticum has the angiographic aspect of a smoke cloud of cigarette, sometimes in low and extra-cranial anatomical position, sometimes more upwards, so that the compensatory circle, depending on the case, can be defined intra or extra-cranial.
This citation, a right one from the theoretic point of view, must however consider the extreme rarity of this disease, present especially in Japan.
Always in the case of anatomical anomalies, the intra-cranial communication between the vertebral and the carotid artery must be reminded. (10)
This rare compensatory circle is not distinguishable, if not angiographically, from the compensatory action of basilar trunk, trough the posterior communicant a. Trough the anterior communicant a., instead the compensatory flow can be carried coming from the contra-lateral internal carotid artery, with normalisation of flow direction in the orbital plexus.
Downstream the Willis circle, furthermore several anastomoses are active between vessels of very little calibre, which though they are important for the distal vascular connection between contiguous encephalic territories, however are not examinable by cervico-encephalic Doppler exam.
The anatomical knowledge of the physiologic or pathologic input paths to orbital plexus clarifies the different behaviour to ophthalmic test in cases of revascularization of plexus by the external carotid artery. Indeed, ophthalmic flow has an inverted direction only in the cases of presence of the superficial branches of external carotid.
If the plexus instead has important interactions with the deep branches, the ophthalmic flow is physiologically directed towards the extern of cranium, causing so a false negative result to Doppler exam.

Elements of Doppler acquisition technique
It is necessary to use a Doppler bi-directional Continous Wave velocimeter, with a high frequency probe for superficial vessels, 8 or 9.5 MHz. With the patient supine, the Operator is at the head of the bed, applying the probe on the prescribed repere points. To avoid uncontrollable movements, the probe must be taken in site by the velocimeter connection cable. As soon as acquired, the Doppler signal must be translated into positive, in analogy to vertebral signal at Tillaux triangle.
The information on the flow direction is inferred only from the compression manoeuvres. The existence of several ophthalmic branches, detectable with Doppler exam, obliges to the acquisition in several elective sites, in case of signal absence in one of them (Flow-chart 1).
The nasal branch (dorsal of nose/angular of nose) is acquired at the superior-medial angle of the orbit, while the supra-troclear is a little bit upwards and laterally. The supra-orbital instead is found putting the probe on the skin of superior palpebra, directed towards the eyebrow, one cm. upwards and laterally to corneal middle point.
Several Authors do not agree about the most informative site of acquisition. The supra-orbital would false partially the results, because originated in almost 30% of cases directly from mean pharyngeal a., branch of the external carotid artery. The nasal branch instead would be more informative, because almost in all cases originated from the ophthalmic. (4)
Other Authors: instead have the opinion that the supra-orbital a. gives more reliable informations, in agreement with angiographic data. (11)
In my routine work, I generally acquire firstly the dorsal/angular of nose, which in my experience is more sensible to compression manoeuvres on the external carotid branches. In any case, before concluding for an absence of ophthalmic flow, the signal acquisition in all cited sites must be attempted. (Flow-chart 1)

General criteria for exam management
Some aspects of the acquisition method are essential to the comprehension of adopted terminology. First, flow direction in a vessel detected by Doppler technique depends on the angle between the ultrasound beam and the vessel axis. In the case of the ophthalmic artery, it is possible to see that putting the probe at the superior medial angle of the orbit and changing a little bit the probe direction, sometimes also flow direction changes unpredictably, so demonstrating that if detected only by Doppler acquisition then flow direction has no reliability.
This exam instead is easy if a compression is achieved of the superficial branches of external carotid artery, while recording on the ophthalmic. It is trivial that an increased flow response gives evidence for a physiologic direction from inwards towards the external of cranium. On the contrary, if ophthalmic flow reduces, zeroes or inverts, we are looking at an inverted flow, which has several pathologic meanings, to be defined later on during the same exam. No conclusion can be drawn instead when the ophthalmic flow is unchanged after the compression manoeuvre. Furthermore, flow invariance can be due also to an erroneous compression manoeuvre, exerted on an inadequate anatomical site and so it must always be studied carefully.
Possibility of modulating ophthalmic flow, by compression of superficial temporal and facial arteries, tells us that these vessels constitute an esocranic flow source, which can also be defined as a primary flow source, i.e. with higher pressure, when the ophthalmic artery inverts. On the contrary, it is a secondary flow source, i.e. with lower pressure, when the ophthalmic artery maintains its physiologic direction.
Generally an endocranic flow source is opposed to the esocranic source. The first one can be defined secondary when the ophthalmic is inverted, primary instead when blood flows out of the encephalon.
Furthermore, primary source flow is detected by the simply recording procedure, while secondary source flow is stressed by suppressing by compression the primary sources.
A first conclusion is possible. Ophthalmic inversion is always sign of pathology, while a physiologic flow direction does not always indicate normal haemodynamics. This happens because flow direction is caused by pressure interactions in downstream vascular districts of internal and external carotids. It is so possible to get a false negative for low percent stenosis of internal carotid artery, where the pressure reduction is low, or in severe percent stenoses, associated to sclerotic pathology of external carotid artery, when to pressure fall in the carotid siphon an analogue hypotensive regimen is opposed in the branches of the external carotid artery.
The set of informations which can be inferred from the ophthalmic test overcomes these quite generic data in their physiopathologic meaning.
Generally, it can be said that the ophthalmic test can be reorganised in three arterial compressions. One is performed in a combined mode on superficial and facial arteries, i.e. on vessels which are too close to the anastomosis. The others on two vessels which are quite distant from it, i.e. separately on the two common carotid arteries, at the base of the neck and lasting one or at maximum two cardiac cycles. These manoeuvres are anyway prohibited in the case that the exam stresses pathologic alterations in compression sites.
Possible haemodynamic responses at ophthalmic level are five: increase, invariance, reduction, zero-flow, inversion. Aside of invariance, because of already described reasons, and of zero flow, because of reasons we will see later on, remaining responses can be grouped into two functionally opposed sets (Table 6).
In a first group we put the ophthalmic flow inversion and reduction, in the second group we insert instead the increase of the same flow. This classification separates functionally antithetical haemodynamic responses, i.e. inversion and reduction take place when higher pressure primary sources (or destinations) are compressed, while flow increase is specific of the exclusion of lower pressure secondary sources. Zero-flow response could rightly be introduced into the first group, only for the closer compressions, but it is better take it alone, because it has no antithetical correspondence, because in this case the secondary source is absent. Furthermore, zero-flow for distant compressions, when it is already demonstrated the existence of a secondary source, is due to the synchronous exclusion of both the sources, primary and secondary.
The use of this classification is immediate.  (Table 7)
If with a distant compression we get a response of the same group of that got with the closer compression, then the vessel compressed at distance has the same functional meaning, i.e. it is also an esocranic source. On the contrary, if we get an antithetical response, compressed vessel has an antithetical function, i.e. it is to be considered an endocranic source. If finally, flow zeroes, it is clear that the vessel feeds in the same time eso and endocranic districts.
An example can clarify simply all that was only told
as theory. In the normal subject (Fig. 2) compression of homo-lateral external carotid branches produces the increment of the ophthalmic flow. Superficial temporal and facial arteries are so definable as esocranic and secondary sources of the ophthalmic flow. Compressing homo-lateral common carotid artery, we get zero-flow, demonstrating then (aside of a case of ophthalmic fed by deep branches of external carotid artery) that the common carotid a. sends its flow to external and internal carotids. This last then is patent and does not have a stenosis capable of changing significantly its pressure regimen. In this case, common carotid a. is an esocranic and endocranic source of ophthalmic flow. If on the contrary the compression of homo-lateral common carotid would cause the ophthalmic flow increase, in the same measure got with the compression of the external carotid branches, then common carotid would be an esocranic source and the result would be typically pathologic, giving evidence of the obstruction of the homo-lateral internal carotid artery and of an effective conpensation trough Willis circle.
Next to these quite paradigmatic cases there is a crowded set of not typical cases, for whom always a haemodynamic reasoning is needed: Sometimes then the examination must go far a little bit from the described scheme, introducing new compressions, with an use which is clear from the careful reading of Flow-charts 1, 2, 3, 4, 5, 6.
The conclusion of this methodological premise is to clarify the aim of the ophthalmic flow examination:
l identify the ophthalmic flow direction
l quantify approximately the lesion of the homo-lateral internal carotid artery
l identify the eso and endocranic sources of ophthalmic flow
l extract from the exam further hints to go deeply into Doppler examination.
 
 

General criteria shown in the previous paragraph are synthetic and referred to quite typical cases of haemodynamic response to compressive manoeuvres, during the acquisition of the ophthalmic Doppler signal. This method allows the Operator to maintain a rational behaviour, paying attention to the physiopathology of encephalic circle during the execution of the exam. Nevertheless, the wealth of response combinations to compressions obliges to the compilation of a diagnostic scheme, which makes it possible the interpretation of more unusual cases.
The model is taken from the language of Informatics, where the flow-chart is used to outline the subsequent logic and operative steps of a programme. Symbols used in this work are listed in Table 8 and the diagram does not present difficulties to be passed over with a specific knowledge.
The flow-chart is constituted by a way to go trough, which branches depending on different responses got to compression manoeuvres. These last are organised into three subsequent compressions (superficial temporal + facial, homo-lateral common carotid, contra-lateral common carotid), executed in an always identical order, to whom sometimes other compression manoeuvres are added.
Among them, the compression of the sub-parotid external carotid, which excludes the flow from the artery, before it emits its deep branches. (12) This manoeuvre can be useful to distinguish revascularizations of orbital plexus by deep branches from those by superficial branches of the external carotid artery. (Flow-chart 4).
Indeed, in the normal subject (Fig. 2) flow increase is got, in the same way as in the first compression, demonstrating that the external carotid artery is only an esocranic source. In the case of revascularization trough the deep branches instead, flow zeroes, demonstrating that the external carotid a. sends flow also in the towards direction in the plexus, qualifying itself as an eso and endocranic source of ophthalmic flow. The compression of sub-parotid external carotid is performed just behind the ascendant branch of jaw.
The compression of contra-lateral superficial temporal and facial is sometimes needed, but banished to a position of low importance, as part of the obtainable information is already comprised in the result of the compression of contra-lateral common artery. The same reasoning holds also for other minor meaning compressions.
In the flow-chart the case of flow invariance after the first compression of the scheme is not reported. This is because invariance is mostly subjected to error. In this case, after having repeated many times the manoeuvre (it is harmless) to avoid manual errors, it had better to execute other closer compressions, to localise a flow source, on which to base the following reasoning.
I.e., compressions of the homo-lateral sub-parotid external carotid or of contra-lateral vessels, which by modulating the ophthalmic flow could clarify their function at the anastomosis level. In case of invariance also in the other sites, it suites to suppose a low pressure regimen in the superficial branches of external carotid and to entry again into the flow-chart, as the answer to the first compression were an increased flow, considering normal the result, in the case that distant compressions do not present new reasons of doubt.
The set of clinical and experimental situations is however so various that it is not possible to predict all cases which theoretically can verify. I.e., particular anatomical situations or parietal stiffness status of vessels can make compressions impracticable at the base of the neck, so making vane the diagnostic possibilities of the ophthalmic test.
Furthermore, compressive manoeuvres in the diagnostic scheme are organised in a completely arbitrary sequence, suited to common practical methods of exam execution, but the flow-chart can be redrawn at the will of the Operator (not without an annoying work) in a completely different way, without minimally affecting the exam result. Planning an optimal scheme is not proposable, because it supposes the exact knowledge of the action of each of the possible collateral circles, making also a difference for each cerebro-vascular pathology.
At last, in the flow-chart are reported assertions which are only interesting working hypotheses (Flow-charts 4, 6), about the quantification of the percent stenosis of the internal carotid artery, on the base of the residual flow after compression of the homo-lateral common carotid, compared to that stressed by compression of superficial branches of the external carotid.
An example of a haemodynamic and clinic reasoning, based on ophthalmic test and referring to Flow-chart 6, is reported completely in Fig. 3.
In conclusion, the ophthalmic test is of quick execution and gives very detailed informations on the status of the encephalic haemodynamics. Ophthalmic acquisition does not require particular technical skill of the Operator and the method can be easily teached also to technical personnel, also if diagnostic responsibility is owned only by medical personnel.
Using a few easy arterial compressions, in not dangerous sites, gives in a short time indications to go deeper into Doppler investigation. So modified, the ophthalmic test seams re-proposable as a screening investigation, as the possibilities of getting false negatives are so greatly reduced.
Not taking away anything from the value of Doppler complete exam of cerebral circle, it is useful to know how to reduce the number of clinical and semeiologic manoeuvres to a quite limited set, controllable by the clinic reasoning.
 
 

C.W. Doppler examination should always be executed in the evaluation of the status of the supra-aortic trunks, the best association being still its execution during Echo
and ColorDoppler examinations.
It is still today the exam of choice for a not invasive and low cost evaluation of compensations in cerebro-vascular pathology.
However there is no superimposition between compensation detected by Doppler and the angiographic one.
The angiographic observation of already active compensatory circles can be inadequate, because others instead can be visible only after activation manoeuvres and not in static conditions.
I.e., in the occlusion of the internal carotid a. with a not inverted ophthalmic, the contra-lateral carotid can intervene trough a patent anterior communicant a., stressed by an increased velocity on the homo-lateral ophthalmic after compression of the homo-lateral common carotid artery.  (Fig. 4)
If compressing the contra-lateral common carotid a. the ophthalmic flow reduces and does not invert, then there is another endocranic source of flow. I.e., the homo-lateral posterior communicant a. can be patent and compensation comes from the homo-lateral vertebral, which has sufficient pressure to fight the contra-pressure of the external carotid, but not so much to win the pressure of the contra-lateral common carotid artery.
In this way a minor compensatory circle has been stressed, which is not visible if not after compression of the more active circle. This circumstance is a major protection factor than the existence of only one compensatory mechanism.
Visualising this condition can be trickily by angiography, while it is easy by C.W. Doppler.
 

1) Franceschi, C.: Exploration Doppler des voies de suppléance des artères cervico-cérébrales. Journal des Maladies Vasculaires. 1981,6, 89-100.
 2) Maroon, J.C., Campbell, R.L., Dyken, M.L.: Internal carotid artery occlusion diagnosed by Dopppler ultrasound. Stroke: 1,122-127,1970.
 3) Brokenbrough, E.C.: Screening for the prevention of stroke: use of a Doppler flowmeter. Regional Medical Program Brochure, Washington/Alaska.
 4) Franceschi, C.: L'indagine vascolare con ultrasonografia Doppler. Masson Italia, 1980, I. ed. francese del 1977.
 5) Sabbà, C., Portincasa, P., Palasciano, G., Albano, O.: Il Doppler ophtalmic test. Interesse e limiti. MIN. ANGIO., 8(2), 1983, 109-114.
 6) Baker, W.H., Hayes, A.C., Littoy, F.N.: Bilateral carotid stenosis: is Multiple-modality Noninvasive testing reliable ? In Dietrich, E.B., Non invasive Cardiovascular diagnosis, II ed., P.S.G., 1981, p.27.
 7) Gutierrez, I.Z., Makula, P.A.: The limited usefulness of the Doppler Supraorbital test and ophthalmic artery systolic pressure measurement in the diagnosis of internal carotid artery occlusive disease. in Dietrich, E.B., Non invasive Cardiovascular diagnosis, II ed., P.S.G., 1981, p.93.
 8) Testut, L., Latarjet, A.: Anatomia umana, volume II, Miologia-Angiologia, V ed., U.T.E.T. , Torino, 1971, pp. 740-743.
 9) Gillilan, L.A.: Potential collateral circulation to the human cerebral cortex. Neurology: 29,941,1974.
10) Youmans, J.R., Scarcella, G.: Extracranial collateral cerebral circulation. Neurology: 11,166,1961.
11) Kaneda, H., Irino, T., Watanabe, M., Kadota, E., Taneda, M.: Semiquantitative evaluation of ophtalmic collateral flow in carotid artery occlusion: ultrasonic Doppler study. J. Neurol. Neurosurg. Psych.: 1979, 42,1133-1140.
12) Franceschi, C.: Manuale delle compressioni per lo studio della circolazione cerebrale e periferica con metodica ultrasonografica Doppler. Fardis ed., 1981.
13) Franceschi., C., Franco, G., Luizy, F., Tanitte, M.: Precis d'Echotomographie Vasculaire. Vigot, Paris, 1986.
14) Prichard, D.R., Martin, T.R.P., Sherriff, S.B.: Assessment of directional Doppler ultrasound techniques in the diagnosis of carotid artery diseases. J. Neurol. Neurosurg. Psych.: 1979,42,563-568.
15) Barnes, R.W., Rittgers, S.E., Putney, W.: Screening for Carotid Artery Disease: Comparison of Plethysmographic and Doppler Techniques. Brochure.
16) Franceschi, C., Vadrot, M., Luizy, F., Jardin, M., Ribadeau-Dumas, C., Franco, G., Bouzac, H.: Echotomographic anatomy of the supra aortic arterial truncs. Ultrasons, 1981, 2, n. 1, 57-61.
17) International School of Ultrasonology. Neurosonology, including Doppler Ultrasound. Directors: E. Borgatti, C. Alvisi. "Ettore Majorana". Centre for Scientific Culture, Erice, March 1983.