Giuseppe Russo
Neurochirurgia, A.O. Cardarelli - Napoli
The natural history of the internal carotid artery (ICA) occlusion is
not still well defined and its treatment remains a controversial point.
Clinically it can present itself with signs of acute, transitory or
developing ischemia; but sometimes it can remain asymptomatic. The observation
of such cases is clearly increased during the last years, thanks to the
diffusion on large scale of ultrasound diagnostics.
Recent perspective studies, on whose base almost 50% of such patients
can present homolateral cerebral infarct within 3 years, have induced to
reconsideration of the assumption by which an asymptomatic ICA occlusion
is a stable and favourable condition.
For long time the treatment of ICA occlusions has been considered contraindicated
and is still topic of debate.
Such a controversial point reflects the difficulty of defining a population
which is susceptible of surgical treatment and available to practise perspective
studies to establish the clinical results in several treatment groups.
Anyway, the improvements of microsurgical, anaesthesiologic techniques
and of perioperative haemodynamic assessment have allowed revascularization
interventions in patients subgroups with ICA acute occlusion, with a low
% of neurological damages and absence of operative mortality (18).
These positive outcomes have moved the attention from the problem of
defining rigid protocols with exact timing limits and "therapeutic windows"
to a functional evaluation of cerebral haemodynamic compensations, suited
to each patient.
This holds also considering the impossibility to know certainly the
elapsed time between the cerebroafferent artery occlusion and the clinical
signs onset; because an arterial occlusion can be compensated for a long
time by efficient haemodynamic compensations and cause neurological deficits
only after their exhaustion.
Inclusion criteria of patients into the cited subgroup require the
lack of severe neurological deficits, of reduced consciousness and of intracerebral
haemorrhage. It is our opinion that a haemodynamic criterion could be added,
given by the presence of an appropriate cerebral vasomotor activity.
Hereafter we want to exhibit elements to sustain this our opinion,
whether taken from scientific literature or from our direct experience.
The haemodynamic and metabolic effect of the atherosclerotic occlusive
disease on cerebral parenchyma is not exclusively dependent on the stenosis
severity of arterial lumen, but also on the intervention of adequate collateral
paths, whose exam is necessary with the aim of a careful evaluation of
the "global effect" of the arterial stenosis on cerebral function.
Cerebral perfusion pressure (CPP) is the difference between the arterial
pressure and the intracranial pressure (ICP), which in physiological conditions
is equivalent to venous pressure.
Regional cerebral blood flow amount (rCBF) is directly proportional
to CPP and inversely proportional to cerebrovascular resistances: rCBF
= rCPP/rCVR. So that as in normal conditions rCPP is constant for a large
range of arterial pressure values, CBF change is to be related to changes
of vascular resistances.
During its evolution, vascular occlusive pathology can cause a CPP
decrease, which at the beginning is compensated by the dilatation of cerebrovascular
resistance compartment (arterioles and metaarterioles), demonstrated by
a blood volume increase.
Autoregulation capability exhausts when microcirculatory vasodilatation
capability is not able to compensate CPP reduction.
ICA stenosis is able to decrease CBF only when it reaches its "critical
value"; at this point, a little further vessel lumen reduction causes a
significant decrease of intravasal pressure and of blood flow.
Anyway, also in presence of significant ICA flow reduction, CBF in
most cases remains within normal limits, thanks to collateral circulation
activation. Boysen and others demonstrated that also in presence of ICA
stenosis which is able to reduce significantly blood flow registered downstream
to stenosis by an electromagnetic flowmeter, CBF values, measured by Xenon
133 endoarterial injection technique, can be within normal limits (4).
Collateral circulation intervention can be detected whether by cerebral
angiography as by Doppler Ultrasonography.
Angiography is still today the reference technique for the evaluation
of the anatomical configuration of Willis Circle; though it is anyway a
hard-working technique, which often requires several angiographic series
to achieve a better definition of the functional configuration.
Blood flow homogenous distribution between brain areas is guaranteed
by an efficient vascular "net", which is able to establish a communication
between main cerebral arteries: anterior, middle and posterior.
Willis Circle with the anterior and posterior communicant arteries,
extracranial carotid circle, with periorbital arteries and internal maxillary
branches as well cortical leptomeningeal anastomoses, are the anatomical
structures of what is commonly defined the cerebral "collateral circulation"
(CCC). Furthermore, the meningeal arteries net can also be recruited in
case of increased metabolic requirement, in specific cerebral areas.
CCC effectiveness is due whether to the integrity of normal morphology
as to the onset of changes of regional pressure values, which cause its
activation.
Really, CBF dynamics is linked to the onset of pressure gradient, by
which blood flow is directed towards zones where conditions of decreased
CPP have been caused.
Inflow pressure decrease at the level of conductance vascular compartment
caused by an arterial lumen stenosis as well as vascular bed increase caused
by a vasodilatation of resistance vascular compartment (arterioles and
metaarterioles) are the conditions which can cause CCC activation, regulating
at the same time its flow.
This means that in an auto-regulating vascular system blood flow amount
which passes through a vascular segment is regulated by the metabolic requirement
of the region which this vascular segment owns to.
Common clinical experiences of asymptomatic patients also with bilateral
carotid occlusive disease demonstrate the essential importance of the effectiveness
of such compensating circles.
Transcranial Doppler Ultrasonography (TCD), owing to its peculiarity
and mainly to low-cost, not invasivity and ready availability, is the ideal
technique for a clinical evaluation of haemodynamic compensations in case
of stenotic occlusive disease of afferent cerebral arteries. But what makes
this method an absolutely original one is the capability of measuring changes
of blood flow, registered in the main arteries of the cranial base, with
a time resolution range of the order of milliseconds.
TCD blood flow changes give immediate and indubitable informations
about the effectiveness of collateral circulation through the anterior
(ACoA) and posterior (PCoA) communicant arteries.
Flow direction inversion in the anterior cerebral artery (ACA) registered
in basal conditions is indubitably
a sign of ICA thrombosis.
In this condition ACA blood flow can be normally directed
(flow direction out of probe) only in the case of coexistence of aplasia
or severe hypoplasia of ACoA.
Anyway, in this case it is always possible to register a significant
asymmetry of the posterior cerebral arteries (PCA) blood flow velocity.
By a more careful exam of TCD result registered during a more prolonged
compression of common carotid artery, we can get further informations about
the intracranial compensation mechanism.
CCA flow stop causes an almost 50% immediate decrease of MCA systolic
velocity and an about 90% gradual preocclusive levels restoration, within
almost 10 sec.
During the time of the flow stop the systolic to diastolic ratio, measured
by pulsatility index, following Gosling (PI), is decreased by almost 60%,
compared to basal values (0.25 vs. 0.65).
Progressive and gradual restoration of velocity values to preocclusive
levels together with decreased PI values are expressions of the effectiveness
of the collateral circulation, activated by the compensatory vasodilatation
of resistance vascular compartment.
During the last ten years, using TCD ultrasonography we examined several
patients with different degrees of ICA stenosis, evaluating the haemodynamic
effect of arterial lumen stenosis on the cerebral blood flow registered
from the middle cerebral artery.
In presence of a > 90% stenosis Lindegaard and others registered a
Vmca reduction of almost 25% comparing to those registered on the healthy
side in the same subjects (8).
Instead, patients we examined, with ICA stenosis within 70% and 90%,
did not present statistically significant hemispheric asymmetries
of blood flow velocity values registered from MCA (Vmca).
In individual cases Vmca on the side of ICA occlusion resulted even
greater than the contralateral one (15).
These observations recall several experiences realised by other authors,
using different exam techniques.
In 1938 Mann and coll. demonstrated that the inflow pressure fall downstream
to a stenotic segment occurs only as a consequence of an essential decrease
of vessel lumen (9). Over a certain stenosis level, further small arterial
lumen decreases produce a sudden fall of the pressure and of the flow amount
downstream to it. The arterial lumen stenosis level over which this phenomenon
becomes significant has been defined "critical stenosis".
In a review on the topic, May and coll. showed that in several considered
studies the stenosis critical value ranged widely within 65% and 95% (10).
Later other studies confirmed Mann and coll. original observations,
underlining also the need of considering the flow amount which in basal
conditions passes through the artery (Fig. 1)
(12).
An almost 70% internal diameter reduction of arterial lumen in the
stenosis site, corresponding to an almost 91% area reduction of considered
artery section is generally defined "critical stenosis"; though in stress
conditions, when the increased cerebral metabolism requires a greater flow
amount, also a 50% stenosis can behave as a critical stenosis (2).
Xenon 133 CBF measurements performed on baboons demonstrated that in
basal conditions MCA stenosis till 90% are not sufficient to produce cerebral
ischemic damages; while cerebral ischemia signs on the side of the occluded
carotid were detectable when arterial hypertension joined to internal carotid
occlusion (11).
In normal conditions in a stenotic artery flow amount (Qp) depends
on pressure gradient between the arterial and the venous compartment (pa-pv),
minus the pressure fall which occurs inside the stenosis (Dps)
and it is inversely proportional to peripheral resistances Rp:
(pa - pv) - Dps
Qp = ----------------
Rp
Increasing stenosis degree Dps increases,
causing a Qp fall unless a simultaneous Rp reduction occurs.
When there is an efficient vasomotor reserve, this seems to occur by
the reduction of arteriolar and metaarteriolar tone (5).
In normal conditions the vascular bed reserve (VBR) is defined as the
resting flow (Qpr) to maximal flow amount (Qpmax) ratio and is proportional
to ratio between resting (Rpr) and minimal (Rpm) peripheral resistances.
Qpmax Rpr
VBR = --------- = ------
Qpr Rpm
In presence of stenosis, instead:
Rpr
Dps
VBR = ----- (1 - -------)
Rpm pa
- pv
When Rp reaches its minimal value (Rpm), corresponding to the maximal
dilatation of resistance vascular compartment, flow to peripheral districts
decreases accordingly to any further increase of vessel stenosis; in this
case the stenosis becomes a critical one.
Following Young, " the main effects of a stenosis on regional blood
flow are today sufficiently known, but it is still necessary to get a greater
number of in vivo measurements. Gathering these data is difficult, owing
to the great number of complex physiologic variables, which must be considered
(21).
Since our first TCD ultrasound analyses of cerebrovascular reserve,
we asked ourselves the following question: which connection exists between
blood flow velocity measure in cm/sec and the flow amount Q (ml/min) in
the middle cerebral artery?
The middle cerebral artery (MCA) is the natural site for TCD haemodynamic
evaluation in case of internal carotid artery stenosis:
I. MCA is in direct anatomical continuity with carotid siphon;
II. MCA, which gives almost 80% of blood flow to cerebral hemisphere,
can be considered a terminal artery (20) as the possibility of activating
collateral circulation to districts fed by it is almost exclusively due
to leptomeningeal arteries, which besides having a high flow resistance,
can intervene only as a consequence of the realisation of a significant
pressure gradient between the anterior and the posterior circles (1);
III. as MCA detection through the temporal bone window requires a very
acute (0 - 15) insonation angle and considering MCA M1 tract as sufficiently
anchored to cranial base, cosine(F) change in Doppler equation is less
than 3.5% (7).
As to quantitative evaluations, it seems correct to assume that MCA
diameter which is, inversely related to velocity, remains constant also
during tests which cause changes of several physiologic parameters (hyperventilation,
hyper-hypocapnia, hypotension, and
so on).
The error of this assumption seems acceptable, because whether in animal
experiments as in man systemic arterial pressure changes cause a less than
5% change of intracisternal arteries diameter; as pCO2 blood changes do
not seem to influence significantly the diameter of MCA main trunk.
Blood flow amount (Q) (ml/min) which passes through a given vessel
segment is given by flow velocity (V) multiplied by the diameter of vessel
lumen (D).
A low correlation was demonstrated by Bishop and coll., between the
absolute values of Xenon 133 cerebral blood flow (CBF) and TCD Vmca values
(r = 0.424, p < 0.01).
A good correlation between CBF and Vmca has been demonstrated by the
same authors, cited parameters being reported as % changes following a
hypercapnia test, compared to basal values (r = 0.849, p < 0.001) (3).
Also Lindegaard and coll. in a selected clinical model reported a wide
individual variation of ICA magnetic flowmeter Vmca to flow (Q) ratio.
The same parameters, Vmca and Qica, showed instead a linear correlation,
if normalised by the arithmetic mean of each subjects (8).
So that, taking into account experimental evidences and theoretical
assumptions, it seems acceptable to consider a strict correlation between
% Vmca changes in each subject, considered as control of himself, and %
flow amount (Q) changes in the same artery.
The cerebral vasomotor reserve can be so measured considering blood
flow changes following the "administration" to the subject under exam of
a physiologic stimulus, able to modify cerebrovascular resistances.
Increase of Carbon Dioxide (CO2) blood concentration is the most powerful
stimulus able to modify CBF, with an effect almost exclusively restricted
to peripheral vascular bed (16). Flow velocity to blood volume ratio in
cerebral arteries of cranial base remains constant, as pCO2 does not modify
intracistern arteries diameter. So that % changes of TCD velocity can be
considered as a direct expression of % changes of flow amount (Q) (5).
% pCO2 change to % CBF change ratio follows a sigmoidal curve, which
is linear for pCO2 values within 25 and 75 mmHg.
Vmca changes measured in basal conditions and following hypercapnia,
hypocapnia or acetazolamide infusion have been used to estimate cerebral
vasomotor reserve (12, 17). The measure, besides establishing in basal
conditions the position on the sigmoidal curve of the subject under exam,
allows us to estimate also the individual residual vasomotor capability.
Hyperventilation, reducing alveolar and arterial CO2 concentration,
causes a vessel tone increase at the level of the resistance vascular compartment,
by a H+ blood concentration decrease, This condition is detected by Doppler
effect, as a mean flow velocity decrease and a flow wave pulsatility (PI)
increase.
% V and PI changes before and following a hyperventilation test provide
a numerical index, defined "reactivity index" (RI), expression of the residual
capability of resistance compartment to change vasomotor tone.
In a group of patients with ICA stenosis > 70% we examined, differences
between V-RI mean value on the stenosis side
(3.5 ± 2.1) and on the healthy side (3.5 ± 1.0) were
not statistically significant (p = 0.8).
This difference was instead highly significant in a single case with
ICA complete occlusion
(V-RI = 4.1 vs. 0.8) (PI-RI = 6.0 vs. ­p;0.1) (15).
The difference between PI-RI on the stenosis side (3.5 ± 2)
and the one of healthy subjects (6 ± 4) (p = 0.04) was instead worth
of attention.
The most interesting data resulted from this study was that in case
of high degree ICA stenosis there is a dissociation between the index of
vasomotor reactivity to CO2, measured as velocity (V-RI) and the same index
measured as flow wave pulsatility (PI-RI).
Though pulsatility index (PI) is conditioned by the changes of several
physiologic parameters, the advantage of expressing vasomotor reactivity
index as PI-RI is that it is not necessary to assume a strict correlation
between V and CBF in the considered conditions.
Complex biochemical and ultra-structural mechanism influence collateral
circulation, by haemodynamic mechanism of auto-regulation and cerebrovascular
reactivity.
TCD has allowed us to measure the residual capability of resistance
vascular compartment to change its vasomotor tone as a response to changes
of cerebral metabolic requirements.
A further forward step, after our first measurements here outlined,
was linked to the possibility of a simultaneous recording of blood flow
from both MCA together with end-tidal CO2 and to off line analysis and
correlation of obtained results (Fig. 2).
Patients with monolateral ICA occlusion were examined at rest, during
spontaneous hyperventilation (hypocapnia test) and rebreathing expirated
air (hypercapnia test) (17).
Analysing registered data, global change of blood flow from hypocapnia
(maximal vasoconstriction) to hypercapnia (maximal vasodilatation), was
considered and defined "full range" reactivity.
Data were compared to those of a group of healthy subjects.
The results show a perfect hemispherical symmetry of the vasomotor
reactivity index to CO2 (RI) measured in healthy subjects, whether by hypocapnia
(RI = 3.3 vs. 3.3) as by hypercapnia test (RI = 3.6 vs. 3.5), as well as
of full range RI (RI = 3.5 vs. 3.4).
RI numerical value denotes % flow velocity change (ml/sec) for each
pCO2 unit (mmHg) change.
The same indexes are clearly lower on the same side of occluded ICA:
hypocapnia RI = 2.3; hypercapnia 1.1, full range 1.4 (Fig.
3).
Such values show a clear reduction and in some cases an exhaustion
of vasomotor capability of resistance cerebrovascular compartment, already
maximally dilated to compensate for CPP fall, produced by inflow pressure
decrease.
Confidence intervals analysis of each reactivity index shows that using
rebreathing hypercapnia test as vasodilator stimulus has a higher capability
to separate disease from normality (difference between means = 1.83, 99%
confidence interval between 1.33 and 2.33; p < 0.0001).
But it is the analysis of each case to give the most significant reflection
and discussion points.
While in some patients ICA occlusion causes the exhaustion of the microcirculatory
capability to change its vasomotor tone
(RI = 0.00), in others such capability can be a little bit less than
that of healthy subjects
(RI = 2.12 vs. 2.16).
Furthermore, while in healthy subjects the "best" vasomotor response
can in individual cases be detected during hypo or hypercapnia test (greater
vasoconstriction or vasodilatation capability), in ICA thrombosis patients
a constant greater alteration of vasodilator capability is evident (hypercapnia
test), to witness the presence of an already reached maximal compensatory
dilatation.
These last outlined data underline the need of an haemodynamic evalutation
of cerebral blood flow compensations, suiting it to each patient, as a
pre-requisite to any treatment of arterial cerebroafferent occlusive lesions.
With this aim we wanted to select an easy feasible test in clinical
practise, able to provide the least as possible discomfort to the patient.
From our data, TCD reactivity test to CO2 seems to satisfy these requirements.
1. Aaslid R: Transcranial Doppler Sonography. Springer Verlag, 1986.
2. Beurger FW: Critical arterial stenosis: a theoretical and
experimental solution. Ann Surg 180: 33-41, 1974.
3. Bishop CCR et al.: Transcranial Doppler measurement of middle
cerebral artery blood flow velocity: a validation study. Stroke 17, 5:
913-15, 1986.
4. Boysen G et al.: The effect of induced hypertension on internal
carotid artery pressure and regional cerebral blood flow during temporary
carotid clamping for endarterectomy. Neurol 22: 1133-44, 1972.
5. Ferguson GG: Extracranial carotid artery surgery. Clin Neurosurg
29: 543-74, 1982.
6. Kirkham FJ et al.: Transcranial measurement of blood velocities
in the basal cerebral arteries using pulsed Doppler ultrasound: velocity
as an index of flow. Ultrasound Med Biol 12: 15-21, 1986.
7. Lindegaard K-F et al.: Precerebral hemodinamics in brain tamponade.
Part 1: clinical studies in blood flow velocity. Neurochirurgia (Stuttgart)
23: 132-142, 1980.
8. Lindegaard K-F et al.: Assessment of intracranial hemodynamics
in carotid artery disease by transcranial Doppler ultrasound. J Neurosurg
63: 890-98, 1985.
9. Mann FC et al.: The effect on blood flow of decreasing the
lumen of a blood vessel. Surgery 4: 249-51, 1938.
10. May AG et al.: Hemodynamic effects of arterial stenosis. Surgery
53: 513-15, 1963.
11. Mendelow AD: The hemodynamic effect of internal carotid artery
stenosis and occlusion. J Neurosurg 66: 755-63, 1987.
12. Moore WS: Effect of flow rate and vessel calibre on critical arterial
stenosis. J Surg Res 26: 1-4, 1979.
13. Ringelstein EB: A practical guide to transcranial Doppler sonography.
In: Noninvasive imaging of cerebrovascular disease. Alan R. Liss, Inc.
74-121, 1989.
14. Russo G et al.: L'esame Doppler transcranico. Ricerca Neurochir.,
Supp. 1, 1988.
15. Russo G et al.: Valutazione emodinamica delle stenosi della arteria
carotide interna. Il ruolo della ultrasonografia Doppler transcranica.
N.P.S. XI, 3: 283-322, 1991.
16. Russo G et al.: Non invasive recording of CO2 cerebrovascular reactivity
in normal subjects and patients with unilateral internal carotid artery
stenosis.
17. Russo G et al.: A bed side test for cerebrovascular CO2 reactivity
measurement. Re-breathing as vasodilator stimulus (in press).
18. Sorteberg W et al.: Blood velocity and regional blood flow in defined
cerebral artery systems. Acta Neurochir 97: 47-52, 1989.
19. Spetzler RF et al.: management of total carotid occlusion. In:
Microsurgical carotid endarterectomy. Bailer JE, Spetzler RF ed. Lippincott
- Raven Pbs. 141-162, 1996.
20. Symon L: Pathological regulation in cerebral ischemia. In Wood
JH ed: Cerebral blood flow. Physiologic and clinical aspects. New York,
Mc Graw-Hill Book Co, 423-24, 1987.
21. Toole JF: Cerebrovascular disorders. 3rd ed. 1-18, 1984.
22. Young DF: Fluid mechanics of arterial stenosis. J Biomed Enginee
101: 157-75, 1979.