Am J Electroneurodiagnostic Technol

49:244-259. 2009
© ASET, Missouri

Transcranial Doppler Series Part III: Interpretation

Heather A. Nicoletto, BS, RVTand

Marilyn H. Burkman, BS, RDCS, RVT

Neurodiagnostic Laboratory
Duke Ur)iversity Hospital
Durham, North Carolina

ABSTRACT. Transcranial Doppler (TCD) is a quick, non-invasive,

and inexpensive way to evaluaie the blood flow in ihe basal cerebral
arteries. TCD is most commonly used to evaluate for vasospasm in
patients with subaraclmoid hemorrhage and for vasculopathy in chil-
dren with sickle cell anemia. It may also be used to evaluate intracranial
vessel narrowing and occlusion: collateral flow due to extracranial
vessel occlusion, arteriovenous malformations, and cavernous carotid
fistulas: and to help confirm cerebral circulatory arrest. Systolic
upstroke, velocity, pulsatility index, and direction of flow aid in the
interpretation of TCD.

KEYWORDS. Cerebral circulatory arrest, sickle cell anemia,

transcranial Doppler, vasospasm, waveform morphology.

INTRODUCTION
Transcranial Doppler (TCD) is gaining popularity among physicians because it
is a quick, inexpensive, and noninvasive way to evaluate the blood flow in the
basal cerebral arteries. The uses for TCD vary widely. The most common uses are
to evaluate for vasospasm in patients who have had a subarachnoid hemorrhage and
10 evaluate for vasculopathy in children with sickle cell anemia. Other less common
uses are to look for vessel narrowing or occlusion in patients wht) present with stroke
and to evaluate for cerebral circulatory arrest. Specifically, the TCD waveform
is analyzed to assess the hemodynamics of each vessel in the head and neck. The
physician can make an interpretation by evaluating the velocity of the blood How. the
shape of the waveform, the direction of flow, and the resistance in each vessel.

Received; February 20, 2009. Accepted for publication: March 11, 2009.

244
 TCD INTERPRETA TION 245

WAVEFORM ANALYSIS
There are many characteristics within the TCD waveform that are used to evaluate
cerebral hemodynamics. Waveform morphology, velocity, pulsatility indices, and
direction of now are analyzed. Turbulence and bruits are heard as well as visually
analyzed. All of these characteristics of the TCD waveform will help the physician
make an interpretation of the study.

Waveform Morphology
The shape of the TCD waveform is called waveform morphology. Each waveform
has systole and diastole, with systole being the peak of the waveform and diastole
being the deceleration of the waveform. In evaluating waveform morphology, the
physician will look at the systolic upstroke. A nonnal TCD waveform in systole
has a quick, sharp upstroke that accelerates to a peak (Zweibel 1992) (Eigure lA).
This upstroke can be altered for various reasons. With a proximal obstruction, the
upstroke becomes slow and the result is a slanted or rounded waveform. This is
referred to as a slowed systolic upstroke or a damped waveform (Figure IB). This
slowed systolic upstroke may also be caused by cardiac function such as myocardiai
dysfunction or aortic valve stenosis. It is important to compare the upstroke in ail
arteries. If the systolic upstroke is slowed due to a proximal obstruction, you will
only seethe slowing in the branches of that artery. If cardiac function is the cause of
the slowed systolic upstroke you will see the slowing in every artery (Carter 1992).

Velocity
Velocity is one of the most crucial components the physician will use to make an
interpretation. Transcranial Doppler uses time average mean flow velocity (MFV)
calculated as:
MFV = (systolic velocity-diastolic velocity/3)+ diastolic velocity
(Katz and Alexandrov 2003).
Each vessel has a normal mean flow velocity. The middle cerebral artery (MCA)
and anterior cerebral arteiy (ACA) have the highest velocities and the posterior
cerebral artery (PCA) and basilar artery (BA) have lower velocities (Table I). In
children, blood flow velocity is relatively high. With aging, blood fiow velocities
decrease (Table 2). When performing a TCD. it is important to know what the normal
velocities for each vessel should be ba.sed on the age of the patient. Velocities that
fall out of this normal range may be an indicator of disease.
When ilow velocities are elevated, it suggests that there is a vessel narrowing
within the site being insonated. This is based on the hemodynamic principle that
246 TCD INTERPRETATION

cm/s • < - 4p
150 Sharp Syslallc l'pitrokri

100
â4 J
50 LLLLlLLi
50
100 •
0 1 2 3 4 5 6 7 t) 9 10s

Date 7/14/2008 1:10:51 PM MeanAp

Probe 2PW Max 93/0
Sample Volumefmm] 6 Mean 67/0
Gainl%] 31 Min 49/0
ISPTA 721» Pl 0.65/0.00
Depth [mm] 30 O/S
ScalefHz] 7519
Label MCA_R

B
IMC4 1 y

1 O -40
10:24:01 AM -fiO
Depth 45
Power 100 -80
: . ^ ^ ^ _
Sample 6
Slowfü .SVülolir lipsnokíS
-1UU
Peak 42
0
Dias 21
Mean 29 -100

FIG. 1A. Transcranial Doppler waveform demonstrating a normal systolic upstroke. The
black lines along the upstroke show a quick, sharp upstroke that accelerates to a peak.
FIG. IB. Transcranial Doppler waveform seen distal to a vessel narrowing or occlusion.
The black lines along the upstroke are slanted to show the delayed systolic peak.
 TCD INTERPRETATION 247

Table 1. Normal mean flow velocities for adults.

Artery Mean Vel<Kiiy

ECICA 30 +/- 9 cm/sec
MCA 55 +/- I2cm/sec
ACA 50 +/- II cm/sec
PC A-PI 39 +/- 10 cm/sec
PCA-P2 40 +/- iücn-i/sec
OA 21 +/- 5 cm/sec
Supraclinuid ICA 41 +/- 11 cm/sec
I'arusellar ICA 47 +/— 14 cm/sec
VA 38 +/- 10 cm/sec
BA 41 +/- 10 cm/sec
ECICA - extracranial internal carotid artery: MCA - middle cerebral artery: ACA - anterior
cerebral artery: PCA - posterior cerebral artery. OA - ophthalmic artery: ICA - internal carotid
artery; VA - vertebral artery; BA - basilar artery

Tabit 2. Normal mean flow velocities (cm/sec) from birth to age IS years.

Middle Internal Anterior Posterior Cerebral Artery

Cerebral Carotid Cerebral Basilar
Age Artery .'\rterv Artery Artery
Oto 10 days 24 + / - 7 25 + / - 6 ¡9+/-6 - - -
11 to 90 days 42 + / - 10 43 + / - 12 33 +/-11 - -
3 to 11.0 months 74+/-14 6 7 + / - 10 50 +/- 11 - - -
1 to 2.9 years 85 + / - 10 81+/-8 55+/-13 50+/- 17 50+/-12 51 +/- 6
3 to 5.9 years 9 4 + / - 10 93 + / - 9 7! +/-15 56+/-13 48 +/- 11 58 +/- 6
6 to 9.9 years 97 + / - 9 93 + / - 9 65+/-13 57 +/- 9 51+/-9 58 +/- 9
10 to 18 years 81 + / - 11 7 9 + / - 12 56+/- 14 50+/- 10 45 +/- 9 46 +/- 8

the velocity of blood flow is inversely related to the area of the vessel lumen (Aaslid
et al. 1984). Wheti flow velocities decrease, it suggests a proximal narrowing or
occlusion because there is a smaller amount of blood flow coming from the
narrowed/occluded area.

Pulsatility Indices
Pulsatility indices (PI) are a comparison of systolic flow to diastolic flow. In
other words, vessels that feed low resistance beds (the brain and organs) have a
sharp systolic upstroke with continuous forward flow throughout diastole, thus a
low pulsatility index (Figure 2A}. A high pulsatility index will be found in vessels
that feed high resistance beds (the extremities and face). They have a sharper
systolic upstroke, a narrower peak in systole, and less flow or reversal of flow in
diastole (Figure 2B ). A normal PI for the basal cerebral arteries is 0.5 to 1,19 (Katz
and Alexandrov 2003). PI is calculated as:
248 TCD INTERPRETATION

Date 7/14/2008 1:29:13 PM MeanAp

Probe 2PW Max 103/33
Sample Vokjmejmm] 8 Mean 74/16
Gaînt%] 19 Min 55/0
ISPTA PI 0.65/2.11
50 D/S
ScalefHz] 7519
Label MCA L

1441

30
60
J
0
0
1
1
-30

-«0 -«0
1I / ' '

45 0 5.6<Vb 8 154 2 44 0 40OA, 8 231 2

PW PW
Depth Gain Power Sample Scale Probe Depth Gain Power Sample Scale Probe

FIG. 2A. Transcranial Doppler waveform from a low resistance vessel. There is continu-
ous forward flow throughout both systole and diastole.
FIG. 2B, Transcranjal Doppler waveform from a high resistance vessel. There is a sharp
systolic upstroke, a narrow peak in systole, and less flow in diastole.
FIG. 2C. Transcranial Doppler waveform from the jugular vein. Fiow in veins is continu-
ous through the entire cardiac cycle with variations in velocity with respiration.

PI = systolic velocity —dia.stolic velocity/mean flow velocity

(Katz and Alexandrov 2003).
PI lower than 0.5 may suggest a proximal narrowing/occlusion or an arteriovenous
malformation.
A proximal narrowing or occlusion causes the arterioles to dilate in response to
lack of oxygen. As the arterioles dilate, the vascular resistance decreases allowing
 TCD INTERPRETATION 249

more diastolic flow. As systolic now remains the same and diastolic flow increases
the PI decreases (Babikian 1993).
Arteriovenous malformations create a decrease in PI because there is an abnormal
connection between the arteries and the veins. Flow in veins is continuous through
the entire cardiac cycle with variations in velocity with respiration (Figure 2C).
Because of this continuous venous flow the flow in the artery becomes less resistant
causing a lower PI.
Pis that are greater than 1.19 may be caused by a distal occlusion. When blood
tlow hits the occlusion it causes flow to stop abruptly. The result is a decrea.se in
diastolic tlow or a reversal of flow in early diastole and little or no How in late
diastole.
Arterial constriction will also cause Pis to elevate. High intracranial pressure
increases resistance in the brain causing a decrease in diastolic tlow. As the diastolic
tlow decreases the PI increases. Age can cause the vessels to lose compliance, or
stiffen, so people over the age of 65 may have high Pis.
Cardiac function plays a role in PI. High Pis may be caused by severe aortic régur-
gitation. In severe aortic régurgitation, blood leaks backwards into the heart during
diastole. This can cause a decrease or loss of diastolic flow in the brain.

Direction of Flow
Antegrade tlow is blood that is flowing in the normal direction. When tlow
reverses from its normal direction it is called retrograde tlow. Flow direction may
reverse when a vessel is acting as a collateral (.sending blood to areas of the brain it
normally doesn't supply). Collateral How occurs when there is a proximal narrowing
or occlusion.
Il is also possible that blood tlow may only partially reverse. A hesitant waveform
is when forward flow brietly stops (partial reversal of flow) during systole
(Figure 3A). An alternating waveform is when the flow reverses during systole but
is antegrade during diastole. It may also be seen as forward flow during systole with
a brief period of tlow reversal during early diastole and forward flow in late diastole
(Figure 3B). Partial reversal of flow may occur during the development of collateral
flow.

Turbulence and Bruits

Normally, flow in the arteries is laminar which means the tlow moves in an
orderly fashion parallel to the vessel walls with the slowest flow at the vessel
walls and the fastest How in the center of the vessel (Zwiebel 1992). When ihere
is narrowing in the artery, the flow becomes disorganized or disturbed. When
insonating within a narrowed segment of vessel, disturbed flow is visualized as a
250 TCD INTERPRETATION

cnVs - 40
too
Ü
50
1
1
n

L mm m
100
r
ut Flow Rtïfi-ïHl

0 1 2 3 4 5 s

Date 5/19/2008 4:01:28 PM MeanAp

Probe 2PW Max 21/73
Sample Vo(ume[mm] 10 Mean 2/51
Gain[%] 31 Min 0/34
ISPTA 720 PI 10/0.8
Depth [mm] 85 D/S
ScalelHz] 6536
Label BA

B
1 . 1 . 1 .

-40
04:38:14 PM II pi fî |j -60
Depth 51
Power 20 -80
1 . 1 . 1 .
Sample 3 Flow Rr^(>-^aI

0
1 f Fotward now f 1
-60
Aultfli'ailt ilav
•100
I • 1 • 1 •

FIG. 3A. Transcranial Doppler waveform demonstrating hesitant flow. Latent flow
reversal is when forward flow briefly stops (partial reversal of flow) during systole.
FIG. 3B. Transcranial Doppler waveform demonstrating alternating flow. There is ante-
grade flow during systole with a brief period of flow reversal during early diastole
and forward flow in late diastole.
 TCD INTERPRETATION 251

LMCA
2 CM 40
09:54:12 AM -60
36
Depth
Power 100
Sample
1
, t,.
""" ,1 -80

200
Peak 2421
Dias 127*
kail -100
Mean 173* 0
1 -^c^—a^.. JÜ: x^ ^: "^

FIG. 4. Transcranial Doppler waveform within a vessel narrowing. Disturbed flow is seen
as bright pixels along the baseline.

bright group of pixels along the baseline (Figure 4). This disturbed flow can also be
heard and is called a bruit. A bruit can sound loud, gruff, or even like a seagull {called
a musical inurniur).

TCD INTERPRETATION
The introduction of TCD made it possible to record intracranial blood flow
velocities. TCD has become an important, noninvastve method for assessing cere-
brovascuiar hemodynamics and for evaluating intracranial cerebrovascular disease.
It can measure relative changes in cerebral blood tlt)w objectively, immediately, and
as often as desired (Otis and Ringelstein 1992). The established applications for
TCD are:
• monitoring for vasospavm in the presence of subarachnoid hemorrhage.
• detection of intracranial stenosis or occlusion in the major basal cerebral
arteries,
• evaluation of collateral flow due to extracranial occlusive disease.
• detection of arteriovenous malformations and cavernous carotid fistulas, and
• evaluation of cerebral circulatory arrest.

Subarachnoid Hemorrhage and Vasospasm

Subanichnoid hemorrhage (SAH) is caused by leakage of blood from a weakened
artery wall or a ruptured aneurysm. Delayed vasoconstriction (vasospasm) of the
252 TCD INTERPRETATION

basal cerebral arteries is a common occurrence following subarachnoid hemorrhage.

Vasospasm generally occurs 3 to 14 days after subarachnoid hemorrhage with
maximal severity between days 7 and 12 (Sloan 1993). The exact mechanism of
vasospasm is not known. However, experimental and clinical studies have shown
that the amount and duration of vessel exposure to blood in the basal cisterns and
subarachnoid space determine the occurrence of vasospasm (Sloan 1993). Vasospasm
can occur in any basal cerebral vessel regardless of where the hemorrhage is.
The evaluation of vasospasm is one of the most common uses of TCD. There
are many published TCD criteria for vasospasm but the one most frequently used
was written by Aaslid et al. (1984). On TCD, vasospasm is seen as an increase in
mean flow velocities. However, an increase in velocity alone is not enough to make
the interpreiation of vasospasm. Patients with SAH are often treated with hypervol-
emic-hypertensive-hemodilution (HHH) therapy. HHH therapy involves maintaining
an elevated blood pressure, a low hematocrit, and keeping the patient well hydrated.
One reason HHH therapy is used is to attempt to prevent vasospasm.
HHH therapy alone can cause flow velocities to increase. This is known as
hyperdynamic flow. Because of this, the Lindegaard ratio was developed (Linde-
gaard et al. 1986). This ratio is calculated using the highest middle cerebral
artery (MCA) velocity divided by the ipsilateral extracranial internal carotid artery
(ECICA) velocity. The Lindegaard ratio helps determine whether the increase in
MCA velocity is due to vasospasm or due to hyperdynamic flow. The Lindegaard
ratio also works for the carotid siphon by dividing the siphon velocity by the ipsilat-
eral ECICA velocity. A modified Lindegaard ratio determines whether increases in
basilar artery (BA) flow velocities are due to vasospastn or hyperdynamic flow. This
is calculated by dividing the highest BA velocity by the average of the right and left
extracranial vertebral artery (ECVA) (Sviri et al. 2006).
Vasospasm in the middle cerebral arteries and internal carotid arteries can be
graded as mild, moderate, or severe. Many studies have demonstrated that a mean
velocily of 120 to 149 cm/sec can be consideted mild, 150 to 199 cm/sec can be
considered moderate, and greater than 200 cni/sec can be considered severe. The
Lindegaard ratio must be used as well and a ratio of 3.00 to 5.99 is considered mild
to modetate vasospasm. A ratio of 6.00 or greater is considered severe vasospasm
(Aaslid et al. 1984. Lindegaard et al. 1986).
Velocities in the anterior cerebral artery (ACA) and posterior cerebral artery (PCA)
will increase due to vasospasm, but they can also increase when they are acting
as collaterals. Collateral flow can occur when the MCA or internal carotid artery
(ICA) is in moderate or severe vasospasm. If the ACA velocity reaches 130 cm/sec
or greater without the presence of moderate or severe ipsilateral MCA or ICA
vasospastn, the ACA can be considered in vasospasm. However, if the ACA velocity
reaches 130 cm/sec and the ipsilateral MCA or ICA is in moderate or severe
vasospasm the interpretation of vasospastn vs. collateral flow should be made. The
 TCD INTERPRETATION 253

same idea applies with the PCA. However, the velocity criteria used is 110 cm/sec
(Sloan 1993).
The interpretation of vasospasm in the vertebral arteries can be made at a velocity
of 85cmysec. In the BA, an interpretationof'may represent vasospasm" can be made
at a velocity of 70 cm/sec with a BA/ECVA ratio of 2.0 to 2.49. Moderate vasospasm
in the BA is interpreted with a velocity of 85 cm/sec and a ratio of 2.5 to 2.99. Severe
vasospasm of the BA is interpreted with a velocity of 85 cm/sec and a ratio of 3.0 or
greater (Sviri et al. 2006).
There are many different published criteria for vasospasm. At Duke University
Hospital, we use the criteria published in Contemporary Neuro.'iur^ery (Newell
and Winn 1989). No matter which criteria are used, each lab should make sure
to internally validate the results with angiography or computed tomography angiog-
raphy and adjust the criteria as necessary to most accurately correlate with those gold
standards.
I

Sickle Cell Anemia

Another common use of TCD is evaluation for vasculopathy. or narrowing of
the basal cerebral arteries, in children with sickle cell anemia. Stroke occurs by
age 20 years in about 11 % of palicnis with sickle cell anemia (Adams et al. 1998a).
The disease is caused by abnormal hemoglobin which binds with other abnormal
hemoglobin molecules within the red blood cell to cause rigid deformation of the
cell. This can cause sludging and congestion of the arteries which then leads to stroke
(Lonergan et al. 2001). The TCD evaluation for narrowed vessels involves looking
for elevated flow velocities along with possible bruits within the basal cerebral
arteries. The most widely used criteria are velocities of 170 to 199 cm/sec, which
is consistent with conditional sickle cell vasculopathy. A velocity greater than
200 cm/sec is consistent with sickle cell vasculopathy (Adams et al. 1998a. Adams
et al. 1998b). These criteria are now being used for all basal cerebral arteries (Adams
etal. 1998b).

Intracranial Vessel Narrowing

Intracranial atherosclerosis is responsible for up to 10% of transient ischémie
attacks (TIA) and stroke (Sloan et al. 2004). TCD can reliably detect stenosis of
the carotid siphon, MCA, vertebral artery (VA), PCA, and BA. If there is a vessel
narrowing within any of these vessels, there will be a focal elevation in velocity.
Focal means that the elevation will be within a 5 to 10 mm segment and all velocities
at depths around the focal narrowing will be low to normal. Besides focal elevations
in velocity, there are other criteria. Compare flow velocities from one side to the
other. There will be a side to side difference in velocities of greater than 30 cm/sec.
254 TCD INTERPRETATION

If the narrowing is signiflcant. downstream effects such as turbulence or a slowed

systolic upstroke may be seen (Figure IB). A bruit may also be seen and heard
within the focal narrowing (Figure 4). ,

Intracranial Vessel Occlusion

The detection of intracranial vessel occlusion can be reliable if specific criteria are
followed. If the vessel is occluded, there will be absence of arterial flow signals at
normal depths. However, the technologist must be able to easily identify vessels that
are also found within this acoustic window, because absence of flow signal could
also indicate a poor acoustic window. For instance, if the MCA is occluded you
should be able to easily identify the ACA and PCA. thus showing that the acoustic
window is satisfactory. Elevated velocities may be seen in the ipsilateral ACA or
PCA. This is because the ACA and the PCA are acting as collaterals to the area of
the brain normally fed by the occluded MCA. There may be upstream effects such
as an increase in pulsatility indices. This happens because flow heading toward the
occluded area is bumping up against the occluded vessel.

Extracranial Occlusive Disease and Collateral Flow

The term hemodynamically significant is used to describe a narrowing of 80%
or greater (including vessel occlusion) which affects the blood flow distally (Rosen-
kranz et al. 1991). On TCD, a change in MCA waveform morphology indicates there
is a hemodynamically significant narrowing or occlusion of the ipsilateral ICA. The
MCA waveform will have a slowed systolic upstroke, and may have lower than
normal flow velocities and pulsatility indices (Figure IB). Also with a hemodynami-
cally significant ICA narrowing comes the presence of intracranial collateral flow.
Three vessels, the anterior communicating artery, the posterior communicating
artery, and the ophthalmic artery, can act as collaterals. One, two, or all three
vessels may act as a collateral at one time so it is beneficial to evaluate for all three
collaterals.
Crossover collateral occurs when the anterior communicating artery moves blood
from the right to the left, or vice versa. On TCD, the ACA on the side contralateral to
the ICA narrowing will have increased velocities. Often times the ACA velocity will
be higher than the MCA velocities on the same side. The ACA on the side ipsilateral
to the ICA narrowing will now be reversed (flowing toward the probe). It may also
have elevated flow velocities (Figure 5A).
Posterior to anterior collateral occurs when the posterior communicating artery
moves blood from the posterior circulation to the anterior circulation. On TCD,
the PCA-Pl segment velocities on the side ipsilateral to the ICA narrowing will
be elevated (Figure 5B). This elevation will be higher than the ipsilateral MCA
velocities by at least 125% (Rosenkranz et al. 1991).
 TCD INTERPRETATION 255

Lie« Ai^_L DM* 1,nia 08 9:30:37 AM

'3Mn«%) 56 O»ptf 66
Sanpli Vokna (mit 10 nota 2 PW
s
8 l^w^ï (piM): 100-3 Scan 10000
8 EPTA 390

8 090 / 1.00
•*) 123./330
056
O Itei 1S&/330 n / a»
240 2.82
•tan 192/60.0 so /
t DS 0.42 / 036
1
Elevated velocity

B C
L-OA

RPCA-P1 Maan
1 COA-
08:27:43 AM M
Depth 57
ki
1
Povwer 50
Sampte 6
Peak 2 1 r -«0
Dias 81 '
Mean 133'
50 0 14% 8 231 2
1 Depth Gain Power Sample Scale Probe

D
L-OA SO •«»r^ 2

Mean
\
1 Pi
FIG. 5A. Transcranial Doppler waveform of an anterior cerebral artery (ACA) acting
as crossover collateral. Flow in the ACA ipsilateral to a proximal occlusion is reversed
(flowing toward the probe). There is also an increase in flow velocity.
FIG. 5B. Transcranial Doppler waveform of a posterior cerebral artery (PCA) acting as
posterior to anterior collateral. Velocities in the PCA ipsilateral to a proximal occlusion are
increased.
FIG. 5C. Transcranial Doppler waveform of an ophthalmic artery (OAt with high
resistance flow directed toward the probe.
FIG. 5D. Transcranial Doppler waveform of an ophthalmic artery acting as an external
to internal collateral. The waveform is low resistance with flow away from the probe
(retrograde flow).
256 TCD INTERPRETA TION

External to internal collateral occurs when the ophthalmic artery (OA) supplies
blood from the external carotid iuiery to the carotid siphon. This is seen on TCD
as reversal of now in the OA. Remember, the OA normally has high resistance
waveform morphology because it is feeding the eye (Figure 5C). When it changes
directit)n it is now feeding the brain and the waveform morphology will become low
resistance (Figure 5D). External to internal collateral is the last collateral to occur. It
will only occur if the brain is not getting enough oxygenated blood from the other
two collateral sources. If this collateral is seen, it is very important to inform the
physician since this patient is at greater risk of having an ischémie stroke.

Vertebrohasilar Insufficiency
Vertebrobasilar insufficiency occurs when there is not enough oxygenated blood
flow to the cerebellum. This can occur due to basilar artery occlusion or narrowing,
but more often it is due to a subclavian steal. When the subclavian artery is narrowed
or occluded at the origin, it causes a pressure difference that changes the direction of
flow in the ipsilateral vertebral artery. Instead of the vertebral artery flowing toward
the brain, it reverses direction and supplies blood flow to the distal subclavian artery.
Under resting conditions this may not affect flow in the ba.silar artery. However,
when the arm is being exercised, the oxygen demand pulls blood to the arm via the
subciavian artery, and the basilar artery flow may start to change direction. This can
be seen as hesitant, alternating, or completely reversed flow (Figures 3A and 3B).
Hesitant or alternating flow may also be seen in the vertebral artery under
other circumstances. If there is a proximal vertebral artery narrowing or occlusion,
alternating or hesitant flow may be seen. This is because the blood flowing antegrade
in the contralateral vertebral artery can backfill the distal portion of the compromised
vertebral artery at the confluence of the basilar artery.

Arteriovenous Malformations and Cavernoas Carotid Fistuias

An arteriovenous malformation (AVM) is a collection of abnormal vessels in
which the arterial circulation flows directly into the venous circulation without inter-
vening capillaries (Giller 1993). Because of the venous connection, the arteries that
feed an AVM will have low pulsatility indices on TCD. These indices will often be
lower than 0.5. A side to side comparison of the Pis in the artery should be done to
see if there is a significant difference. The flow velocities in the feeding arteries will
be higher than normal because of the large volume of blood being carried to the AVM
(Figure 6).
Cavernous carotid fistulas are an abnormal connection of the cavernous internal
carotid artery (carotid siphon) and the cavernous sinus. This can be a congenital
condition, but is more often caused by trauma to the head. The TCD appearance is
 TCD INTERPRETATION 257

Label PCAF2_L Ùtm

Gwi(%) 40 OMhlnvrq 67
Swnpi« Vohme <irai4 B (Vote 2 fW
R M ar irrtM) 54 05 Seal* (Ht) 9600
BPTA 210

*, 200/106 " '« ' 0«^LowPI

M» 890/171 " °™ ' °»
P*^ 540/IM SO 4« , 163
^ 06 022 / 0.61
Elevated velocity
FIG. 6. Transcranial Doppter waveform of a posterior cerebral artery (PCA) that is
feeding an artariovenous malformation (AVM). There is an increase in flow velocity and
a pulsatility index (PI) of less than 0.5.

virtually identical to that of an AVM. However, the high flow velocities and low Pis
will be seen only in the cavernous internal carotid artery.

Cerebral Circulatory Arrest

Angiography and EEG remain (he mosl widely used studies to evaluate for the
irreversible or complete loss of brain functions. However. TCD has value when
sedative drugs render EEG unreliable or when there is risk of harm to potential donor
organs from angiographie dyes (Ducrocq et al. 1998). The development of cerebral
circulatory arrest is identified by changes in TCD waveform morphology. Oscillating
flow is when forward flow in systole is nearly equal to reversal of flow in diastole
(Figure 7A). Systolic spikes are a very short velocity peak seen only in systole
(Figure 7B). Both of these signals arc consistent with cerebral circulatory arrest
on TCD. No identifiable flow is the final stage of cerebral circulatory arrest. The lack
of identifiable flow may be due to inadequate windows, .so it is imperative to have a
recent TCD study thai sht)ws adequate windows with flow.
There are guidelines that every lab should follow when using TCD as a con-
firmatory test for cerebral circulatory arrest. Other conditions, such as intoxication,
hypothermia, severe arterial hypotension, and metabolic disorders, can cause an
increase in cerebral resistance and must be excluded. Make sure the patient has not
had a decompressive craniectomy which can interfere with the development of the
intracraniiil pressure. The TCD must demonstrate oscillating flow or systolic spikes
in the basilar artery, bilateral MCAs, and bilateral ICAs on two examinations at an
interval of at least 30 minutes apart. If all of these criteria are met the diagnosis of
cerebral circulatory arrest may be confirmed (Ducrocq et al. 1998).

CONCLUSION
The analysis of the TCD waveform makes it possible to assess the basal cerebral
arteries for various disease processes. Changes in flow velocity can lead to an
258 TCD INTERPRETATION

I I [ r 11 I. i I
2PW
Ma« 11/5
S
ISPT* ?zo PI a.Mu.sa
so CVS
70«
MCA R

B
|.

LaM 1r26ax 17 8-07:43 AW

1Ó0

2S0 SO CBptr 46
I)
Swnpl* VoàMne (mn) 6 Pobt . 2 PW
1 1 1 1 1 1 1 I 8 tavw :nW):
ISPÍA 390
1003 Scale (He) 6600

I 1 I i 1 1 1 1 o

5 00/ S.0O
4.75 / 133
IMn
? lataii 430/13.0 n 0.8B
g SO S.60 1 zeo
D6 0.12 0.38
1 1
•

50 ! „ ' • 1 . ' • I.. • • ' . • • > . -

FIG. 7A. Transcranial Doppler waveform of oscillating flow. There is a sharp forward
flow in systole with reversal of flow in diastole. This waveform is consistent with cerebral
circulatory arrest.
FIG. 7B. Transcranial Doppler waveform of a systolic spike. There is brief forward flow
during systole with no flow during diastole. This waveform is consistent with cerebral
circulatory arrest.

interpretation of vasospasm in the presence of subaracbnoid hemorrhage, sickle cell

vasculopathy, or intracranial vessel narrowing. Low pulsatility indices may indicate
proximal vessel occlusion or the presence of an AVM or cavernous carotid fistula.
Increased pulsatility indices can suggest distal vessel occlusion or increased intracra-
nial pressure. Changes in flow direction may result from a proximal vessel narrowing
or occlusion or may indicate the presence of collateral flow. Knowledge of cerebral
hemodynamics is an important part of TCD interpretation. However, obtaining the
patient's history and understanding other possible disease processes is also needed in
order to accurately interpret the TCD.

ACKNOWLEDGEMENT
We would like to thank the Duke University Hospital Neurodiagnostic Lab staff
for their support, and especially Rebecca Rendahl, R. EEG T.. RPSGT, BS and Mike
Blake for their assistance, with this paper.
 TCD INTERPRETA TION 259

REFERENCES
Aaslid R. Huticr P, Nornes H. Evalualion of cerebrovascular spasm with transcranial Doppier
ultrasound- J Neumsurg l')S4: 60:37-41.
Adams RJ. McKie VC. Bramhilla D, Carl E, Gallagher D, Nichols FT, Roach S. Abboud M,
Berman B. Driscoli C. Files B, Hsu (., Hiirlei A, Miller S. Olivieri N. Pegclow C. Scher
C. Vichinsky E. Wang W. Woods G, Kutler A. Wrighl E, Hagner S. Tighc F, Lewin J, Cure J,
Ziiiiiiicnnari RA. Waclawiw MA. Stroke prevention trial in sickle cell anemia. Control Clin
Trials 1998a: 19:110-29.
Adams RJ. McKic VC, Hsu L, Files B. Vichinsky E, Pegelow C. Abboud M. Gallagher D. Kutlar
A. Nichols FT. Bonds DR. Brambilla D. Woods G. Olivieri N. Driscoli C, Miller S. Wang
W. Hurlett A. Scher C. Berman B. Carl E. Jones AM. Roach ES. Wrighl E. Zimmerman
RA, Waclawiw M, Prevention of a first stroke by transfusions in children with sickle cell
anemia and abnormal results on transcranial Doppler ultrasonography. N Ene J Med 1998b-
.Í39:S-II.
Babikian VL. Transcranial Doppler evaluation of patients with ischémie cerehrovascular disease. In:
Babikian VL. Wechsler I.R (Editors). Transcranial Doppler Ultrasonography. First edition. St.
Louis: Mosby-Year Book. Inc: 1993. p. 87-104.
Carter SA. Hemodynamic considerations in peripheral and cerebrovascular disease. In: Zweibel WJ
(Editor). Introduction to Vascular Ulirasonography. Third edition. Philadelphia: WB Saunders
Company: 1992. p. 3-17.
Ducrocq X, Hassler W. Morilake K, Newell DW, vonReutem GM, Shiogai T. Smith RR. Consensus
opinion on diagnosis of cerebral circulatory arrest using Doppler-sonography: Task Force
Group on cerebral death of the Neurosonotogy Research Group of the World Federation of
Neurology. J Neurol Sei 1998: 159:145-50.
Giller CA. Arteriovenous mallomiations. In: Babikian VL, Wechsler LR (Editors). Transcranial
Doppler Uttrasonography. St. Louis: Mosby-Year Book Inc: 1993. p. 128-38.
Katz ML. Alcxandrov AV. Á Practical Guide to Transcranial Doppler Examinations. Littleton:
Summer Publishing: 2003, p. 25-48.
Lindegaard KE. Bakke SJ. Sorieherg W. Nakstad P, Nornes H. A non-invasive Doppler ultrasound
method for the evaluation of patients with subarachnoid hemorrhage. Acta Radlol SuppI 1986'
369:96-98.
LonerganGJ. CiineDB. Abhondan/.oSL. Sickle cell anemia. Radiographies 2001: 21:971-94.
Newell DW. Winn HR, Evaluation of vasospasm using transcranial Doppler Contemp Ncurosiirg
1989: 11(I8}:IO-15.
Otis SM. Ringelstfin EB. Transcraniaf Doppler sonography. In: Zweihel WJ (Editor), [ntroduction
to Vascular Ulirasonography. Third edition. Philadelphia: WB Saunders Company 1992
p. 145-71.
R{)scnknin/ K. Langer R, Felix R. Transcranial Doppler sonography: collateral pathways in internal
carotid artery obstructions. Angiology 1991: 42:819-26.
Sloan MA. Detection of vasospasm Ibllowing subaracbnoid hemorrhage. In: Babikian VK. Wechs-
ker LR (Editors). Transcranial Doppler Ultrasonography. St. Louis: Mosby-Year Book. Inc;
1993. p. UI5-27.
Sloan MA. Alexandrov AV, Tegeler CH, Spencer MP. Caplan LR. Feldmann E, Wechsler LR,
Newell DW. Gomez CR. Babikian VL, Lefkowitz D. Golman RS. Armon C. Hsu CY, Goodin
DS. Asse.ssment: Transcranial Doppler ultrasonography. Report of the Therapeutics and
Technology Assessment Subcommittee of the American Academy of Neurology. Neurology
2004:62:1468-81.
Sviri GE, Ghodke B. Britz GW. Douville CM, Haynor DR. Mesiwala AH. Lam AM, Newell
DW. Transcranial Doppler grading criteria for basilar arterv vasospasm, Neurosurcery 2006-
59:360-66.
Zweibel WJ. Spectrum analysis in Doppler vascular diagnosis. In: Zweibel WJ (Editor). Introduc-
tion lo Vascular Ultrasonography. Third edition. Philadelphia: WB Saunders Company: 1992,
p. 45-65.