511
Carotid Blood Flow in Man
Determined by Video
Dilution Technique: I. Theory,
Procedure, and Normal Values
Bo. M. T. Lantz'
Arthur B. Dublin
John P. McGahan
Daniel P. Link
The blood flows in the common, internal , and external carotid arteries were determined as a percentage of the cardiac output by video dilution technique in 20 normal
subjects during routine angiography. Nine women and 11 men , ages 19-63 years ,
displayed a mean flow in the common carotid of 8.5% (SO ± 0,9%; n = 40); internal
carotid, 5,3% (SO ± 1.0%; n = 24); and external carotid , 3 .2% (SO ± 0.4%; n = 24).
Relative flow is calculated by a modification of the Stewart-Hamilton principle. The
technique is fast, Simple, highly accurate, and avoids the errors connected with
previous videodensitometric mean transit time techniques. The method can be used in
routine angiography without prolonging the catheterization procedure or adding to the
patient's risk or cost.
There is currently no accurate c lin ical technique for measuring blood flow in
individual arteries in a noninvasive manner. Numerous experim ental methods
have been advocated, including electromagnetic flowmeters, radionuclide procedures, Doppler sonographic techniques, and videodensitometry. None is
widely accepted and very few have been specifically appli ed to the determination
of carotid blood flow in man.
Review of Previous Techniques
Received March 2, 1981; accepted after revision July 7, 1981 .
Presented in part at the annual meeting of the
American Society of Neuroradiology, Ch icago, IL,
April 1981 .
' All authors: Departm ent of Di agnostic Radiology, Uni versity of California Davis Medi ca l Center,
1 09R Professional Bldg ., 4301 X St., Sacramento,
CA 95817 . Address reprint requests to B. M . T.
Lantz.
AJNR 2:511-516, November/ December 1981
0195-6108 / 81 / 0206-0511 $00 .00
© Ameri can Roentgen Ray Society
Our present knowledge of carotid blood flow is based mainly on reports of
direct electromagnetic flow readings in man during surgery. In the 1960s Hardesty et al. [1, 2] applied electromagnetic flowmeters around the common carotid
artery in 11 patients with neck tumors at the time of dissection . The flow readings
were obtained during general anesthesia (Fluothane) and with the internal jugular
vein ligated on the side of arterial exposure. By clamping the external carotid
artery, the mean blood flow of the internal carotid was estimated at 3 64 ml / min
(range, 289-494 ml / min). In four patients , 70 % of the common carotid flow went
to the internal carotid . Assuming that the blood flow in both vertebral art eri es
was equal to the flow in one internal carotid , total cerebral blood flow was
calculated to be 1,090 ml / min. In a similar study, Kristiansen and Krog [3]
estimated the mean common carotid flow in five patients to be 500 ml / min, of
which two-thirds went to the internal carotid and one-third to the external carotid.
Anoth er study of 18 patients with aneurysms and tumors showed a considerable
range of variation of carotid flow in different patients . [4].
The supply of the brain by two internal carotid and two vertebral arteries
together with the drainage via two jugular veins excludes the application of
conventional techniques for blood flow determination in organs supplied by a
single artery and vein. In addition, anastomoses between the arteries and mi xing
of venous blood from different sources in the jugular veins exc lud e methods that
give quantitative results elsewhere.
It is generally assumed that the cerebral circulation takes about 13% -14 % of
512
LANTZ ET AL.
the cardiac output in the adult man [5, 6]. In a survey of 82
cases examined by the original nitrous oxide technique of
Kety and Schimdt [7] , Wade and Bishop [5] reported a mean
cerebral blood flow of 53 ml / min / 100 g brain tissue, corresponding to a total flow of 740 ml / min; the average weight
of the brain was taken to be 1,400 g. This in turn would be
calculated to be 12.8% of the average cardiac output of
5,800 ml / min in adults. Other investigators, using the Stewart-Hamilton indicator dilution method [8-10] and the Fick
principle with external scintillation detectors [11 -14], have
reported a total cerebral blood flow in the range of 7081,247 ml / min.
A few radiologic techniques using densitometry have been
used in the past. In 1966, Hilal et al. [15, 16] described a
new indicator dilution technique for blood flow determination
in individual carotid arteries. The optical density of injected
contrast medium at constant rate was determined on the
radiographs. The concentration of the contrast material in
the vessel was calculated from the measured diameter of
the artery and from the quantitative comparison of the
density of the opaque medium in the artery and that of a set
of tubes containing known concentrations of the same contrast medium. In patients older than 50 years, the normal
average internal carotid artery blood flow obtained by this
method was 331 ml / min and the normal average common
carotid artery flow was 516 ml / min, well in agreement with
the previous values obtained by electromagnetic flowmeter.
Interest for arterial blood flow determination was renewed
with the development of videodensitometry [17]. The high
dynamic response of this technique makes it well adapted
to study circulatory hemodynamics. The device was used
by several groups [18-24] to calculate blood flow according
to the mean transit time technique . Their interest was exclusively directed toward the coronary blood flow and no
estimates of carotid flow were reported. The transit time
technique has three major disadvantages compared with
the video dilution technique . First, it requires accurate measurement of the arterial segment volume . Inaccuracies of
magnification, vessel diameter, segment length , and, to a
lesser extent, deviations of the vessel out of the fluoroscopic
plane lead to severe errors of calculated flow. Second , the
transit time technique assumes that the flow in the artery is
constant, which is clearly not true. Measurements obtained
vary depending on what phase of the cardiac cycle was
used. Transit time measured during a complete cardiac
cycle would avoid this problem, but only a segment of
thoracic aorta is large enough to contain the flow of a
complete cardiac cycle. The capacity of all other peripheral
arteries restricts the transit time in the segment of the
fluoroscopic field to well under one cardiac cycle. Third,
clinical interpretations of blood flow expressed in absolute
units of milliliters per minute is difficult. It varies with parameters such as the patient's weight, height, and body surface.
Flow through an individual artery is evaluated most easily
as a fraction of a common patient denominator such as the
cardiac output. Hence, the transit time technique has not
achieved general clinical acceptance.
Video dilution technique offers clear advantages over all
previous techniques in being a fast, simple , and highly
AJNR :2. November ( December 1981
accurate method of determining carotid blood flow that can
be performed during routine angiography. It has been developed and thoroughly tested in a hydrodynamic flow model
[25, 26] and extensively validated in the animal model [27]
before being applied to the clinical setting.
METHODS
By the video dilution technique, it is possible to estimate the
blood flow in any selectively catheterized artery as a percentage of
a reference flow . It is performed during fluoroscopy by recording
two consecutive injections on videotape. The method is an indicator
dilution technique building on the Stewart-Hamilton principle where
the sampling catheter has been replaced by a videodensitometric
window. The integrated densitometric area of the dilution curves
represents a mass-time curve in contrast to the concentration-time
curve obtained by conventional indicator dilution techniques. The
area, A, is directly proportional to the amount of injected contrast
medium , M, and inversely proportional to the flow, 0, at the site of
injection. Thus, A = k(M / O), where k is a proportionality constant
affected by the rad iation intensity and the gains of th e television
c hain and recording equipment. Fortunately , k need not be determined as it will eventually cancel out in subsequent ca lculations.
If two contrast injections are made in a circulatory system and
the dilution curves recorded downstream over the same cross
section of a selective artery , then the flows at the two injection sites
can be calculated according to 0, 10 2 = (M, . A2 )/ (M 2 • A,). This
relationship has been extensively studied in a hydrodynamic flow
model where the ratio 0, / 0 2 has been determined by video dilution
technique and compared with volumetric flows [25, 26]. The technique has been validated in an animal model (20 dogs , 389 measurements) comparing the video dilution technique and electromagnetic flow readings [27]. A nested random effects analysis gave a
correlation coefficient of 0.99 (r 2 = 0.98). The 95% co nfidence
limits for the c ombined data were -2.8% and +2.4 % [27] .
Video dilution technique is currently used as a standard procedure performed in conjunction with routine angiography in our
institution. Over the past year more than 100 patients undergoing
neuroangiography have had carotid flow measured . The technique
[28-30] will be reviewed briefly.
By percutaneous technique (Seldinger) in th e groin, a stand ard
cereb ral catheter (5 French JB 2, Cook Co.) is placed with th e tip
of the catheter in the ascending aorta under fluoroscopic control
2-4 cm beyond the aortic valve. The image intensifier is then
ce ntered and locked in a midline positi on cove ring the pro ximal part
of th e aortic arch and th e main arteries of the neck with the head in
a straight supine position (fig. 1). An injection of 20-30 ml reg ular
cerebral contrast medium (m eg lumine diatrizoate , Conray-60, Mallinc krodt) by pressure injector with highest allowable injection rate
is rec orded on videotape during fluoroscopy with locked vo ltage
and amperag e. During th e recording, the pati ent holds his breath.
Th en th e catheter is positioned with the tip in th e left co mmon
ca rotid artery and a manual injection of 1 ml contrast materi al is
recorded . With the same position of the image intensifier and th e
patient (position 1), the catheter is now pl aced in th e right co mmon
ca rotid artery for recording of a 1 ml manual injec ti o n of co ntrast.
Dilution curves obtained by the densitometric window cove rin g
c ross sectio ns of the left and right co mmon ca rotid arterie s may be
obtained on line or after the procedure by repl aying the tape. An
estim ate of th e left co mmon carotid flow as a fraction of th e card iac
output is seen in figure 1. The position of the densitometric window
over th e left co mon carotid artery has to be the sa me during the
aorti c arc h and th e selective injection . Th e same reference injec ti on
in th e arch is used for ca lc ulation of th e flow in the right common
carotid .
AJNR:2, November / December 198 1
VIDEO DILUTION FOR CAROTID BLOOD FLOW
Fig. 1.-A, Cardiac output measured
over left com mon carotid artery afte r 30
ml contrast injection in ascending aorta
is calculated as 2 1 0 in arbitrary units. B ,
Left co mmon carotid fl ow measured at
same site after 1 ml contrast inject ion
into art ery is calculated as 18.5 in arbitrary units. Left common carotid flow as
fraction of cardiac output is th en 18.5 /
210 x 100 = 8.8% .
Position I
Position 2 .
Slep 2.
f~
A. Cardiac Output
513
VIDEO
WINDOW
A. Right Common Carotid Flow
Mol
A '040
Qo
Fig . 2.-A, right common carotid flow
measured over right internal caro tid after
1 ml injection into right common carotid
is calc ul ated as 25.0 in arbitrary units.
B, Right internal carotid flow at same site
after 1 ml injection into art ery is calculated as 15.9 in arbitrary units. Flow ratio
of right intern al ca rotid / right common
carotid is then 15.9 / 25 x 100 =
63 .8 %.
B.
Left Common Carotid Flow
D
1000 0 25 .0
B. Right Internal Carotid Flow
Mol
M =I
A 054
Q
io x
0
~4
A 063
Q=~X IOOOOI5 . 9
x 1000 0 185
,-"II~
1
Determination of the right internal carotid flow as a fraction of the
right common carotid flow is illustrated in figure 2. The image
intensifier is moved ,to position 2 and 1 ml of contrast material is
injected twice into both common carotid and internal carotid arteries, respectively . Both dilution curves are recorded with the window
covering a c ross section of the internal carotid without movement
of the patient or the image intensifier during the two consecutive
injections. A similar procedure is performed in a third position
covering the left common and internal carotid arteries.
The manual injections were performed with a 1 ml tuberculin
syringe. Before each injection , the catheter was carefully filled and
there was a 2 min interval between injections to avoid effects of
contrast material on the circulation. All recordings were performed
on standard angiographic equipment including a three-phase generator and a cesium iodide image intensifier (25.4, 15.2 cm). The
25.4 cm input was used in most procedures. Contrast injections
were recorded during fluoroscopy with locked voltage and amperage on a 1.9 cm Sony Umatic videocassette recorder. After angiography, dilution c urves were obtained by replaying th e videotape an d
placing the densitometric window in appropriate position . Th e
videodensitometer (T. Strad, Siemens-Elema, Sweden) had an output voltage proportional to the logarithm of the incident radiation to
provide linearity . The area under the densitometric curves was
determined by planimetry . All measurements were obtained in conjunction with routine angiography in the interim while waiting for the
serial film s to be processed.
Materials
Normal values of blood flow in the common, internal , and external
carotid arteries were obtained from patients undergoing routin e
angiography who met th e fOllowing c riteria: (1) norm al angiog ram
(and CT if applicable), (2) negative laboratory findin gs, (3) no
2
neurolog ic symptoms at th e time of examination or when d ischa rged, and (4) c linicall y judged as normal. Nine women and 11
men 19-63 years of age were examined by video dilutio n techniqu e.
Some of the pati ents we re admitted in co nnection with traumatic
accidents, others for neurological workup after previous neurolog ic
symptom s. In all patients, ang iography was requ ested and the fl ow
studies performed as a co mplementary proced ure with the patient 's
co nsent. All patients we re examined in th e supine position in a
co nscious state with mild sedation (Dem erol, 50 mg) preceding th e
ang iog raphi c proced ure. No co mpli catio ns were seen during th e
procedure or during hospitali zati on. Th e en tire tim e for th e fl ow
stud y did not usuall y exceed 10 min , wi th add itional flu oroscopic
time not more th an 100 sec . Additi onal co ntrast material used for
th e fl ow studies was within the range of 25-35 ml. In onl y eig ht
patients , th e com mon carotid artery flow was determined as a
percentage of the ca rdi ac output. In another 1 2 patients, a co mplete
workup was obtained with flow determination of the co mm on, internal , and ex tern al carotid arteri es. The external carot id flow was
calc ulated from th e difference between the co mmon and internal
carotid fl ows.
Results
Individual estimates in 20 normal subjects are displayed
in figure 3 and table 1. No differentiation by age group or
gender was determined. The mean blood flow in the right
common carotid (n = 20) and the left common carotid (n
= 20) was equal to 8.5 % of the cardi ac output (SO ± 0.9).
A slightly larger flow was obtained in th e right intern al
carotid , 5.4 % (SO ± 0.9) , compared with the left sid e, 5.3 %
(S .D. ± 1 .0), in 12 patients. However , the differe nce was
not signifi cant (p < 0.01). The co mbined flow of th e right
514
LANTZ ET AL.
®
CD
3.3% (0.4)
+--- - - --8.5% (0.9)
100%
Fig. 3.-lntern al and external carotid blood flow derived from individual
measurements in 12 norm al human subj ects. Common carotid flow is corresponding figure from individual measurements in 20 subjects .
and left common carotid arteries in 20 patients was 17.0%
(SO ± 1.6) and the corresponding combined figure for the
right and left internal carotid arte ri es in 12 patients was
10.7% (SO ± 1.6).
The mean common carotid blood flow calculated from 40
ind ividual estim ates bilaterally in 20 patients was 8 .5 % of
the card iac output (SO ± 0 .9). The corresponding figure for
all internal carotid arteries (n = 24) was 5.3 % (SO ± 1.0)
and for the external carotid arteries (n = 24), 3 .2% of the
cardiac output (SO ± 0.4).
Discussion
In our estimates, the carotid blood flow was expressed as
a fraction of the cardiac output. Since the reference injection
for cardiac output was performed in the ascending aorta 24 cm from the aortic valve , the assumption is made that this
flow is equal to the left ventricular output. This is not strictly
true as the coronary blood flow of about 4 % has been
negl ected. Some of the pressure-injected contrast material
in the ascending aorta will probably mi x in the coronary
sinus. Our estimate of cardiac output might then be about
2 % less than the true left ventricular output giving a corresponding overestimate of the carotid flows. A 2 % correction
of this error would give a blood flow in the internal carotid
of 5.2 % instead of the reported 5.3 %. This small error has
here been neglected assuming that the cardiac output is
equal to the flow in the proximal part of the ascending aorta.
The mean blood flow through the common carotid , 8.5 % ,
internal carotid, 5.3 %, and external carotid, 3 .2 % of the
cardiac output, would correspond to the absolute amount of
493 , 307, and 186 ml / min, respectively , assuming the
average cardiac output in man to be 5,800 ml / min. These
figures correspond reasonably well with previously reported
figures obtai ned by the electromag netic flowmeter [1-4]
and by the technique described by Hilal et al [15, 16]. In our
study , the internal carotid received 62.4 % of the flow in the
AJNR :2. November/ December 1981
common carotid. This also agrees with previous reports [13 , 15, 16].
The total blood flow through both internal carotids in 12
patients was 10.7 % (SO ± 1.6) of the cardiac output. This
figure reflects only a part of the cerebral circulation. Additional supp ly is provided by both vertebral arteries, presumably to a lesser extent. In addition, both internal carotid and
vertebral arteries give off extracerebral branches. The internal carotids supply the ocular muscles, the lacrimal gland,
the nasal mucosa, and the neural tissue of the orbits via the
ophthalmic artery . The vertebral arteries similarly have several small branches supporting the neck before entering the
skull. However, these extracerebral branches are not likely
to carry a significant amount of blood [6]. From preliminary
estimates, the flow in each of the vertebral arteries has been
calculated to be 1.9% of the cardiac output (Lantz BMT,
Link OP , Holcroft JW, Foerster JM, unpublished data). This
means that the total inflow of blood to the skull through both
intern al and vertebral arteries would be 14.5% of the cardiac
output under normal resting conditions. Thus, the blood
supply of the brain with exclusion of extracerebral branches
of the internal carotid and vertebral arteries would amount
to about 12% -14 % . This figure corresponds very well to
data published on total cerebral blood flow [5, 6]. Total
cerebral blood flow estimated by indicator dilution methods
[8-10] and by external scintillation detectors [11 -14] is
reported in absolute units to be 708-1,247 ml / min . Converted to relative measures of the average cardiac output of
5,800 ml / min, the corresponding fractional cerebral blood
flow would be 12.2% -21 .5 %. The total cerebral blood flow
of 14% of the cardiac output estimated by the video dilution
technique would amount to 812 ml / min in the average adult
patient.
Absolute flow estimates in milliliters / time unit is not possible with the video dilution technique. However, this is the
strength of the method, as the absolute flow estimates have
no c lini cal relevance unless related to other patient parameters such as weight, height, or body surface. It expresses
regional artery blood flow as a percentage of cardiac output.
The data show that the estimates of flow at rest in the
carotids are fairly constant in a normal population with small
deviations from the mean value. Experience is still too small
for differentiation of flow data in regard to gender and age
groups, but preliminary results suggest no significant difference in this regard . A continuous accumulation of data will
answer these questions.
When using the video dilution technique in the clinical
setting, several sources of error should be considered regarding the effect of contrast material on circulation, mixing
of contrast with blood, and the recording technique . These
issues were addressed thoroughly in a recent review [31].
However, practical advice when using the video dilution
technique is outlined briefly.
It is well known that angiographic contrast media have
effects on the cereb ral hemodynamics [32-34]' This effect
is transient and the blood flow usually returns to normal
within 30-60 sec . To avoid local and general effects on
circulatory hemodyn amics, it is advisable to wait about 2
min before an injection is repeated. It is also preferable that
AJNR :2, Nove mber / December 1981
VIDEO DILUTION FOR CAROTID BLOOD FLOW
515
TABLE 1 : Carotid Blood Flow Determination by Video Dilution Technique in 20 Normal Subjects at Rest
RCC
Subject No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
.... . . . . .
. . . .. . . . .
.. .... . . . . .. .... . . . . .
.... . . . .. ..
. .. . . . . . .
. ... .. . . .
... .. . ..
..... .. .
Mean
Standard deviation
Total no. of subjects
Note .- Arte ri es are abbrev iated as:
and LEG = left external ca rotid .
Rce
8.3
7.9
9.5
7.8
9.3
8 .3
8 .1
7.7
8 .2
8 .7
7 .9
11 .0
8 .8
8 .6
8 .8
7.1
7.0
9 .9
8 .5
8.9
8 .5
0.9
20
RI C
RE C
LCC
5.3
5.4
4 .3
7.2
5.9
6 .3
5.1
4 .2
3.9
6.0
5.4
5.8
2.9
3.3
3.6
3.8
2.9
2.3
3 .7
2.9
3 .1
3 .9
3. 1
3. 1
8. 7
7.6
8. 7
8.2
8 .1
8 .1
8.5
8 .2
8.0
9 .3
9 .8
9 .9
7.8
7.6
8 .3
6.9
7.9
10.4
8 .8
8 .5
5 .0
6. 2
5 .9
6.4
5 .0
4 .0
4 .2
3 .9
4 .8
7.1
5 .6
5 .2
5.4
0.9
12
3 .2
0 .5
12
8 .5
0.9
20
5 .3
1.0
12
Ll C
LEC
RCC
RI C
+ Ll C
REC
+ LEC
3. 0
3.1
3. 9
3. 5
2.8
3. 6
4 .1
3. 0
3.1
3.3
3.2
3 .3
10.3
11 .6
10.2
13.6
10.9
10.3
9.3
8. 1
8. 7
13 .1
11 .0
11 .0
5 .9
6 .4
7.5
7.3
5.7
5 .9
7.8
5.9
6.2
7.2
6.3
6. 4
3 .3
0.4
12
17 .0
1.6
20
10.7
1.6
12
6.5
0 .7
12
= right comm on ca ro tid, RIC = right intern al c aroti d , RE C = rig ht ex tern al c arotid ,
the patient hold his breath during contrast recordings to
avoid cyclic fluctuations of the densitometric baseline. It is
extremely important that two consecutive recordings are
made with the densitometric w indow over the exact same
cross section of the artery without movement of the patient
or the image intensifier in between . The only d isadvantage
of the technique seems to be the requirement of patient
cooperation. In our experiments , about 2 % of all recordings
have to be deleted for this reason .
Recirculation of contrast material through the venous
system can be avoided by repositioning the electron ic window. Also , spillover of contrast material from the selective
artery in the aorta can be checked easily by replay ing the
tape . In those instances, the recordings should be deleted.
Contrast injections in the aorta are performed in a direction opposite to the flow. Good mi xing of contrast materi al
with blood is anticipated. However, w ith selective injection,
the contrast material is injected in the same direction as the
flow and perfect mixing is not anticipated . The latt er does
not seem to play an important role, as the densitometric
window is covering a complete cross section of the artery ,
recording the total amount of passing contrast material. This
function has been studied by comparing a series of injections in the superfic ial femoral artery in dogs in the same
and the opposite direction of the flow [31, 35]. There was
no sign ificant difference between the techniques (p < 0.01) .
As a precaution, we have usually maintained a distance of
at least 2 cm between the catheter tip and the w indow and
at least 3 cm between the catheter tip in the comon carotid
and the carotid bifurcation .
By determining blood flow in individual carotid arteries ,
+ LCC
17 .0
15 .5
18.2
16. 0
17.4
16.4
16.6
15.9
16.2
18 .0
17 .7
20 .9
16.6
16.2
17 .1
14.0
14.9
20 .3
17 .3
17.4
Lee =
left common ca ro tid ,
lie
= left in tern al ca ro ti d,
the video dilution technique is capable of giving important
physiologic information in a variety of clinical situations. Th e
extent of cerebral ischemia caused by arterial obstructive
disease, intrac ranial aneurysms , and emboli can be evaluated before surgery . Also in the treatment of stroke and
brain injury and in determining the prognosis in comato se
patient s, th e video diluton tec hnique may be of great impo rtanc e . It can be performed in any angiographi c suite w ith
the addition of a videodensitom eter . Th e tec hnique is highl y
acc urate and gives information about hemodynami c s that
cannot be obtained easily by any other means. It does not
contribute to the patient's risk or cost and there are no
contraindications other than those for angiography.
Further papers in this series will deal w ith the influence of
vascular disease and intracranial mass effects upon the
c arotid blood flow.
REFERENCES
1.
2.
3.
4.
5.
Hardesty WH , Rob erts B, Toole JF, Roy ster HP. Studies of
carotid artery blood fl ow in man . N En gl J M ed 1960;2 63:
944 - 946
Hard esty WH , Roberts B, Toole JF, Royster HP. Stu dies of
carotid art ery fl ow. Surg ery 1961 ;49: 251-256
Kri sti ansen K, Krog J. Electromagneti c studi es on th e blood
fl ow through th e carotid system in man. Neurology (N Y)
1962;12: 20- 22
Tindall GT, Odom GL, Cupp HB, Dill on ML. Stud ies on carotid
artery fl ow and pressure. Observations in 18 pati ents during
graded occlu sion of proxi mal carotid artery . J Neurosurg
1962;19 :91 7-923
Wade OL, Bi shop JM . Ca rdiac outp ut and regional blood flow.
516
LANTZ ET AL.
Oxfo rd, En gland: Blackwell , 1962 :93
6. Milnor WR . Reg ional ci rc ulations. In : Mountcastl e VB , ed .
M edical physiology. St. Louis : Mosby, 1980;2 : 1 094 -11 07
7. Kety SS , Schmidt CF. Th e nitrous oxide method for quantitative
determin ation of ce rebral blood fl ow in man: theory, procedure
and normal values . J Clin In vest 1948;2 7 : 47 6-483
8 . Shenkin HA , Harm el M H, Kety SS . Dynamic anatomy of the
cereb ral c irc ulation. Arch Neurol Psychiatr 1948;60 : 240- 252
9 . Nylin G, Sil versk iold BP , Lofsted t S, Regnstro m 0 , Hedlund S.
Studies on cerebra l blood fl ow in man using radioactive labelled erythrocytes. Brain 1960;83: 293-335
10. Helli nger FR , Bloor BM , McC utc hen JJ . Total cerebral bl ood
flow and oxygen co nsumption using th e dye-dilution meth od .
A study of occlusive arterial disease and cerebral in fa rcti on. J
Neurosurg 1962 ; 19 : 964-9 70
11. Lewis BM , Sokoloff L, Wec hsler RL , Weutz WB, Kety SS. A
meth od fo r con tin ous measurement of ce rebral blood fl ow in
man by means of rad ioacti ve kry pton (Kr-7 9). J Clin In ves t
1960;39: 707 -716
12 . Reinmu th OM , Schein berg P, Bronne B . Total cerebral blood
flow and metaboli sm. Arch Neuro/1965; 12: 49- 66.
13 . Sapirstein LA . Measurement of ceph alic and cerebral blood
flow frac ti ons of the card iac outpu t in man. J Clin In vest
1962;4 1 : 1429 -1 435
14. Steiner SH, Hsu K, Olin er L, Behn ke RH . Th e measurement of
ce rebral blood flow by extern al isotope counting . J Clin In vest
1962 ;4 1 : 222 1-2232
15. Hilal SK , Resch JA , Am platz K. Determin ation of the carotid
and regional cerebra l blood flow by a radiographic technique.
A c ta Radiol [ Oiagnj (Stockh) 1966; 5 : 232 - 240
16. Hilal SK . Human caroti d artery flow determin ation using a
rad iog raphic tec hnique. In vest Radio/ 1966 ;1 : 11 3-1 22
1'7 . Wood EH, Sturm RE, Sanders JJ . Data processng in cardiovascular physio logy w ith particu lar reference to roentgen vi deodensitometry. Proc Mayo Clin 1964;39: 849-8 65
18. Ru ti shauser W , Bu ssman WD , Nosed a G, Meier W . Wellauer J .
Bl ood flow measurement th roug h single co ronary arteri es by
roentgen densitometry. I. A co mpari son of flow .measured by a
rad iolog ic tec hniq ue appli cab le in th e intact organism and by
elec tromag neti c flowmeter. AJR 1970; 109: 12-20
19. Ruti shauser W , Noseda G, Bu ssman WD. Preter B . Blood flow
measu rement throug h single coronary arteries by roentgen
densitometry . II. Ri ght co ronary artery flow in consciou s man.
AJR 19 70 ;109:2 1-24
20 . Ru tishauser W . Application of roe ntgen densitometry to blood
flow measurement in models, animals and in intact co nsc ious
man . In Heintzen PH, ed . Roentgen- , c ine- and videodensitometry . Stu ttgart: Thi eme, 197 1 : 1 08 -11 8
2 1. Sm ith H, Frye R, Davi s G, et al. Measurement of flow in
22 .
23 .
24.
25 .
AJNR :2, November/ December 1981
saph enous vein-co ron ary artery grafts by roentg en videodensitometry. Circulation [suppl 2]1971 ;44: 1 07
Smith H, Frye R, Donald 0, et al. Roentgen videod ensitometri c
measurement of coro nary blood fl ow . Determination from simultaneous ind icator-diluti on c urves at selected sites in th e
co ronary c irc ulation and in coronary artery-saph enous vein
grafts. Proc Mayo Clin 1971 ;46: 800-80 6
Bursch J, Johs R, Kirbach H, Sc hnurer C, Heintzen P. Accuracy of videodensitometri c flow measurement. In : Heintzen PH ,
ed . Roentgen- , cine- and videodensitometry . Stuttg art : Thieme ,
1971 ; 11 9 -1 32
Rosen L, Silverm an NR. Vid eod ensitometric measure ments of
blood fl ow using c ross-correlati on techniques. Radiology
1973 ;109: 305-3 10
Lantz BMT. A methodologic in vestiga tion of ro entgen videodensitometric measurements of relative flo w. Berk eley: University of Californi a, 1974.
26 . Lantz BMT. Relative flow measured by roentgen vid eod ensitometry in hydrodynamic mode l. Ac ta Radiol [Oiagnj (Stock h)
1975;16:50 3- 5 19
27. Holcroft JW , Lantz BMT, Link DP , Foerster JM . Th e video
dilution tec hniq ue accurately measures bl ood flow. Surg Forum
1980;31 :324- 32 5
28. Lantz BMT, Foerster JM , Lin k DP , Holocroft JW. Angiographi c
determin ation of ce rebral blood flow . Acta Radiol [Oiagnj
(Stockh) 1980;21 : 147-1 53
29. Lantz BMT, Dublin AB , Don ald PJ, Lin k DP, McGahan JP.
Cerebral bl ood fl ow determin ation by video dilution techniq ue
in a patient with malignant c hemodectoma. Acta Radiol [Oiagnj
(S tock h) 1980; 2 1 : 69 7- 704
30 . Lantz BMT, Dublin AB , McGahan JP, Lin k DP . Assessment of
cerebral bl ood flow in man by video dilution tec hnique. A
pre liminary report. In vest Radio/1981 ; 16 : 18 1-1 8 7
3 1. Lantz BMT , Foerster JM , Lin k DP, Holcroft JW. Determination
of relative blood flow in single arteries: new video dilution
technique. AJR 1980;13 4 : 1161-1168
32 . Greitz T. Radiologic study of the brain c irculation by rapid
seri al angiography of th e ca rotid artery. Ac ta Radiol [Oiagnj
(Stockh) 1956;[suppl 140]
33 . Hilal SK . Hemodynamic c hanges associ ated with the intraarteri al injection of contrast media. Radiology 1966;86: 615633
3 4. Greitz T, H i rn ell G. Bradyca rdi al reactions during c erebral
angiog raphy; a compari son of Isopaqu e sodium , Isopaque B,
Hypaq ue and Urog rafin . Ac ta Radiol [Oiagnj (Stockh)
1967;[suppl 2 70]: 75-86
35. Holc roft JW, Lantz BMT, Foerster JM , Link DP: Video dilution
technique . An accurate measure of blood flow during routin e
angiograph y. Arch Surg 1980;115 : 1299-1 30 3