EurJ VascSurg 1,227-234 (1987)
F l o w E s t i m a t i o n by D S A in Bypasses w i t h or W i t h o u t a Distal A r t e r i o - v e n o u s Fistula* Gfinter Stelzer,t Dennis P. van Berge H e n e g o u w e n Klinik Oberwald, Fachklinik fiir Gefdfl und Enddarmerkrankungen, Postfach 1 1 4 9 , 6 4 2 4 Grebenhain-i, West Germany Flow of contrast medium in bypasses with or without an arteriovenous fistula (A VF) has been measured using a densitometric technique. Modern digital subtraction angiography (DSA ) equipment has the ability to produce images related to the density of the bolus of contrast passing down the vessel and to measure flow from them. Three studies were performed; (1) the method was tested in a model where it was compared with direct flow measurement; (2) A flow study was carried out in four groups of different below knee bypasses; (3) A study to examine the distribution offlow in adjunctive arteriovenous fistala was performed. The first study showed that the method was reliable and highly reproducible; the mean relative variance of DSA measured flow to the realflow being 3.05% and never more than 10%. The second study showed mean flow for various bypasses as shown below: (1) l Ofem.pop, bypasses (61-20 2 ml/ min). 120 ml/min; (2)lOfem.tib. bypasses (41-190 ml/min ). 120 ml/min; (3) 1Ofera.rib, bypasses with AVF(243~155 ml/min). 321 ml/min; (4) lOpop.pedal, bypasses with AVF(112-382 ml/ min). 261 ml/min. The differences between group 1 and 2 and group 3 and 4 are significant (P < 0.005). The third study showed that relatively little of the blood entering an A VF passes distally into the arterial bed (Mean: 11%, 6-19?/o). In conclusion, DSA flow measurement is a reliable method of flow measurement. An AVF increases graft flow 2-3 times but only about 10% of the bloodpasses distally into the arterial system. Key Words: DSA; Flow measurement-A V Fistula.
Introduction With the introduction of digital subtraction angiography (DSA) a new tool for densitometric flow measurement has become available. At the same time the trauma of angiography can be reduced to a single puncture of the femoral artery with a gauge 20 needle. By reducing the quantity of contrast medium to 5 - 1 0 ml a further reduction in the risk of angiography can be achieved. Modern DSA equipment has the ability to assess density of the image and thus evaluate it more fully. Although the usual DSA-camera is only able to differentiate between 250 steps in the grey scale (the h u m a n eye differentiates between more than 2000 steps), compared with normal * Presented at the Inaugural Meeting of the European Society tbr Vascular Surgery, 7-8 May 1987.
"["Engineer for Informationin Medicine. Please address reprint requests to: Dennis P. van Berge Hengouwen, Klinik Oberwald, Postfach 1149, 6424 Grebenhain-1, West Germany. 0950-821X/87/040227+08503.00/0
© ] 987 Grune &Stratton Ltd
angiography this represents an 8-fold improvement. Since January 1985 a series of flow measurements on different types of bypasses in the lower leg have been performed.
M a t e r i a l s and M e t h o d s Three different studies were carried out. First: The method was tested in a model and compared with simple direct flow measurement. Second: Flow was measured in four different types of below knee bypasses; ten femoro-popliteal bypasses, ten femoro-tibial bypasses, ten femoro-tibial bypasses with an arterio-venous fistula (AVF) integrated in the distal anastomosis and ten popliteal-pedal bypasses also with a distal AVF. Third: In bypasses with a distal AVF the distribution of blood flow was examined.
228
G. Stelzer and D. P. van Berge Henegouwen
Fig. l. The contrast image and calculations at two different ROIs. F start shows the moment where the contrast starts to flow and F max is the moment of maximum density. C = Coin. F start; the moment when the contrast inflow starts. F max; the moment of maximum density (see text).
Comparison offlow A reversed vein s e g m e n t w i t h a l e n g t h of 2 0 c m w a s interpositioned in a n o r m a l i n t r a v e n o u s infusion set. By v a r y i n g the pressure in t h e system, different flows could be achieved. A t the distal end of the t u b e t h e liquid w a s collected a n d the v o l u m e w a s m e a s u r e d . The flow w a s c a l c u l a t e d by dividing the v o l u m e by time. S i m u l t a n e o u s l y a D S A - f l o w m e a s u r e m e n t (DSAFM) was performed u s i n g a n ELSCINT ECLAT l a 700 w i t h a n e x p o s u r e rate of 3.3/s. A small bolus (1 ml) of c o n t r a s t w a s injected into t h e infusion set t h r o u g h a T c o n n e c t i o n placed just p r o x i m a l to the interposed vein at a flow r a t e of 5 ml/s. As the c o n t r a s t passed t w o different points (Regions of interest = ROI) in the vein these events were s h o w n g r a p h i c a l l y (Fig. 1). The t r a n s i t - t i m e b e t w e e n t h e t w o ROIs could t h e n be m e a s u r e d exactly. For r e a s o n s of reproducibility two different points in these curves were compared, first at the onset of inflow (F start) a n d second at the m o m e n t of m a x i m u m d a r k e n i n g (F max). The only p r o b l e m w a s a n a c c u r a t e m e a s u r e m e n t of the diameter, c o n s e q u e n t l y a m e t h o d w h i c h allowed direct m e a s u r e m e n t of the b y p a s s d i a m e t e r from the DSA picture was developed. As a reference a coin (c) of k n o w n d i a m e t e r w a s placed n e x t to the vein, allowing c o n v e r s i o n of t h e Eur ] Vasc Surg Vol 1, August 1987
m e a s u r e d d i a m e t e r in pixels to millimeters. It was importa n t for the c a m e r a a n d the l o g a r i t h m i c amplifier to be a d j u s t e d exactly. This w a s done by a d j u s t i n g the logarithmic amplifier u s i n g i m a g e s w i t h a k n o w n stepwise i n c r e a s i n g density. First a profile of t h e c o n t r a s t - d e n s i t y of t h e vessel w a s m a d e , a 2-fold c o n t r a s t density therefore d o u b l e d t h e d a r k e n i n g to the c a m e r a . If the vessel is circular, flow c a n be c a l c u l a t e d using t h e formula: Flow = r 2 . r c . V r = d i a m e t e r (cm), v =
(cm3/min),
Distance b e t w e e n ROI 1 a n d ROI 2 (cm) T r a n s i t time (s) 0 . 6 0
(cm/min)
This study was done using b o t h a n a r r o w (4 mm), a n d a wide (8 m m ) vein segment.
Flow measurement--in
patients
F o u r g r o u p s of t e n bypasses were studied. All patients h a d b e e n treated for severe p e r i p h e r a l ischaemia, with loss of tissue. On the t e n t h p o s t o p e r a t i v e d a y a n anglo-
D S A F l o w Estimation
229
gram was performed using conventional DSA methods. In order to measure flow a coin of k n o w n diameter had to be placed in the angiography-area as close to the bypass as possible. The common femoral artery was punctured with a 20 gauge (0.9 ram) needle. After checking the puncture site, a small bolus ( 5 - 1 0 m l ) Of non-ionising contrast medium was injected at a flow of 5 ml/s. The DSA film was stored in the computer and after the anglogram was completed flow was measured using two different ROIs over the bypass. To make the measurement more accurate the distance between these two [lOis was made as large as possible within the length of the angiography field ( 1 4 - 3 0 cm). The femoro-popliteal (Group A) and the femorotibial (Group B) bypasses were all performed using reversed autologous greater saphenous vein from the femoral artery to the distal popliteal artery or to one of the lower leg arteries. The femoro-tibial bypasses with distal AVF (Group C) were also performed using reversed autologous vein, but because of severe outflow obstruction already shown in the preoperative angiogram a small side to side AVF was integrated into the distal anastomosis. The popliteal pedal (Group D) bypasses were also constructed using reversed autologous saphenous vein from the distal popliteal artery to the dorsal pedal artery or the distal posterior tibial artery below the level of the ankle joint. Due to the very limited outflow of this kind of bypass an AVF was again integrated into the distal anastomosis. 1
Distribution of flow in A V fistulae In the third study an estimation of the distribution of blood in the distal anastomosis was performed. With the DSA-unit used by us it is not only possible to reproduce the contrast stream graphically, but also to process the data mathematically. Therefore the integral of the darkening of the region of interest is calculated for every picture taken (subtraction). The values calculated are inserted into a Chebycheff-Polynome Formula to the n.th-power. Again under strict conditions provided that the camera and logarithmic amplifier are adjusted correctly, the integral of the Chebycheff-Polynome is related in a linear m a n n e r to the flow of contrast medium. By comparing the ROI over the bypass and the ROI over the recipient artery it is possible to calculate the distribution of contrast, i.e. (blood) (Fig. 2). The distribution was calculated using the following formula: 1S2 (rec.) Flow to the receiver (%) = 1~2 (byp.)" 100
Fig. 2. The DSAFMofa popliteal--pedal bypass from the distal popliteal artery to the dorsal pedal artery. In the lower part of the figure the different curves of the passage of the contrast over the ROIsis shown.
1S2 = Integral of CA-change due to flow of CA (tl/t2); t l = time the inflow starts; t2 = time the outflow ends.
Results
The first study showed that DSA was a reliable way of measuring flow when compared with a direct flow measurement method. The results comparing the two techniques are shown in Table 1. The onset of flow (Flowstart) as well as m a x i m u m flow (Flowmax) can be used as measuring points in this model. The m e a n relative variance between the two methods was 3.05%, and never exceeded 10% (Fig. 3). The results of the second study are shown in Tables 2 - 5 and Fig. 4. There were no complications from the arterial puncture or the injection of contrast in these 4[0 EurJ Vasc Surg Vol 1, August 1987
G. Stelzer and D. P. van Berge Henegouwen
230
Table 1. Comparison between volume flow and DSA measured flow (Flow start and Flow max.). All values in ml/min. Figs. 4 and 5
Wide vein (0.8 ram) Real flow
Narrow vein (OM~mm)
F start
F max
44
46
45
60
65
116
Real flow
F start
F max
42
38
42
64
51
49
53
110
110
122
123
120
150
152
145
172
178
172
187
184
196
207
205
198
300
302
289
300
290
302
375
385
365
375
392
369
studies. The m e a n flow in the femoro-popliteal bypasses group (A) was 1 2 0 m l / m i n ( 6 1 - 2 0 2 m l / m i n ) . The femoro-tibial bypasses group (B) also h a d a m e a n flow of 1 2 0 m l / m i n ( 4 1 - 1 9 0 m l / m i n ) . The highest flow was found in the femoro-tibial group with a n AVF (C) at the distal anastomosis. M e a n flow i n this group was: 321 m l / rain ( 2 4 3 - 4 5 5 m l / m i n ) . On the other h a n d this group showed the lowest postoperative increase i n a n k l e /
400
brachial pressure index (A/B.I). The postoperative A/B.t was O. 72. The group with very distal bypasses (D) from the distal popliteal artery to the foot arteries h a d a m e a n flow of 261 m l / m i n ( 1 1 2 - 3 8 2 m l / m i n ) . The differences between group A + B (without a n AVF at the distal anastomosis) a n d group C + D (with AVF) is highly significant ( M a n n W h i t n e y U - W i l c o x o n r a n k s u m W test P < 0.005). In addition c o m p a r i n g group A with C or D a n d group B with C or D is highly significant ( M a n n W h i t n e y U Wilcoxon r a n k s u m W test P < 0.005). The differences b e t w e e n group A a n d B to one a n o t h e r a n d group C and D to one a n o t h e r were not significant. The distribution study showed that only a small a m o u n t of the contrast e n t e r i n g the distal a n a s t o m o s i s actually passes into the distal arterial system. Mean distal arterial flow in the group of femoro-tibial bypasses with distal AVF was: 9% (6 16%), in the group of popliteal bypasses: 11.8% (8-19%). (Tables 4 a n d 5).
Discussion Densitometric studies have already been applied in diagnostic procedures in cardiology a n d i n some respects are more reliable t h a n the original catheter studies. 6 Fiedler
-
350 •:.:.:,:.:~ ::
300
I i
250
o"
i!!i!ii!i~ii ~i~
:::
i!iiiiiiil i~ili~ililf!i!iii
200
iiii~ililili!:!i!
150
.... i i!i :=iii iil i iii iiiiii
I00 50
..........:............ ,';';':'I';G::::: :: c.::::::::;?:~:::: ;?:::::::!~iiii: :i: :':i:i:i~i:ii:.?
10
.L.
v
c
.o
a
g.
Reol flow
-5
-10
Fig. 3. The relative variance and flow in the DSAFMcompared with real volume flow measurement. EurJ Vasc Surg Vol 1, August 1987
D S A F l o w Estimation
231
Table 2. Flow studies by DSAFM in ten femoro-popliteal (below knee) bypasses: Group A
Patient
Age
Preop. A/B.I
Postop. A/B.I
Flow in Outflow*
ml/min
1
ES
f
73
0.00
1.25
3
184
2
HS
f
81
0.62
0.85
3
112
3
HA
m
76
0.27
0.47
<1
88
4
JR
f
66
0.21
0.95
1
61
5
BS
m
78
0.29
0.94
3
202
6
WF
f
76
O.00
0.56
7
KV
m
74
0.16
1.00
2
102
8
SP
m
77
0,23
0.85
2
138
9
HL
m
63
0,19
1.05
1
97
f
81
O,31
1.40
3
145
Mean
120
10 EU
75
* Outflow scale; 3 = three vessels; 2 = two vessels; 1 = one vessel, < 1 = segment. A/B.I = Ankle/Brachial pressure index.
et al. showed that the densitometric estimation of a sten-
osis in the superficial femoral artery is a reliable method of assessing the degree of a stenosis, s The sensitivity of DSA flow measurement (DSAFM) depends on the accuracy of measurement of the distance from the vessel, its diameter and the determination of transit time. By placing a piece of metal of k n o w n diameter in the angiography area, the distance and diameter can be converted from pixels to millimeters. Since the
law of Lambert-Beer also applies to X-rays using the graphic image the density and the diameter can be checked to ensure that it is a real circle. 16 By making the distance between the two ROIs over the bypass as long as possible, the transit time increases making the DSAFM more exact. Although the m a x i m u m exposure rate of this equipment is 12/s, an exposure rate of 3.3/s was chosen. The main reason for doing this was to improve the quality of the pictures, caused by a low resolving power
Table 3. Flow studies by DSAFM in ten femoro-tibial bypasses: Group B
Patient
Preop.
Postop.
Age
A/B. I
A/B. I
Outflow*
Flow ml/min
1
JT
m
67
0.25
1.05
1
174
2
KD
m
67
0.35
0.95
1
190
3
MZ
f
75
0.29
0.89
1
120
4
DM
f
76
0.22
1.35
1
98
5
FG
m
71
0.26
1.50
1
180
6
HG
m
65
0.19
0.95
1
88
7
MJ
f
73
0.31
1.58
1
87
8
EW
m
79
0.25
1.05
1
68
9
LN
f
77
O.11
1.58
1
41
f
62
0.36
1.06
1
155
Mean
12 0
10 MS
Eur] Vasc Surg Vot 1, August 1987
232
G. Stelzer and D. P. van Berge Henegouwen
Table 4. Flow studies by DSAFM in femoro-tibial bypasses with distal AVF: Group C Patient
Age
Preop. A/B. I
Postop. A/B. I
Outflow* (scale)
Flow ml/min
Distrib flow in %
1
HR
m
75
0.53
1.24
<1
311
11
2
OK
m
77
0.35
1.06
<1
357
8
3
GG
f
78
O.00
0.45
<1
455
10
4
LN
f
77
0.15
0.45
<1
352
8
5
EK
f
82
0.36
0.57
<1
301
16
6
HL
m
83
0.23
0.53
<1
253
8
7
RE
f
79
0.38
0.69
<1
243
6
8
MT
f
74
0.28
0.59
<1
299
6
9
HI(
f
67
0.11
0.48
<1
298
5
f
56
0.36
0.65
<1
345
8
Mean
321
9
10 LB
at the high exposure rate. As expected for flow rates in excess of 500 ml/min a faster exposure rate is necessary. 4 The excellent agreement seen between DSAFM and actual flow measurement irrespective of the diameter of the vein clearly demonstrates the accuracy of the diameter measurement. The mathematical processing of the graph with the Chebycheff polynome helps to find the exact point where the contrast starts to flow in the ROI (F start) and the point of m a x i m u m density (F max). As s h o w n in the first study both points can be used as fixed points to measure the transit time.
The second part of the study showed that this method can differentiate between high and low flow in bypass grafts. These grafts were chosen because they have no branches which can disturb the pressure system and diminish turbulence. Another reason was because data on flow in these grafts is already available 3 for comparison. Lastly, we wanted to discover the flow characteristics of AVF when incorporated into a distal anastomosis. 1 Different kinds of bypasses were used to find out if there were any differences in flow. In this study the DSAFM was performed on the tenth postoperative
Table 5. Flow studies by DSAFM in ten popliteal (below knee)--pedal bypasses: Group D
Patient
Age
Preop. A/B. [
Postop. A/B. I
Outflow* (scale)
Flow ml/rnin,
Distrib flow in %
1
RF
m
62
0.40
1.O6
<1
311
10
2
JK
m
73
0.40
1.00
<1
112
12
3
WM m
57
0.45
1,45
<1
294
8
4
Eli
f
78
0.50
0.96
<1
351
12
5
HB
m
46
0.00
1.23
<1
382
10
6
KB
m
68
0.25
0.86
<1
372
13
7
KK
m
66
0.43
0.81
<1
138
14
8
WL m
82
0.31
0.95
<1
218
9
9
KHB m
71
0.39
1,37
<1
188
19
60
0.38
1.02
<1
244
ll
Mean
261
] 1.8
10 KF
m
Eur i Vasc Surg Vol 1, August 19 8 7
D S A F l o w Estimation
500
450
400
350
000
3OO
m
-~ 25o o~ 200
•
233
system. If this pressure is higher than the preoperative pressure in the area then according to Poiseuille's law more blood will pass distally than before, irrespective of the presence of an AVF. A steal phenomenon can only appear if there is an inflow stenosis or if the graft has a stenosed area in it. 14 Comparing our values for distal outflow with the data from electro-magnet flow measurement for very distal bypasses and bypasses with distal AVF, the levels are very similar. 2,12 This again suggests that our flow measurements are accurate. In conclusion, the DSAFM gives us the opportunity to study and measure flow in bypasses in vivo and can help to assess the complex haemodynamics of adjunctive arteriovenous fistulae at the distal anastomosis.
O
•
Q•
150
%
I00
%
•
•
00
References
50
Group A
GroupB
GroupC
GroupD
Fig. 4. DSAFM of four groups of difference below knee bypasses; Group A: 10 fem.-pop, byp., Group B: 10 fem.-tib.byp, Group C: 10 fem.-tib, with distal AVF, Group D: 10 pop.-ped.byp, with distal AVF.
day w h e n relatively stable conditions were available. Most published flow studies have been done intraoperatively immediately after the circulation is restored. 3,r,8 This is probably the most unfavourable time because the circulatory balance of the leg is seriously disturbed by the previous ischaemia and the effects of anaesthesia. ~° In spite of these objections, published results of intraoperative flow are very similar to those found in this study. 7,9,. Furthermore these data show that flow depends on the quality of the outflowtract. 7 If a poor outflowtract is suggested preoperatively by the angiogram, a low flow in the bypass can be expected.~ ~ Since low flow predicts an early occlusion, we have increased flow by creating a distal AVF. u,13 This leads to 2-4-fold increase in flow in the bypass. Whether this leads to improved short and longterm results remains to be seen. 1,2 The last part of the study allowed us to investigate the haemodynamics of the distal AVF. Controversy has existed as to the efficacy of an AVF, some authors suggesting that it actually makes things worse by stealing blood from the foot. Unless there is a proximal stenosis or the bypass itself is stenosed the pressure at the distal anastomosis will increase to that of the proximal arterial
1 VANBERGEHENEGOEWENDP, STELZERG, DAUTZENBERGTn, HELM1GL, EHRESMANN U. Pedal and distal lower leg bypasses with a distal arteriovenous fistula. A preliminary report (submitted) 1987. 2 DARDIK H, SUSSMANB, KAHN M, SVOBODAJI, MENDES D, DARDIK L Distal arteriovenous fistula as an adjunct to maintaining arterial and graft patency for limbsalvage. Surgery 1983 ; 94: 478-486. 3 SUMNERD. Application of the electromagnetic flowmeter in reconstructive vascular surgery. In: Rutherford RB ed., Vascular Surgery. Philadelphia: W. B. Satmders, 1977; 135-145. 4 FIEDLER V, SfJMPELMANN R, PETERS P, E. Quantative FluSmesstmg der A. Carotis im Modellversuch sowie bei Patienten durch Kombination yon Sonographie und digitaler Subtraktionsangiographie. In: Digitale Radiographie, Referate und Vortrdge. Konstanz: SchnetzorVerlag, 1984;215-218. 5 FIEOLER V. Densitometrische Bestimmung des relativen Stenosegrades mit einer Handelsfiblichen DSA-Anlage. In: Vogler E, Schneider GH eds. Digitale Bildgebende Verfahren Integrierte digitale Radiologie. Bedim Schering, 1986: 803-807. 6 HmGINS CAB, NORRIS SHL, GERBERKH, SLUTSKYRH et al. Quantitation of left ventricular dimensions and function by digital video substraction angiography. Radiology 1982 ; 144: 461-469. 7 BARNERHB, JUDDDR, KAISERGC, WILLMAN VL, HANLONCR. Blood flow in femoropopliteal bypass vein grafts. Arch Surg 1968 ;96: 619-627. 8 BARNERHB, KAMINSKIDL, CODDJE, KAISERGC, rVVILLMANVL. Hemodynamics of autogenous femoropopliteal bypass. Arch Surg 1974; 109:291-293. 9 MUNOTr~El), DARLINGRC, MORANJM, BtJCKLEYM], LINTON RR, AUSTEN WG. Quantitative correlation of distal arterial outflow and patency of femoropopliteal reversed saphenous vein grafts with intraoperative flow and pressure measurements. Surgery 1969;65: 197-206. 10 TERRYHJ, ALLANIS, TAYLORGW. The relationship between bloodflow and failure of femoropopliteal reconstructive arterial surgery Br]Surg t 9 7 2 ; 5 9 : 5 4 9 - 5 5 !. 11 CAePELENC JR HALL KV. Electromagnetic flowmetry in clinical sargery. Acta Chir Scand [(Suppl) l 1967; 368 : 3-27. 12 DEANRH, YAO IST, STANTONPE, BERGANJJ. Prognostic indicators in femoropopliteal reconstructions. Arch Surg 1975 ; 110:1287--1293. 13 LITrr.E JM, SHEIL AGR, LOEWENTHALJ, GOODMANAH. Prognostic value of intraoperative blood-flow measurements in femoro-popliteal bypass vein-grafts. Lancet 1968 ; 2: 648. 14 BILLET A, Q U ~ L LA, POLITO WF, DAGHER FJ. The vascular steal phenomenon: An experimental model. Surgery 1984; 96: 923-928. Eurj Vasc Surg Vol 1, August 1987
234
G. Stelzer and D. P. van Berge H e n e g o u w e n
I 5 MENOZIANJO, LA MORTE WW, CANTELMONL, DOYLE], SIDAWVAN, SAVENORA, The preoperative angiogram as a predictor of peripheral vascular runoff. Am ] Surg 1 9 8 5 : 1 5 0 : 3 4 6 - 3 5 2 . 10 Bt)Rscn J, ]oils R, HEINTZEN PH, Untersuchungen zur Gfiltigkeit des Lambert-Beerschen Gesetzes bei r6ntgenologischen Kontrastmittelmessungen. Fortschr. R6ntgenstr 1969; 112 : 2 5 9 - 2 6 6 .
EurJ Vasc Surg Vol 1, A u g u s t 1987
] 7 BUCHBINDERD, PASCHAR, VERTAM], ROLLINSDL, RYAN TJ, SCHULER JJ, FLANIGAN DP. Ankle Bypass: Should we Go the Distance? Am ] Surg 1 9 8 5 ; 1 5 0 : 2 1 6 - 2 1 9 .
Received 15 April 1987