Eur J Vasc Endovasc Surg 13, 439-442 (1997)
Residual Arteriovenous Fistulae after "closed" in situ Bypass Grafting:
an Overrated Problem
L. C. van Dijk ~, H. van Urk 2, J. S. Lam6ris ~ and C. H. A. Wittens .3
1Department of Radiology and 2Department of Vascular Surgery of the University Hospital Rotterdam "Dijkzigt" and
3Department of Surgery, St Franciscus Gasthuis, Rotterdam, The Netherlands
Objectives: To prospectively evaluate the incidence and consequences of residual arteriovenous (AV)-fistulae after
"closed" in situ bypass grafting.
Methods: In 34 patients, 35 "closed" in situ bypasses were performed. Postoperative assessment of residual A V-fistulae
and bypass patency was performed with duplex scanning.
Results: Postoperative mortality was 3%. During 35 "closed" in situ bypass procedures 216 side branches were coil
embolised. Postoperatively 39 AV-fistulae were detected (15% of the total number of 216 + 39 =255 side branches). Of
these, 13 (5%) closed spontaneously. Fifteen (6%) remained unchanged and 11 (4%) were treated. In three patients four
asymptomatic residual A V-fistulae were treated. In four patients seven symptomatic A V-fistulae were treated for: decreased
distal bypass flow in one; persistent leg oedema in one; pain and redness of the skin in two. One-year primary patency
was 80% ( S E 8.4%). Residual A V-fistulae were detected in none of six bypass occlusions during follow-up.
Conclusion: Residual A V-fistulae detected following "closed" in situ bypass grafting only need treatment if they are
symptomatic, which is uncommon.
Key words: in situ bypass; Arteriovenous fistulae; Saphenous vein; Endovascular; Coil embolisation.
In situ vein infrainguinal bypass is one of the preferred
techniques for treatment of lower extremity ischaemia.
The standard "open" technique, using ligation of the
side branches via a long skin incision, is associated
with wound complication rates up to 44%. 1'2 To reduce
the number of postoperative wound complications,
"closed" in situ bypass techniques, using peroperative
endovascular coil embolisation of vein side branches,
have been developed. 3'4 A disadvantage of these
"closed" techniques is the high incidence of post-
operative residual arteriovenous (AV)-fistulae. Cikrit
et aI., 5 who used the electronically steerable nitinol
catheter system as first used by Rosenthal et al., 3 re-
ported that 39% of the patients were treated for re-
sidual AV-fistulae after "dosed" in situ bypass grafting.
We have reported that 42% of the patients were treated
for residual AV-fistulae after "closed" in situ bypass
grafting, using a co-axial catheter system in a ran-
domised study comparing the "closed" technique to
the "open" technique. 6 Rosenthal et aI., however, only
* Please address all correspondence to: C. H. A. Wittens, Department
of Surgery, St Franciscus Gasthuis, Kleiweg 500, 3045 PM Rotterdam,
treated 6% residual AV-fistulae after refinement of the
electronically steerable nitinol catheter and increased
operator experience with the "closed" technique. 7
The purpose of this study was to evaluate the in-
cidence and consequences of residual AV-fistulae after
"closed" in situ bypass grafting using a co-axial cath-
eter system for peroperative endovascular coil em-
bolisation of the vein side branches.
Patients and Methods
From June 1992 to March 1996, 35 "closed" in situ
bypasses were performed in 34 patients (23 male, 11
female) in the St Franciscus Gasthuis, Rotterdam. The
mean age was 73 years (range 55-90) and indications
for operation were non-healing ulcers or necrosis in
16, restpain in seven and life-style limiting/disabling
claudication in 11. Fourteen (41%) patients had a smok-
ing history, 17 (50%) had ischaemic heart disease, 17
(50%) had hypertension and six (18%) were diabetic.
Seventeen femorodistal popliteal and 18 femorocrural
bypasses were performed.
The "closed" in situ bypass technique used was
described by Wittens et al. 4 The main characteristics
1078-5884/97/050439 + 04 $12.00/0 © 1997 W.B. Saunders Company Ltd.
L.C. van Dijk et al.
i early postoperative mortality (no AVF)
6 occlusions (no AVF) ]
21 bypasses: 39 AVF
7 bypasses without AVF
Fig. 1. Results of follow-up of 35 bypasses. AVF = AV-fistulae; B = bypass.
of this method are: the valvul0tomy procedure is
performed "blindly" with a disposable variable valve
cutter; the side branches of the vein are coil embolised
via a co-axial embolisation catheter under fluoroscopic
control. The disposable materials needed for this pro-
cedure (valvecutter, embolisation catheter and 10 em-
bolisation coils) are commercially available as a set
(Cook Europe, Denmark).
During the follow-up period duplex scanning was
performed within 4 weeks after the operation, every
3 months postoperatively in the first year and every
6 months thereafter for assessment of graft patency
and occurrence of AV-fistulae. The Kaplan Meier life-
table method was used to estimate the probability of
bypass patency. Primary patency rates were calculated
without taking into account operations and/or in-
terventions for treatment of residual AV-fistula.
The postoperative mortality was 3%: one patient died
in the first week following the operation because of
cardiac failure. In 35 "closed" in situ bypasses a total
number of 216 side branches (mean 6.2 per patient)
were coil embolised during the bypass procedure. The
mean follow-up was 21 months (range 1-48). Bypass
patency rates are shown in Fig. 1. The 1 year primary
patency rate was 80% (s.E. 8.4%). During the follow-
up period six bypasses occluded. In none of these
patients had residual AV-fistulae been detected during
follow-up before occlusion. Postoperatively, in 21 of
the 35 bypasses a total number of 39 AV-fistulae were
detected. The total number of vein side branches de-
tected intraoperatively and postoperatively was there-
fore 255. In other words 15% (39) of the total number
of side branches were only detected postoperatively
13 AVF (12B) spontaneous closure
15 AVF (9B) unchanged
7 AVF (4B) treated
4 AVF (3B) treated
as AV-fistulae. Figure 2 shows the outcome of these
AV-fistulae. A total number of 13 (5%) of these AV-
fistulae in 12 patients closed spontaneously after a
mean follow-up period of 8 months (range 1-30).
Fifteen AV-fistulae (6%) in nine patients remained
unchanged for a mean follow-up period of 13 months
(range 1-22) and 11 (4%) were treated: in three patients
four asymptomatic residual AV-fistulae were treated;
in two patients 1 week and in one patient 1 month
postoperatively. Seven symptomatic AV-fistulae in four
patients were treated. The reasons for treatment were
as follows: two AV-fistulae causing continuous ret-
rograde diastolic flow distally in one patient were
treated 8 months postoperatively with cessation of
diastolic flow. Two AV-fistulae were associated with
persistent leg oedema in one patient and treated suc-
cessfully 3 months postoperatively. Pain and redness
of the skin was caused by three AV-fistulae in two
patients and were treated successfully 2 days post-
All but three residual AV-fistulae were detected at
the first duplex examination. In one patient an AV-
fistula was detected at the first examination and three
additional AV-fistulae were detected 2 months later.
These four AV-fistulae remained asymptomatic for the
follow-up period of 23 months.
Management of residual AV-fistulae after in situ bypass
grafting is a controversial issue. Symptoms attributed
to residual AV-fistulae include bypass occlusion,
diminished bypass function, leg oedema and skin
symptoms. 8 In this study no bypass occlusion was
associated with detected residual AV-fistulae. How-
ever, other investigators have reported that residual
AV-fistulae can cause bypass occlusion, especially the
Eur J Vase Enctovasc Surg Vol 13, May 1997
AV-fistulae After "Closed" in situ Bypass
(22) (18) (14) (10) (61
3 6 9 12
15 18 21 24 27 30
Fig. 2. Primary patency rates. Numbers between brackets indicate bypasses at risk.
segment distal of the AV-fistula. 9-11 In this study duplex
scanning showed continuous retrograde diastolic
blood flow in a bypass segment distal of two large
residual AV-fistulae in one patient. Treatment of these
AV-fistulae restored normal blood flow. Leopold et al.
reported similar experiences treating impaired bypass
function, without occlusion, by closure of residual AV-
fistulae. 12 In one patient in our study, persistent leg
oedema was associated with residual AV-fistulae and
treated successfully by AV-fistula ligation. Chang et al.
reported successful treatment of persistent leg oedema
by closure of residual AV-fistulae in four patients after
216 in situ bypasses. 8
Besides bypass failure and leg oedema, other symp-
toms attributed to residual AV-fistulae include redness
and tenderness of a focal skin area. These AV-fistulae
with a connection to the skin tend to occlude spon-
taneously, s Therefore a conservative attitude towards
these AV-fistulae is a good option. We started by
obliterating all residual AV-fistulae in patients suf-
fering skin redness and pain. However, after two
patients we changed our policy to a more conservative
attitude and treated skin symptoms by local ap-
plication of a compressive bandage. Symptoms sub-
sided in all patients within 1 week. Accordingly, we
did not close any other residual AV-fistulae for this
reason. Leather et al. reported only 1 patient with skin
necrosis after 1038 in situ bypasses. 13 In the early
phase after the introduction of "closed" in situ bypass
grafting, we treated three patients for AV-fistulae that
were asymptomatic. This policy was also changed to
a more conservative approach, after which no
asymptomatic residual AV-fistulae were treated any
The incidence of residual AV-fistulae detected in this
study is rather high. Rosenthal et al. reported that 6%
(three patients) of the patients were postoperatively
treated for residual AV-fistulae with a diameter larger
than 2 mm. Numbers of smaller residual AV-fistulae
were not mentioned. Rosenthal et al.'s group had coil
embolised 258 side branches during 53 "closed" in situ
bypass procedures (average of 4.9 side branches per
bypass). 6 In this study we coil embolised an average
of 6.2 side branches per bypass operation. We think that
the technical differences of the embolisation catheters are
not a logical explanation for the difference in residual
AV-fistula incidence. Theoretically the electronically
steerable nitinol catheter as used by Rosenthal et al. (CRI,
Catheter Research, Inc., Indianapolis, IN, U.S.A.) is ideal
for selective catheterisation of difficult angled side
branches. We learned that with a co-axial system (Cook
Europe, Denmark)it is also possible to selectively cath-
eterise difficult angled side branches. A probable ex-
planation for the different incidence of detected residual
AV-fistulae could be that we also detected AV-fistulae
smaller than 2 ram. Small tributaries might have been
missed during intraoperative fluoroscopy due to com-
petent valves at the orifice. These sufficient valves can
dilate under arterial pressure and become incompetent
resulting in residual AV-fistulae.
After gaining some experience with the "closed" tech-
nique, we developed a policy not to aim at closure of all
vein side branches in all cases: side branches that were
very small or were very difficult to catheterise and coil
Eur J Vasc Endovasc Surg Vol 13, May 1997
L.C. van Dijk et aL
embolise within 10 minutes per-operatively were left
open, because we felt that these side branches were
usually of no consequence. An advantage of this policy
is that the peroperative embolisation procedure took
only 20-40 min, since no time was lost with prolonged
attempts for catheterisation of small side branches and/
or additional skin incisions for surgical ligation. In this
series one-third of the AV-fistulae occluded spon-
taneously and 15 of the persistent fistulae remained
asymptomatic during follow-up. Only four patients
with residual AV-fistulae were treated for symptoms
and the two with local skin redness and pain could
probably have been treated conservatively. Therefore it
is reasonable to assume that only in two (6%) of the 35
bypasses did residual AV-fistulae need treatment. No
AV-fistula related occlusions occurred and bypass pat-
ency rates were comparable to other series of "closed"
and "open" in situ bypasses. 6'~4'15 Because asymptomatic
AV-fistulae do not threaten bypass patency, specific fol-
low-up for AV-fistula detection seems unnecessary.
In conclusion, we have shown that the incidence of
detected residual AV-fistulae after "closed" in situ by-
pass grafting was 15% of the total number of vein side
branches. When residual AV-fistulae are detected after
"closed" in situ bypass grafting, they only need treat-
ment in a minority of patients that are symptomatic.
1 REIFSNYDER % BANDYK D, SEABROOK G, KINNEY E, TOWNE JB.
Wound complications of the in situ saphenous vein bypass
technique. J Vasc Surg 1992; 15: 843-850.
2 SCHWARTZ ME, HARRINGTON EB, SCHANZER H. Wound com-
plications after in situ bypass. ] Vasc Surg 1988; 7: 802-807.
3 ROSENTHAL D, HERRING MB, O'DONOVAN TG, CIKRIT DF, COMER-
OTA AH, CORSON JD. Endovascular infrainguinal in situ sa-
phenous vein bypass: a multicenter preliminary report. J Vasc
Surg 1992; 16: 453-458.
4 WITTENS CHA, VAN DIJK LC, Du BOlS NAJJ, VAN URK H. A new
"closed" in situ vein bypass technique. Eur J Vasc Surg 1994; 8:
5 CIKRIT DF, DALSING MC, LALKA SG et al. Early results of en-
dovascular-assisted in situ saphenous vein bypass grafting. J
Vasc Surg 1994; 19: 778-787.
6 VAN DIJK LC, VAN URK H, Du BOlS NAJJ et al. A new "closed"
in situ vein bypass technique results in a reduced wound com-
plication rate. Eur J Vasc Endovasc Surg I995; 10: 162-167.
7 ROSENTHAL D, DICKSON C, RODRIGUEZ FJ et al. Infrainguinal
endovascular in situ saphenous vein bypass: ongoing results. J
Vasc Surg 1994; 20: 389-395.
8 CH~NC BB, LEOPOLD PW, KUPINSKY AM, KAUFMAN JL, LEATHER
RP, S~AH DM. In situ bypass hemodynamics: the effects of
residual A-V fistula. J Cardiovasc Surg 1989; 30: 843-847.
9 SHEARMAN CP/GANNON MX, GWYNN BR, SIMMS MH. A clinical
method for the detection of arteriovenous fistulae during in situ
great saphenous vein bypass. J Vasc Surg 1986; 4: 578-581.
10 GALLAND RB, YOUNG AE, JAMIESON CW. In situ vein bypass. A
modified technique. Ann R Coil Engt 1981; 63: 186-188.
11 DUNDAS P. The in situ vein bypass. ] Cardiovasc Surg 1970; 11:
12 LEOrOLD PW, KUPINSKY a]~, CORSON JD et aL Hemodynamic
observations related to in situ bypass arteriovenous fistulae. Vasc
Surg 1987; 21: 265-270.
13 LEATHER RP, SHAH DM, KARMODY AM. Infrapopliteal arterial
bypass for limb salvage. Increased patency and utilization of the
saphenous vein used in situ. Surgery 1981; 290: 1000-1008.
14 DONALDSON MC, MANNICK JA, WHITTEMORE AD. Femoral distal
bypass with in situ saphenous vein: long term results using the
Mills valvulotome. Ann Surg 1991; 213: 115-121.
15 LEATHER RP, SHAH DJ, CraANC BB, KAUFMAN JL. Resurrection
of the in situ saphenous vein bypass 1000 cases later. Ann Surg
1988; 208: 435-442.
Accepted 11 September 1996
Eur J Vasc Endovasc Surg Vol 13, May 1997