From the Department of Vascular Surgery, Sint Franciscus Hospital,aand the
Department of Surgery, Isala Klinieken, location Weezenlanden Hospital.b
Competition of interest: nil.
Reprint requests: C. H. A. Wittens, MD, PhD, Sint Franciscus Hospital,
Department of Vascular Surgery, Kleiweg 500, 3045 PM Rotterdam (e-
Copyright © 2001 by The Society for Vascular Surgery and The American
Association for Vascular Surgery.
0741-5214/2001/$35.00 + 0
Endoscopic versus open subfascial division of
incompetent perforating veins in the treatment
of venous leg ulceration: Long-term follow-up
Johannes E. M. Sybrandy, MD,aWijnand B. van Gent, MD,aEngelbertus G. J. M. Pierik, MD, PhD,b
and Cees H. A. Wittens, MD, PhD,aRotterdam and Zwolle, The Netherlands
Purpose: Subfascial division of incompetent perforating veins seems to be a successful treatment for patients with venous
leg ulceration (CEAP 6). For postoperative wound complications, endoscopic techniques are more common than open
subfascial division of incompetent perforating veins (Linton procedure). We investigated the long-term results of ulcer
healing and recurrence rates and compared them with preoperative and postoperative duplex findings.
Methods: Patients with venous ulceration on the medial side of the lower leg were randomly allocated to endoscopic explo-
ration or open exploration by means of the modified Linton approach. Ulcer healing and recurrence rates were documented.
Results: Thirty-nine patients were randomly allocated to exploration, 19 patients to open subfascial division of incom-
petent perforating veins (Linton group), and 20 patients to subfascial endoscopic division of incompetent perforating
veins (SEPS group). During the follow-up period, four patients in the SEPS group died, all of causes other than the
venous leg ulcer. Because of a squamous cell carcinoma that had developed in the venous ulcer, one patient in the SEPS
group underwent a below-knee amputation. In a mean follow-up period of 50.6 months, the venous ulceration of all
18 patients in the Linton group who were available for follow-up initially healed. The recurrence rate in this group was
22% (4 patients). In the SEPS group, the mean follow-up period for 19 patients was 46.1 months, with the ulceration
healing in 17 patients and a recurrence rate of 12% (2 patients). The presence of deep venous incompetence (DVI) did
not influence the recurrence rates (P = .044, Fisher exact test), but it significantly influenced the development of new
incompetent perforating veins (3 of 10 without DVI; 7 of 10 with DVI; P = .011, binomial test).
Conclusion: The long-term follow-up results of the endoscopic division of perforating veins are comparable with those
of the open division of perforating veins (modified Linton procedure). (J Vasc Surg 2001;33:1028-32.)
Venous ulceration was known to be a problem as early
as the fourth century BC. The incidence of venous ulcera-
tion of the lower leg (CEAP class 6) in the current
Western population is 1% to 2%.1
The association between venous disease and the ulcer
on the lower limb was probably made by Hippocrates.2
He was also the first to propose compressive bandaging as
a treatment. Since then, many possible treatments and
procedures have been tested. The procedures, as described
by Linton,3Cockett,4and Dodd,5were often successful in
treating venous ulceration, but also had major disadvan-
tages, such as a high percentage of wound problems.
Subfascial endoscopic division of incompetent perfo-
rating veins (SEPS) in the treatment of leg ulcers is now
successfully used. Since the introduction of endoscopic
techniques, several short- and midterm clinical series have
validated high healing rates and low recurrence rates.6-8
Endoscopic techniques have fewer postoperative wound
complications than the open subfascial division of incom-
petent perforating veins (Linton procedure).7-10
We present long-term results of ulcer healing and
recurrence rates with SEPS and the Linton procedure.
All patients were part of a randomized trial, in which
open (Linton) perforating vein surgery was compared
with the SEPS procedure. All patients had an active (open)
ulcer on the medial part of the lower leg. Patients were
randomly allocated to the Linton or SEPS group by means
of opening sealed envelopes. Stratification occurred for
presence or absence of superficial venous incompetence,
for primary or recurrent ulceration, and for presence or
absence of diabetes mellitus. Patients who had arterial dis-
ease were excluded (ankle/brachial index < 0.8).
Age, sex, duration of the present ulcer period, primary
or recurrent ulceration, and size of the ulcer were docu-
mented for each patient. All patients underwent a physical
examination. Duplex ultrasonography was performed
before the operation and 6 weeks and 48 months after
surgery. Duplex ultrasonography was performed with a
Philips P700 and a 7.5-MHz linear array transducer
(Philips Medical Systems, Santa Ana, Calif). Patients were
examined in a near upright standing position. Reflux was
defined as more than 0.3 seconds of retrograde flow.11,12
The same investigator performed each duplex scan.
Open subfascial exploration was performed by means
of the modified Linton approach. An incision on the
medial side of the lower leg down from the level of Boyd’s
perforator to the medial malleolus was made through the
fascia. All perforating veins in the subfascial compartment
were ligated.13For endoscopic subfascial exploration, a
mediastinoscope was used without a thigh tourniquet.14
When patients had concomitant superficial venous incom-
petence, it was treated by means of flush saphenofemoral
ligation and stripping of the long saphenous vein from the
groin to just below the knee.
Patients were mobilized the first postoperative day and
treated with ambulatory compression therapy until the
ulcer healed. Elastic stockings (class II) were prescribed
for lifetime when ulcer healing occurred. The patients
returned to the outpatient clinic at 1, 2, 6, 12, 24, and 52
weeks and 48 months after surgery.
An unscheduled interim analysis was performed
because of a high incidence of severe wound complications
in the Linton group. The results of this interim analysis
confirmed a highly significant difference in wound com-
plications between the two groups. After these results
were known, it was considered unethical to continue allo-
cating patients to the open subfascial exploration, and the
study was prematurely stopped. Patients who were already
randomized were considered for follow-up.
Ulcer healing and recurrences were scored. Recurrences
were defined as an ulcer on the medial side of the same
lower leg after initial healing.
In a period of 15 months (February 1994 to April
1995), 39 patients were included in the study. Twenty
patients were allocated to the SEPS group, and 19 patients
were allocated to the Linton group. Both groups appeared
well matched for various characteristics (Table I).
None of the patients underwent a second operative
ligation of perforating veins or flush ligation and stripping
of the long or short saphenous vein after the initial opera-
tion. Sclerotherapy was not performed in any of the
patients during follow-up.
All patients had one or more incompetent perforating
veins on preoperative duplex ultrasonography. About half
of the patients (17) had deep vein incompetence, and more
then half of the patients (27) had superficial incompetence.
Table II lists the wound complications and postopera-
tive complications, and Figs 1 and 2 show the recurrence
and healing rates. These were published earlier by Pierik et
al.9The Fisher exact test was used as a means of assessing
differences in proportions; the Mann-Whitney test was
used as a means of assessing differences in length of hospi-
Six weeks after surgery was performed, four patients in
the SEPS group still had incompetent perforating veins,
compared with none of the patients in the Linton group.
The average number of persisting incompetent perforating
veins in these four patients was 1.5, compared with 0.3 for
the complete group. There were no patients with postop-
erative superficial incompetence.
After 48 months, duplex ultrasonography was only per-
formed on 23 patients (11 in the Linton group, 12 in the
JOURNAL OF VASCULAR SURGERY
Volume 33, Number 5
Sybrandy et al
SEPS group; Table IV). There were various reasons why all
patients did not undergo a duplex examination after 48
months, the most important being older age (as old as 94
years; 5), death (4), amputation (1), and the inability to
stand long enough to perform the examination (4).
Five of 11 patients in the Linton group had new incom-
petent perforating veins at the medial side of the lower leg.
In the SEPS group, five of 12 patients had new incompe-
tent perforating veins at the medial side of the lower leg
Table III. Operation data and clinical results
(n = 19)
(n = 20)
Mean operating time
Mean blood loss (mL) 170 (30-300)
Hospital stay (d)
41 (19-70)43 (20-90) NS
NS, Not significant.
Table II. Postoperative wound complications after open
and endoscopic division of perforating veins
Linton (n = 19) SEPS (n = 20) P value
Table I. Patient characteristics
Linton (n = 19)SEPS (n = 20)
Deep venous incompetence
Duration of present ulcer
Total duration of ulceration 140 (9-480)
Table IV. Number of patients with incompetent perfo-
Linton (n = 19) SEPS (n = 20)
6 weeks postoperative
48 months postoperative (new) 45% (5/11)
100% (19/19) 100% (20/20)
0% (0/19)20% (4/20)
(Table V). The development of new incompetent perforat-
ing veins was significantly (P = .024, χ2test) influenced by
the presence of deep venous incompetence. There was no
relation to preoperative superficial incompetence.
Follow-up was complete in all patients, with the
exception of two patients (1 in each group). During a
follow-up period of 5 years (mean follow-up period in the
Linton group, 50.6 months; mean follow-up period in the
SEPS group, 46.1 months), four patients (all in the SEPS
group) died of causes other than venous leg ulceration. A
below-knee amputation was performed in one patient in
the SEPS group, because of a squamous cell carcinoma
that had developed in the venous ulcer.
Healing of the ulcer occurred in all patients who were
seen in follow-up in the Linton group. In the SEPS group,
the ulceration healed within 4 months after surgery in 17 of
20 patients. This difference did not reach statistical signifi-
cance. One ulcer failed to heal (6 months after operation,
the leg was amputated because of a squamous cell carcinoma
in the ulcer). The other two patients had ulcers that healed
after more than 1 year (Fig 1). Both patients had incompe-
tent perforating veins postoperatively, and both patients had
recurrent ulceration at the medial side of the lower leg.
Recurrences occurred in four patients in the Linton
group (22%) and in two patients in the SEPS group (12%;
JOURNAL OF VASCULAR SURGERY
1030 Sybrandy et al
Fig 2), which was not statistically significant (P = .044,
Fisher exact test).
We also investigated the relation between deep venous
incompetence and new incompetent perforating veins. Of
the 23 patients who underwent a duplex examination after
48 months, 11 patients had deep venous incompetence.
Deep venous incompetence was defined as reflux in the
popliteal vein, the superficial femoral vein, or both. Of the
patients with deep venous incompetence, 7 of 10 had new
incompetent perforating veins, in contrast to only three of
13 patients without deep venous incompetence (Table
VI). This finding was significant (P = .024, χ2test). Four
patients had persistent incompetent veins because of tech-
nical surgical causes. There was no postoperative superfi-
Surgical treatment of incompetent perforating veins
has been reported to give good results in the healing of
venous ulceration. Various surgical techniques used in the
past have all been abandoned, mainly because of wound
complications.5,6,9,15We started a prospective randomized
trial in February 1994 to determine wound complications,
healing, and recurrence rates after open or endoscopic
division of incompetent perforating veins. This study was
discontinued because there was a high morbidity rate
related to the modified Linton procedure. Therefore, the
anticipated 80 patients were not included, and analyses
were performed on the randomized 39 patients. Although
this is a small number of patients and statistics are there-
fore not optimal, we thought the long-term results were
In the clinical parameters (healing and recurrence),
there were no statistically significant differences. Healing
rates were comparable in both groups, 95% in the SEPS
group and 100% in the Linton group. The recurrence
rates were also comparable, 12% in the SEPS group and
22% in the Linton group. Deep venous incompetence,
superficial venous incompetence, and the number of
incompetent perforating veins, persistent or newly devel-
oped, did not influence the recurrence rate. This study
only analyzed the possible differences between the two
techniques and not the cause of recurrent ulceration. The
small number of patients studied precludes an extensive
extrapolation about the cause of failure.
However, a high number of newly developed perforat-
ing veins were detected in both groups after 48 months, a
Table V. Number of incompetent perforating veins at
medial side of the lower leg
Linton (n = 19)SEPS (n = 20)
6 weeks postoperative
48 months postoperative
0.9 (n = 11)1.3 (n = 12)
Fig 1. Ulcer healing life table. 1, Linton group; 2, SEPS group.
Fig 2. Ulcer recurrence life table. 1, Linton group; 2, SEPS group.
JOURNAL OF VASCULAR SURGERY
Volume 33, Number 5
Sybrandy et al
finding significantly influenced by the presence of deep
venous incompetence. We have no evidence-based explana-
tion for this important finding. It can, however, be theo-
rized that the persistent high venous pressures in the calf
related to deep venous incompetence cause new incompe-
tent perforating veins to develop. Unfortunately, there is no
satisfying procedure available for treating deep venous
incompetence. The results of surgery for deep venous insuf-
ficiency are acceptable only in the hands of a select group of
experienced surgeons.16-19The relation between deep
venous incompetence and development of incompetent
perforating veins needs to be further investigated. All four
patients who died and the one patient who underwent a
below-knee amputation were in the SEPS group. This
could explain the 51⁄2-month, not statistically significant, dif-
ference in the follow-up periods between the two groups.
Six weeks postoperatively, only four patients, all mem-
bers of the SEPS group, still had incompetent perforating
veins on the medial aspect of the lower leg. This finding
correlated with the healing rate published by Pierik et al.9
Although the number of incompetent perforating
veins after 48 months was larger in the SEPS group (1.3
vs 0.9), no significant difference could be determined.
Also, no relation could be shown between incompetent
perforating veins and the recurrence of ulceration.
Although earlier publications on perforator surgery
did show low recurrence rates, this study showed, after 4
years, a recurrence rate of 16.2% (6 of 37). This finding
was also confirmed by Gloviczki et al.8This is still lower
than the recurrence rates after conservative treatment
alone, but it is not as low as expected. This again shows
the necessity of a prospective trial. The high recurrence
rate of incompetent perforating veins is also a concern,
although in this small series it did not show any influence
after 4 years. It is not properly documented whether all
patients in the SEPS group underwent a systematic dissec-
tion of the intermuscular septum. Our own analysis of
whether septum dissection influences the outcome of an
SEPS procedure (J. H. Geselschap, submitted for publica-
tion) and the anatomy study by Mozes and Gloviczki20,21
showed a systematic dissection of the septum to be
mandatory. To minimize the risk of missing perforating
veins, we now always perform this septum dissection.
Another adjustment we have made in the technique is the
preoperative duplex mapping with skin marks at the site of
perforating veins. Furthermore, the material with which
we perform the procedure has been greatly improved. We
now use the video-assisted Olympusscope (Olympus
Winter & Ibe, Hamburg, Germany), which allows
mechanical dissection to be combined with CO2insuffla-
tion. This results in a much better exposure of the subfas-
cial space. When we started using this video-assisted
system, we introduced the use of a tourniquet. We did this
mainly because of the high light-absorbing qualities of red
blood cells. It greatly improved visibility and exposure.
In 1997, Pierik et al9showed that the Linton proce-
dure was obsolete because of wound complications. In this
study, we have presented the long-term follow-up of these
patients. There were no differences in healing or recur-
rence rates in long-term follow-up. We therefore conclude
that, with the knowledge that the morbidity of an SEPS
procedure is significantly less, with equal healing and
recurrence rates, SEPS is preferred to open division of per-
forating veins. The subfascial endoscopic procedure is the
method of choice for dissecting incompetent perforating
veins on the medial side of the lower leg.
We thank I. M. Toonder, RVT, for performing the
1. Nelzen O, Berqvist D, Lindhagen A. The prevalence of chronic lower-
limb ulceration has been underestimated: results of a validated popu-
lation questionnaire. Br J Surg 1996;83:255-8.
2. Adams EF. The genuine works of Hippocrates. London: Sydenham
3. Linton RR. The post thrombotic ulceration of the lower extremity: its
etiology and surgical treatment. Ann Surg 1953;138:415-32.
4. Cockett FB. The pathology and treatment of venous ulcers of the leg.
Br J Surg 1956;43:260-78.
5. Dodd H. The diagnosis and ligation of incompetent perforating veins.
Ann R Coll Surg Engl 1964;34:186-96.
6. Gloviczki P. Subfascial endoscopic perforator surgery: indications and
results [review]. Vasc Med 1999;4:173-80.
7. Sparks SR, Ballard JL, Bergan JJ, Killeen JD. Early benefits of subfas-
cial endoscopic perforator surgery (SEPS) in healing venous ulcers.
Ann Vasc Surg 1997;11:367-73.
8. Gloviczki P, Bergan JJ, Rhodes JM, Canton LG, Harmsen S, Ilstrup
DM. Mid-term results of endoscopic perforator vein interruption for
chronic venous insufficiency: lessons learned from the North
American Subfascial Endoscopic Perforator Surgery Registry. The
North American Study Group. J Vasc Surg 1999;29:489-502.
9. Pierik EGJM, van Urk H, Hop WCJ, Wittens CHA. Endoscopic ver-
sus open subfascial division of incompetent perforating veins in the
treatment of venous leg ulceration; a randomized trial. J Vasc Surg
10. Sato T, Goff CD, Gregory RT, Walter BF, Gayle RG, Parent FN III,
et al. Subfascial perforator vein ablation: comparison of open versus
endoscopic techniques. J Endovasc Surg 1999;6:147-54.
11. Pierik EGJM, Toonder IM, van Urk H, Wittens CHA. Validation of
duplex ultrasonography in detecting competent and incompetent per-
forating veins in patients with venous ulceration of the lower leg. J
Vasc Surg 1997;26:49-52.
12. van Ramshorst B, van Bemmelen PS, Hoeneveld H, Eikelboom BC.
The development of valvular incompetence after deep vein thrombo-
sis: a follow-up study with duplex scanning. J Vasc Surg
13. Browse NL, Burnand KG, Thomas ML. Diseases of the veins: pathol-
ogy, diagnosis and treatment. London: Hodder and Stoughton; 1988
14. Pierik EGJM, van Urk H, Wittens CHA. Endoscopic ligation of per-
forating veins using a mediastinoscope. Surg Laparosc Endosc
15. Stuart WP, Adam DJ, Bradbury AW, Ruckley CV. Subfascial endo-
scopic perforator surgery is associated with significantly less morbidity
Table VI. Relation between DVI and new perforating
veins in 23 patients after 48 months
Patients with new perforating veins
Patients without new perforating veins
JOURNAL OF VASCULAR SURGERY Download full-text
1032 Sybrandy et al
and shorter hospital stay than open operation (Linton’s procedure).
Br J Surg 1997;84:1364-5.
16. Raju S, Hardy JD. Technical options in venous valve reconstruction.
Am J Surg 1997;173:301-7.
17. Gloor B, Lagadier J. Surgical treatment of deep venous reflux. Eur J
Vasc Endovasc Surg 1997;13:98-100.
18. Raju S, Fredericks RK, Neglen PN, Bass JD. Durability of venous
valve reconstruction techniques for “primary” and post-thrombotic
reflux. J Vasc Surg 1996;23:357-67.
19. Masuda EM, Kistner RL. Long-term results of venous valve recon-
struction: a four- to twenty-one-year follow-up. J Vasc Surg
20. Mozes G, Gloviczki P. Surgical anatomy for endoscopic subfascial
division of perforating veins. J Vasc Surg 1996;24:800-8.
21. Geselschap JH, Wittews CMA. The effect of septum dissection during
the SEPS procedure. Submitted.
Submitted Jun 26, 2000; accepted Jan 19, 2001.