Randomized trial of stripping versus high ligation combined with sclerotherapy in the treatment of the incompetent greater saphenous vein

Randomized Trial of Stripping Versus High Ligation Combined With Sclerotherapy in the Treatment of the Incompetent Greater Saphenous Vein Peter H. Rutgers, MD, Peter J.E.H.M. Kitslaar, MD, Maastricht. The Netherkmds

This prospective randonfized sttidy c o m l m r e d Ihe l r e a t m e n l of g r e a l e r saphenous vein hlsul~eieney by sli-ipphlg and local a~adsions of varicose veins with high ligation of the saphenofcmor,-d jmlction (crossectomy) e o m b h l e d ~4th s e l e r o c o m p r e s s i o n therapy. O f 156 consecutive patients, 8 9 legs were randondy allocated to stripping and 92 to Ifigh ligation. At follow-up of 3 months and 1, 2, and 3 years after t r e a t m e n t , clitfieal aud D o p p l e r ultrasound restdts, and c o m plaints and cos,netic restdts, as judged by the imtient and the surgeon, were scored. At 3 years, 6 9 limbs ht the stripping g r o u p ( 7 8 % ) and 73 limbs in the ligation g r o u p ( 7 9 % ) were available to follow-up. T h e cosmetic restdts, both judged by the patient and the surgeon, were siglfifieantly better (P <0.05) hi the stripped limbs than in the limbs with Ifigh ligation and sclerotherapy. Cliifical and D o p p l e r ultrasotmd evidence of reverse flow in the saphenous vein was sig~fificmltly less (P < 0 . 0 0 1 ) alier the stripphlg operation. The results o1' I r e a t m e n l of isolated saphenous vehl insufficiency by stripping operation, therefore, were s u p e r i o r to those obtained by high ligatioq c o m b i n e d with sclerotherapy. many year's, both operative treatment and sclehave found use in the treatment of varicose Fveins,~orrotherapy operative treatment being the oldest (Celsus 25 BC to AD 50), injection therapy having had its introduction in 1853 by Cassaignac. I The popularity of each therapy has varied throughout time, and disagreement still persists concerning the indication lot"each form of treatment. The stripping operation became popular in the early 1950s. Before that time, a high saphenous ligation with retrograde injection of sclerosant was commonly performed, resulting in a high reculTence rate. The stripping operation was founded on the pretnise that insufficiency in a vein that has been removed cannot recur.-" Following the modification of sclerotherapy by Fegan, ~ this therapy found renewed widespread use. In several publications, results of sclerotherapy comparable with those of operative treat-

From the Department of Surgery, University Hospital. Maastricht. Slate University Limburg. The Netherlands. Requests for reprints should be addressed to Peter J.E.H.M. Kitslaar, Department of Surgery. University Hospital. P.O. Box 5800. 6202 AZ Maastricht, The Netherlands. M:muscript submitted February 17, 1993. and accepted in revised form August 5. 1993.

ment have been reportedY' In 1974, Hobbs 7 demonstrated that the best primary treatment for dilated superficial veins and for incompetent perlb.'ating veins in the lower part of the legs was injection compression therapy. Surgery was more successfttl and long-lasting when there was involvement of either of the saphenous systems with junctional incompetence. It was demonstrated that the period of observation after treatment was of importance: unsatisfactory results were a function of the period of observation. 7.s The long-term results of Fegan's method were uncertain but the cost-saving was so great that Doran and White '; have advised that it be used always as the first procedure. In judging the result of treatment, there was only a 40% agreement between surgeon and patient as to what constituted a good result) ° Despite well-defined criteria, two or more investigators may evaluate a given result differently) ~ Jakobsen ~2 concluded that the primary treatment lot" incornpetence of the saphenous vein should be a radical operation. Rivlin ~3 fotmd stripping to the ankle unnecessary and advised a more selective operation of the proximal saphenons vein, and recently trials have been published in which selective stripping compared favorably with standard stripping. H'j5 Long-saphenous-vein-saving surgery Ibr varicose veins has been advised based on the assumption that removal of the saphenous vein is unnecessat'y if insufficient perforating veins have been ligated. ~' This prospective study was started to investigate whether treatment of incompetence of the greater saphenous vein combined with local varicosities by standard stripping and avulsion of varicosities gives medium-term results comparable with the results of high ligation combined with sclerotherapy of tributaries. In order to study the effect of both treatments on local henaodynamics, standard noninvasive tests were used during follow-up.

PATIENTS AND METHODS Both limbs of 268 consecutive patients (536 lirnbs) presenting at our surgical outpatient departrnent with vat'ices were examined for this study. Complaints tot" each of the limbs were registered separately. Exatnination consisted of clinical and Doppler ultrasound testing. All clinical investigations were performed by two experienced vascular surgeons and consisted of inspection, palpation of the varicosities, atld tourniquet tests. The objective of the tourniquet tests was to find retrograde filling of the greater saphenous vein on release of a high thigh tourniquet. The Doppler ultrasound investigations were pertbnned with a bidirectional continuous-wave Doppler device, which was the usual ultrasound equipment for venous testing at the time this long-term study was stalled. The re-

THE AMERICAN JOURNAL OF SURGERY"

VOLUME 168 OCTOBER 1994

311

INCOMPETENT SAP|tENOUS VEIN/RUTGERS AND KITSLAAR

stilting signals were judged acoustically by two experienced vascular laboratory technicians. During exanaination, the patient was in the supine position for the examination of the fernoral veins and the posterior tibial veins. The popliteal veins were tested in a prone position. The erect position was used for examination of the greater saphenous vein at the saphenofemoral junction and along its course to the below-knee level and for the examination of the lesser saphenous vein. During proximal muscle compression, reflux can be heard if there is valvular inco.npetence. Occasionally, a short refluxing signal can be heard (less than I second), which was not considered as being pathologic in nature. Reflux at the saphenofemoral jtmction was tested by the Valsalva maneuver. Based on these tests, 181 legs, belonging to 156 patients, were considered to have an isolated incornpetence of the greater saphenous vein combined with local varicosities. A total of 78 patients (89 legs) were randomized to the standard stripping operation combined with surgical avulsion of varicosities, and 78 patients (92 legs) to high ligation of the saphenofemoral junction (in Europe often termed crossectomy) combined with sclerocompression therapy of the varicosities. The randonaization procedure was per|brined very rigidly and irrespective of the size of the varices. For practical reasons, both limbs of patients with bilateral greater saphenous vein incompetence were treated in a similar manner (11 patients in the stripping group and 14 patients in the high ligation sclerotherapy group), which accounts for the difference in treated limbs in both groups. The stripping operation is pedbrmed under general anesthesia. Through groin incisions, the proximal tributaries of the long saphenous vein are ligated and divided. The saphenous vein is divided flush at the saphenofemoral junction. Then an incision is made directly anterior to the medial malleolus at the ankle, exposing the saphenous vein. After distal introduction of the stripper, the vein is ligated and divided. Care is taken to identit3, and isolate the saphenous nerve from the adventitia of the vein. Local varicosities are avulsed through multiple stab incisions, and stripping is then pedbrmed from ankle to groin. After wound closure, compressive bandages are applied. One day alter the operation, these bandages are changed for compressive stockings. Stockings are maintained lbr 4 weeks. Hospital admission usually lasts 2 to 3 days. The saphenofemoral ligation in the high ligation-sclerotherapy group is performed as an outpatient procedure under local infiltration analgesia. Saphenofemoral ligation is done in the same way as described in the stripping group. Preceding the ligation, the injection sclerotherapy is performed as described by Fegan but slightly rnodified, using a I% solution of ethoxysclerol. Injection sites are selected and marked at variable intervals depending on the size of the varices. With the patient standing, the needie is inserted into the filled vein, which is then emptied by digital pressure. Immediately after injection, tbam rubbet" pads are applied over the injection sites and adjacent veins and covered by a layer of compressive bandages. The maximurn dose of ethoxysclerol used was 12 mL per patient per session. In both patient groups, residual or recurrent varicosities during follow-up were treated by sclerotherapy if requested 312

THE AMERICAN JOURNAL OF SURGERY'

by the patient. Both treatment groups were cornparable with regard to sex and age characteristics. A total of 75% of patients were female, in both groups. Twenty-five pet'cent of the patients had a standing occupation, whereas 75% had sitting work. A family history of varicosis was present in 76% of the patients. The main reasons for seeking treatment were the same in the male and female patients, with a not significantly different proportional occurrence among both sexes, For the majority of patients, cosmetic reasons were most prorninent (75% in men, 95% in women), followed by pain in the legs (60% in men, 63% in women) and cranaps at night (38% in men and in women). Follow-up lit 3 months and I, 2, and 3 years after treatment consisted of a structured interview, clinical investigation, and Doppler ultrasound testing. Patient complaints were noted and compared with the complaints belore treatment by asking the patients to classify them as absent, unchanged, or worse. Cosmetic restdts were judged by the patients as excellent, moderate, or poor. Cosmetic results were also rated by one surgeon not involved in the initial treatment according to the tollowing criteria: (I) no visible or palpable varicose veins (excellent)" (2) visible or palpable varicose veins less than 5 mm in diameter (fair); and (3) varicose veins with a diameter of more than 5 mm or visible incompetent main trunks (poor). Results according to complaints or cosmetic assessment were finally classified as "good" (excellent or fair cosinesis, or absence of complaints) or "'bad" (cosmetic assessment poor, or complaints unchanged or worse). Clinical and Doppler tests were done to demonstrate the absence or presence of reverse flow at the saphenofemoral junction and in the region of the normal course of the greater saphenous vein. At 3 years, 69 limbs in the stripping group (78%) and 73 limbs in the high ligation-sclerotherapy g,'oup (79%) were still available for a full clinical and heqlodynamic evaluation. The patients not attending at the scheduled follow-up were contacted by telephone if possible. The majority indicated that they had not sought further treatment because of an absence of symptoms and none of them had been treated elsewhere lbr varices. STATISTICAL ANALYSIS The percentages of symptomatic and cosmetic lhilures and of clinical and Ilemodynanlic recun'ences of incompetence of the greater saphenous vein for both treatment modalities were compared by simple chi-sqtmre tests. The relation between Doppler evidence of persisting reflux over the saphenofemoral junction and along the course of the saphenous vein lbr both treatment modalities, on the one hand, and complaints, cosmetic assessment by the patients, and cosmetic assessment by the surgeon, on the other hand, was analyzed by means of Fisher's exact test. The initial allocation of patients to one of the treatment groups was maintained throughout the study, according to the "'intention to treat" principle. RESULTS From the 92 legs in the high ligation-sclerotherapy group, 20 received their sclerotherapy in 2 sessions and 3 legs un-

VOLUME 168 OCTOBER 1994

INCOMPETENI" SAI)IIENOUS VEIN/RUTGERS AND KITSL, AAR

High ligation

Stripping

% loo Free o f complaints

50 0 100

Good cosmesis patient Good cosmesis surgeon

50 0 100

Figure 1. Proportion of legs in patients who were free of complaints with good cosmetic results as judged by the patient and as judged by the surgeon at 3 months and 1, 2, and 3 years after treatment for the stripping group and for the high ligat~on-sclerotherapy group.

50 0 3

12

24

36

3

12

24

36

Months

derwent 3 sclerotherapy treatments. Five of the 92 legs had to be treated surgically during the observation period. Indication Ibr re-operation was incompetence of the lesser saphenous vein once. a remaining greater saphenous vein insufficiency three times, and once an exploration of the groin was performed for suspected saphenofernoral reflux that proved negative. In the stripping group of 89 legs, additional sclerotherapy was necessary in 12 limbs on patients' request during the follow-up period. This sclerotherapy was performed at a mean interval of I 0 months following operation. In 2 cases, a re-operation was needed. Both times the indication for reoperation was incompetence of the lesser saphenous vein resulting in persisting complaints. Stripping resulted in darnage to the saphenous nerve in 27 limbs (33%), causing sensory loss or a distressing traumatic neuritis invoh,ing the medial aspect of the ankle. In the majority of the cases, the neuralgia persisted for 1 year: in 4 cases, however, it persisted up to 3 years. In Figures 1 and 2, results of follow-up are given at 3 months and at I, 2, and 3 years after treatment. Chi-square testing showed significant (P <0.05) differences for good cosmetic results as judged by the patients themselves (54% versus 72%) as well as by the surgeon (39% versus 61%), and for clinical (47% versus 10%) and Doppler ultrasound (46% versus 15%) evidence of reverse flow in the saphenovs vein at 3 years. In all cases, the results were better for the stripped legs th:m for the legs treated by high ligation-sclerotherapy. No statistical differences could be found between the two groups concerning posttreatment fi'eedom of complaints at any follow-up time. By means of Fisher's exact test. a significant association (P <0.05) between the surgeon's assessment of the cosmetic r e s u l t s and Doppler evidence of saphenous reflux could be demonstrated at 3 years of follow-up in the high ligationsclerotherapy group only. There appeared to be no association between saphenous reflux and complaints or the patient's cosmetic evaluation in either of the two treatment groups at 3 years of follow-up.

High

ligation

% Ioo Clinical testing

.

Stripping .

.

.F

.

so 0

CW Doppler testing

I

I00

s0

~

~

o

3

12

24

ss

I s

12

24

II 36

Months

Figure 2. Proportfon of legs with clinical evidence and with Doppler ultrasound evidence of residual or recurrent reflux in the greater saphenous vein at 3 months and 1, 2, and 3 years after treatment for the stripping group and for the high ligation-sclerotherapy group.

COMMENTS The objectives of radical treatment of varicose veins are the elimination of points of reflux into vein trunks and of dilated tributaries. Only two studies have been published in which treatment of varicose veins entirely by sclerotherapy was compared with an exclusively surgical treatment. 7L' Both studies were performed before the era of noninvasive functional venous testing. In these studies, the classification of patients and the evaluation of treatment were based on clinical grounds only. which one would consider insufficient at the present time. In 1968, Hobbs j7 reported that, at 2 years, the results of sclerotherapy for various forms of superficial venous valvular incompetence cornpared favorably with those of surgical treatment. In 1974, however, he reported that the I-year "cure" rate of 82% in the sclerotherapy group had fallen to a mere 7% at 5 years and only 30% still remained "'improved." In contrast, 20% of the surgical group remained "'cured" and 80% were still "'improved'" at 5 years. 7 In Jakobsen's study, ~2 which included only saphenous varicose veins, the objective clinical success of sclerotherapy (37%) was significantly less than the success of radical surgery (90%) at the

THE AMERICAN JOURNAL OF SURGERY"

VOLUME 168

OCTOBER 1994

3]3

INCOMPETENT SAPHENOUS VEIN/RUTGERS AND KrFSLAAR

end of a 3-year follow-up period. The subjective appreciation of success by the patient was less divergent (70% versus 93%). The poor long-term success of treatment of incompetent saphenous veins solely by injection-cornpression therapy does not warrant its application. ~8 The question as to which is the best treatment for long saphenous vein insufficiency has been addressed in three randornized studies, comparing stripping of the vein with high ligation/2,~6 Both treatment modalities were combined with surgical ligation of perforators in all three studies and with either surgical avulsion Ls''¢'or injection-sclerotherapy of local varicosities) ~-The studies did not lead to uniform conclusions. The patient's appreciation of long-term successes were equal for both methods of treatment in all three studies, partly because of a bias against stripping due to a high incidence of paraesthesia in the limbs treated with long stripping fi'om the ankle to the groin. The surgeon's assessment, based on the observed extent of persisting or recurrent varicosities, favored stripping in two of the studies and was equal for both methods in the other study. In our study, combining sclerotherapy with high saphenous ligation resulted in patient satisfaction in over 90% of patients. However, after 3 years, the Doppler ultrasound examination showed that in almost 50% of these patients, evidence that saphenous reflux remains or has redeveloped exists. In this way, the intention of the treatment (ie, abolishing the reflux) can be considered a failure. Persisting reflux apparently does not influence the patient's appreciation of cosmetic results and plays no role in persisting complaints. A possible explanation for the difference in results between the two forms of therapy is the fact that the high ligation in the stripping group was performed under general anesthesia, perhaps allowing a better groin exposure and accordingly a more radical ligation of the saphenous vein and its tributaries. Most authors stress the importance of a radical elimination of all side branches during a high saphenous ligation. The traditional total stripping of the greater saphenous vein or ligation of the saphenofemoral junction was based on the assumption that reflux invariably occurs at this junction. The assumption of the saphenofemoral escape as the sole source of reflux into the saphenous system now seems incorrect. LsRecently, in only 64% of truncal varicosities, escape points were found to be localized at the saphenofemoral junction; in the remainder, proximal escape points were localized elsewhere and of nonjunctional nature.~-sThese nonjunctional escapes are missed by high ligation, whereas they are in fact treated by radical as well as selective stripping. Stripping enables the incompetent thigh communicating veins to be removed together with the greater saphenous vein. ]3 The essential step in the treatment of long saphenous vein insufficiency is not the removal of the saphenous trunk as such, but the elimination of junctional incompetence (by high ligation) and of escape via thigh perforators (by local surgery or stripping of the saphenous vein). ~'~'~6The 33% damage to the saphenous nerve in our series is alarming but not different from findings by other investigators. 19.2°When this trial was started, it was the practice that the vein be stripped from ankle to groin. Now, in our clinic 314

like in most others, full-length stripping has been abandoned in favor of knee-to-groin stripping. ~3'~4'2~ The period of observation might be short in our series. Although in both groups the number of successful results declined during the 3 years of follow-up, the results remained significantly better for the stripping group compared with the high ligation-sclerotherapy group. As Jakobsen ~-' notes, there is no reason to assume that these relationships should change after a longer period of follow-up. CONCLUSION The medium-term hemodynamic and cosmetic results of treatment of isolated insufficiency of the greater saphenous vein by the stripping operation are superior to those obtained by high ligation combined with sclerotherapy. REFERENCES I. Rowden Foote R. Varicose Veil,s. London: Butterworth & Co;

1952:I-2 I. 2. Lofgren ET. The treatment of long saphenous varicosities and their recurrence: a long term follow-up. In: Bergan JJ, Yao JST, eds. Surger).' of the Veins. Orlando: Grune & Stranom 1985:285-299. 3. Fegan WG. Continuous compression technique of injecting varicose veins. I_zmcet. 1963:2:109-I 12. 4. Reid RG, Rothnie NG. Treatment of varicose veins by compression sclerotherapy. Br J Surg. 1968;55:889-895. 5. Stother IG. Bryson A. Alexander S. The treatment of varicose veins by compression sclerotherapy. Br J Surg. 1974;61:387-390. 6. Sigg K. Treatment of varicose veins by injection sclerotherapy:a method practised in Switzerland. In: Hobbs JT. ed. The Treatment o.[Vemms Disorders. Lancaster: MTP Press Limited; 1977:113-138. 7. Hobbs JT. Surgery and sclerotherapy in the treatment of varicose veins. Arch Surg. 1974;109:793-796. 8. Van der Stricht J. Technik und Indikationen der Chirurgischen Behandlung von Varizen und ihrer Kombination mit Vertidung. Zemra/b/ PhleboL 1966;5:39---45. 9. Doran FSA, White M. A clinical trial designed to discover if primary treatment of varicose veins should be by Fegan's method or by an operation. Br J Surg. 1975;62:72-77. 10. Raj TB, Makin GS. A randomised controlled trial of two lorms of compression bandaging in outpatient sclerotherapy of varicose veins. J Stlrg Res. 1981:31:440 ~44. I I. McAdam WAF. Hon'ocks JC, DeDombal FT. Assessment of the results of surgery for varicose veins. Br J Stlrg. 1976;63:137-140. 12. Jakobsen HJ. The value of different forms of treatment for varicose veins. Br J Sttt'g. 1979;66:182-184. 13. Rivlin S. The surgical cure of primary varicose veins. Br J Surg. 1975;62:913-917. 14. Koyano K, Sakaguchi S. Selective stripping operation based on Doppler ultrasonic findings lbr primary varicose veins of the lower extremities. SttrgetT. 1988; 103:615-619. 15, Munn SR, Morton JB, Macbeth WAAG, McLeish AR. To strip or not to strip the long saphenous vein? A varicose vein trial. Br J Surg. 1981:68:426-428. 16. Hammarsten J, Pedersen P, Cederlund CG, Carnpanello M. Lon8 saphenous vein saving surgery for varicose veins. A long term lbllow-up. Eur J Vase Surg. 1990;4:361-364. 17. Hobbs JT. Treatment of varicose veins. A random trial of injection compression therapy versus surgery. Br J Surg. 1968;55: 777-779. 18. Eklof B. Modern treatment of varicose veins. Br J Sttrg. 1988:75:297-298. 19. Jacobsen BH,Wallin L. Proximal or distal extraction of the internal saphenous vein? Vasa. 1975;4:240-242.

THE AMERICAN JOURNAL OF SURGERY" VOLUME 168 OCTOBER 1994

INCOMPETENT SAPHENOUS VEIN/RUTGERS AND KITSLAAR

20. Keith LM, Smead WL. Saphenous vein stripping and its complications. Surt~ Clin North Am. 1983;63:13(]3-1312. 21. Ramsasty SS, Dick GO, Futrell JW. Anatomy of the saphenous nerve: relevance to s,'lphenous vein stripping. Am J Sm'q. 1987;53:274-277. EDITORIAL

COMMENT

Primary varicose vein disease is widely prevalent, and although rarely leading to serious consequences, the local discomfort and cosmetic concerns atnply justify treatment. Treatment, however, should be as minimally invasive and cost-effective as possible, consistent with extended relief" and an acceptable cosmetic result. The European surgical community has a long and rich experience treating venous disease and Drs. Rutgers and Kitslaat" from Maastricht offer valuable insights in a randomized trial of total saphenous stripping compared to high saphenous ligation and variceal sclerotherapy. The authors randomized patients with major saphenous incompetence, a subgroup of only 181 of 536 legs (34%) with varicosities. On follow-up, the patient satisfaction was

rated excellent in both groups. This result is not surprising, however, because variceal excision or sclerotherapy alone is commonly recognized to be sufficient treatment for most patients, even with saphenous incompetence, and the occasional need for later local retreatment is not catastrophic. Although stripping proved to be a more effective control of reflux than high ligation, there was a disturbing incidence of saphenous neuropathy. It is worth emphasizing that the authors themselves no longer strip the entire saphenous vein, as evaluated in this study, but only strip the saphenous vein above the knee. Although reflux may not be alleviated by saphenous-sparing procedures, and saphenous reflux may admittedly be a major contributing factor in the tbrmation and retbrmation of vat-ices, empirically it is the varices themselves that are the source of the cosmetic concern and local discomfort. This paper should convince us that high ligation of the saphenous vein does not have an identified role in the treatment of varices, and in the great majority of patients treating the varices alone is appropriate therapy. Marshall W. Webster, MD

THE AMERICANJOURNALOF SURGERY" VOLUME168 OCTOBER 1994 315