From the American Venous Forum

Can phlebectomy be deferred in the treatment of varicose veins?

Daniel L. Monahan, MD, Roseville, California

Objective: This study was designed to observe the clinical sequelae of varicose veins after great saphenous vein (GSV) ablation and to assess possible predictability of spontaneous varicose vein regression.

Methods: Patients with symptomatic varicose veins secondary to GSV insufficiency treated with radiofrequency ablation (RFA) were enrolled in the study. Up to five of the largest varicose veins in each limb were mapped, sized, and documented before RFA. No varicose vein was treated either at the time of RFA or within 6 months postoperatively. Varicose vein status was recorded at follow-up visits.

Results: Fifty-four limbs in 45 patients were included. A total of 222 varicose veins were documented before RFA (4.1 ± 1.1 varicose veins per limb) with an average size of 11.4 ± 3.7 mm. During the follow-up period, complete resolution of visible varicose veins was seen in 13% of limbs after RFA alone, and 63 (28.4%) varicose veins spontaneously resolved. A further 88.7% (141/159) of varicose veins decreased in size an average of 34.6% (4.3 ± 3.4 mm). Preoperatively, 19.4% of varicose veins were above the knee and 75.7% were below the knee. Complete varicose vein resolution was 41.9% (18/43) above the knee and 25.6% (43/168) below the knee. For the above-knee varicose veins, 88.4% (38/43) were located medially, and all the resolved ones (47.4%, 18/38) were medial varicose veins. Resolution rates of the 168 below-knee varicose veins were 30.6% (33/108) of medial, 23.1% (6/26) of anterior, 20.0% (3/15) of lateral, and 5.3% (1/19) of posterior.

Conclusions: Great saphenous vein ablation resulted in subsequent resolution or regression of many lower-limb visible varicose veins. With further study, the predictability of varicose vein regression may perhaps be increased, which can then direct the treatment strategy to further leverage the advantages of minimally invasive endovenous procedures. ( J Vasc Surg 2005;42:1145?9.)

The traditional surgical treatment of symptomatic varicose veins arising from great saphenous vein (GSV) insufficiency has involved eliminating or reducing venous hypertension in the GSV, accompanied by elimination of visible varicosities by stab (avulsion) phlebectomy.1 In fact, the complete extirpation of all visible varicosities is considered by many to be an inviolable tenet of varicose vein surgery.2 Depending on the extensiveness of varicosities to be eliminated, the performance of multiple phlebectomies, usually under general anesthesia, contributes to postoperative pain and morbidity as well as prolonged recovery and delayed return to normal activity.

The development of minimally invasive treatment of the GSV with radiofrequency ablation (RFA) eliminated part of the substantial surgical trauma involved in treating these patients.3 In the hands of many surgeons, it is still accompanied by multiple phlebectomies. The study reported here arose out of a desire to minimize surgical trauma in treating these patients, yet retain completeness and durability of treatment.

A strategy was initiated to perform the GSV ablation procedure without concomitant phlebectomy and to subsequently treat visible varicose veins with sclerotherapy. Patients were seen within 3 days after the ablation procedure, with the intent to perform sclerotherapy at that time. After pursuing this strategy in approximately 30 patients, it was observed at the postoperative visit that the varices in many patients were substantially diminished from their preoperative size. Occasionally, patients deferred sclerotherapy for varying reasons, and on their subsequent return, further regression?and some disappearance?of their varices was noted. Some of these patients had such complete regression of their varicose veins after GSV ablation alone that no further treatment of the varices was pursued.

This study was undertaken in an attempt to address the question of whether complete elimination of surface varicosities is an obligatory accompaniment to treating the GSV to achieve a satisfactory and durable symptomatic and cosmetic outcome. The study represents a preliminary observational experience with a staged approach to treating varicose veins that result from chronic superficial venous hypertension. This approach involves treating the GSV with radiofrequency ablation, with delayed treatment of residual varicosities at 6 months.


Starting in January 2003, 49 patients referred to this office were enrolled. During the enrollment period, an additional 25 potential study patients were treated with the same strategy but not enrolled: 16 patients were not able to comply with follow-up, 8 had hospital-based procedures with no pre-operative vein measurements, and 1 patient declined. Four of the 49 patients enrolled were lost to follow-up. Of the 45 patients included, there was no selection bias based on clinical features. No patient with a history of deep venous thrombosis or with abnormal deep veins on ultrasound examination was included in the study.

All patients were advised of the treatment strategy and protocol for the study, as well as standard treatment options, including multiple phlebectomies. Signed informed consent was obtained for all patients.

All patients underwent a duplex ultrasound diagnostic study that used a 10-MHz probe in an office setting. The standard examination included an assessment for patency and incompetence throughout the length of the great and lesser saphenous veins, with patients reclining at 60°. When results were questionable, the patients were re-examined while standing. Perforators were sought at standard anatomic locations as well as in relation to the presence of otherwise unexplained varicosities. The deep veins were also assessed, above, at, and below the saphenofemoral and saphenopopliteal junctions.

Patients included in the study underwent successful RFA of the GSV of one or both legs. Ablation failed in three limbs in three patients who underwent bilateral treatment. These three limbs were excluded from the study. The extent of ablation extended from the saphenofemoral junction to the lowest point of reflux in the GSV, but not below the upper calf.

The RFA procedure was performed under local and tumescent anesthesia. Most patients were treated in the office, and an oral sedative (5 mg diazepam) was taken preoperatively. Those treated in an outpatient surgery center setting received intravenous sedation. No patient received a prescription for postoperative analgesia. All patients were seen 1 to 3 days postoperatively, and postoperative follow-up at 2 and 6 months was completed.

Preoperatively, and at each postoperative visit, symptoms were assessed and recorded. CEAP clinical classification was also determined. For each limb, the five largest visible varicosities were identified for measurement at their maximum diameters with the patient standing. Only varices linked to the GSV were included. Those varices linked to the lesser saphenous vein or incompetent perforator veins were also managed within the staged strategy, but not included in the study.

table1The clinical end point of the study was the change in size of the varicose veins in response to saphenous ablation alone. An electrocardiogram caliper and a millimeter scale were used to determine the external diameter of the vein on the skin. These measurements were recorded on a clinical follow-up form. At the end of the 6-month period, patients were given the option of sclerotherapy treatment of any remaining varices, if needed. No patients requested, or were treated with, phlebectomy. All treatment was done on an ambulatory basis without interruption of daily activities, other than for the time taken for treatment and the follow-up sessions themselves.


The 54 limbs of the 45 patients included in this report underwent successful ablation of the GSV. The preoperative CEAP distribution is shown in the Table. In the 54 limbs, 222 varicose veins were documented before RFA. An average of 4.1 ± 1.1 varicose veins, with an average diameter of 11.4 ± 3.7 mm, were documented per limb. Four patients with five treated limbs never returned after their 72-hour postoperative check-up, and 10 patients with 10 treated limbs did not return after the 2-month evaluation. Thirty-one patients with 39 treated limbs completed the 6-month study. Two of the patients who did not complete the follow-up protocol moved and could not be contacted. The other patients were contacted but declined further follow-up, reporting satisfaction with their present status.

figure1The following results reflect measurements obtained at each patient?s last visit during the study period. During the follow-up period, 28.4% (n = 63) of varicose veins spontaneously resolved, 4.5% were resolved at 2 months, and another 23.9% had resolved by 6 months. Overall, complete varicose vein resolution was seen in 13.0% of limbs after RFA alone. Among 159 varicose veins that did not resolve, Figure 1 141 (88.7%) decreased in size, on an average of 34.6% (4.3 ± 3.4 mm)(P < .05, paired t test). Ten (6.3%) varicose veins did not change in size, and four (2.5%) varicose veins increased in size by an average of 18.6% (1.3 ± 0.5 mm).

Varicose vein distribution before treatment is shown in Fig 1. Preoperatively, 19.4% and 5.0% of varicose veins were either above the knee or at the knee area, respectively, and 75.7% were below the knee. As illustrated in Fig 2, complete varicose vein resolution was 41.9% (18/43) for above-knee locations, 18.2% (2/11) at the knee area, and 25.6% (43/168) for varicose veins documented below the knee.

figure2For the varicose veins located above the knee, 88.4% (38/43) were located medially, and all the resolved ones were medial varicose veins (18/38, 47.4%). For the 168 below-knee varicose veins, 64.3% were medial, 15.5% anterior, 11.3% posterior, and 8.9% lateral. Resolution of below-knee varicose veins was observed in 30.6% (33/108) of medial locations, 23.1% (6/26) of anterior locations, and 20.0% (3/15) of laterally located varicose veins. Of the 22 varicose veins located posteriorly above or below the knee, only one resolved spontaneously during the follow-up period.

In addition to spontaneously resolved varicose veins, some varicose veins decreased to a size of ≤6 mm—or in four instances, thrombosed—and required no further cosmetic procedure. Of the 159 unresolved varices, 98 were assessed at 6 months. Of these, 47 had regressed to ≤6mm in diameter and generally required no further treatment (Fig 3).

After the 6-month study period, of the 39 limbs of 31 patients that completed the 6-month protocol, 16 limbs (41.0%) had no further treatment after RFA. Twelve limbs (30.1%) underwent ultrasound-guided sclerotherapy of incompetent tributaries of the GSV associated with persistent varices. Three limbs (7.7%) underwent sclerotherapy for surface varices. Eight limbs (20.1%) had both ultrasoundguided and surface sclerotherapy.

During the study period, no new varices or recurrences of regressed varices were observed. As noted, three limbs failed ablation during the study period and were excluded from the analysis, since the object was to assess the effect of the varices when the underlying source of venous hypertension was eliminated.

No serious complications occurred in the study group. Mild cases of superficial phlebitis in the untreated varices have been observed in this practice. Because of this, a compression stocking is prescribed to be worn for a week after GSV ablation. No superficial phlebitis occurred in this study group, however.

A review of the charts of those 25 patients not included in the study, though not containing objective verification of varicose vein regression, revealed relatively similar events in further treatment. Thirty-six limbs were treated, of which 6 limbs (16.7%) had no further treatment after GSV ablation, 15 limbs (41.7%) underwent subsequent ultrasoundguided sclerotherapy for incompetent tributaries, and 13 limbs (36.1%) had sclerotherapy of surface varices. Two limbs (5.6%) underwent both ultrasound-guided sclerotherapy and sclerotherapy of surface varices. The higher rate of subsequent interventions compared with the study group reflected the desire to proceed to treatment of persistent varices sooner rather than following a deferment period. Consequently, with less time allowed for regression, more varices were visible for treatment.

In a phone survey of the study group in January 2005, which represented 12 to 24 months of follow-up, 36 patients with 46 treated limbs were contacted. Compared with their preoperative status, 30 reported absence of symptoms, and six reported improvement. With regard to visible varices before surgery, 16 reported absence of varices, and 20 reported improvement. None reported new symptoms or new varicosities. There was 100% satisfaction with the treatment strategy.


Historically, the complete extirpation of visible varicose veins has been a fundamental tenet in the optimal treatment of varicose veins arising from GSV insufficiency. In a review of the history of varicose vein surgery in the United States, no prior study was identified in which the GSV alone was treated and phlebectomies deferred. In reports by Homans4 and Mayo5 at the beginning of the 20th century, the need for complete removal of visible varicosities was encouraged to prevent recurrence. This dogma was repeated throughout the remainder of the last 100 years. It was, and is, believed that complete removal of the surface varicosities interrupts incompetent tributaries and reduces recurrence rates. Despite technologic innovations and the ensuing improvements in diagnosis and treatment, this tenet has apparently never been challenged or confirmed by direct study.

In the United States, emphasis was directed to treating primarily the GSV. Homans4 is credited with first pursuing this treatment goal with high ligation of the GSV along with complete division of proximal tributaries.6 Eventually, perfection of GSV stripping was described by Myers 1 and Lofgren.7 de Takats8 staged treatment with high ligation of the GSV followed by sclerotherapy, and recorded the phenomenon of regression of varices between treatments, but never reported the consequences of leaving the surface varices untreated.

figure3Before & After Radiofrequency Ablation In Britain, Fegan9 and Hobbs10 later described treatment that focused on sclerotherapy techniques for ablation of incompetent perforator veins. Interestingly, they both noted regression of varicosities after ablation of underlying incompetent perforator vein sources of venous hypertension, 9, 11 and Fegan even asserted that the peripheral varicosities could recover venous wall tone and valvular competence.9 He cited the frequent regression in the puerperium of varicosities arising during pregnancy as an example of this event. However, despite Fegan?s assertion and regardless of the approach taken to eliminate underlying sources of venous hypertension, the complete extirpation of visible varicosities has generally been forcefully maintained.2,12

The observations reported here suggest that an approach focusing on eradication of sources of venous hypertension may decrease the necessity of treating surface varicosities. Suggest is the operative word here, for longer follow-up and greater numbers of patients will be needed to determine what place, if any, this strategy will ultimately have in the treatment of varicose veins. This approach provides the opportunity of savings in both time and cost in the treatment of surface varicosities. The attendant pain, disability, and cosmetic consequences of surgical phlebectomy are also avoided while satisfactory cosmetic and symptomatic outcomes are still achieved.

A surface varicose vein may represent a first-order tributary from the GSV, or perhaps a second, third, or higher order. The resolution or persistence of a surface varix after ablation of the GSV may reflect this relationship. It is interesting to note that most of the varicosities were medially located, especially in the above-knee region. In aboveknee varicose veins, 41.9% resolved spontaneously after GSV ablation, which may suggest their direct relationship with the GSV. All of the resolved above-knee varicose veins were medially located. Posterior varicose veins were less prone to resolve spontaneously, whether the location was above or below the knee.

The persistence of varicosities often reflected an incompetent tributary of the GSV. Treating these persistent tributaries with ultrasound-guided sclerotherapy, rather than directly treating the surface varices, usually resulted in further regression of surface varices and presumably helped prevent recurrence. These additional treatments are nearly painless, take little time, and allow continuation of normal activity. The ultimate place and timing of these additional treatments will require further specific study.

It must be remembered that data were collected for the five largest varices on a given limb. Regression of smaller varices, though not documented, seemed to be more complete. In some instances, a previous varix could still be appreciated as a barely visible or palpable dilation of 3 to 6mm, having become cosmetically irrelevant to the patient. Because symptomatic resolution was so excellent with the ablation procedure alone, many patients turned down cosmetic sclerotherapy for barely noticeable residual varices. These untreated varices are, of course, the center of future evaluation in these patients. Durability of varicose vein regression, affirmed in our relatively early follow-up, awaits future reports.

The current study was designed to challenge conventional treatment strategies. Further follow-up of this study group is planned and will be reported. Many questions still need to be answered in future studies, such as whether aggressive treatment of incompetent tributaries and perforators after saphenous vein ablation will result in greater and more durable regression of varicose veins, and how we can reliably predict which varices will or will not regress.

This observational study had certain methodologic weaknesses; however, even with its weaknesses, the results are felt to be compelling enough to warrant notice and further study. These preliminary data suggest the need for a larger randomized study that would be designed to verify the current findings. Though patient numbers are relatively small, if regression reliably and durably occurs, then it should be observable in even a few patients.

Only the five largest varices of each limb were documented. The caliper technique was, admittedly, fairly simplistic, and it did not reflect the complete extent of varicose vein involvement of a given limb. In retrospect, a method such as grading by photograph, perhaps accompanied by computer-assisted image analysis, might give more accurate information. Ultrasound measurements would be helpful in demonstrating the long-term fate of veins that regress beyond visual discernment.

Even with these limitations, though, regression was demonstrated with very high statistical significance. The failure in patient compliance with completing the study protocol perhaps reflects our general patient population.

The phone survey results are important even though they are not objective and their validity can be questioned. Symptom eradication and good cosmetic outcome are the primary goals of these patients. If at 12 to 24 months these goals are still fulfilled, this validates treatment durability to some degree. Objective follow-up is necessary, however, and these patients will be recalled in the future for documentation of their outcomes.


This study represents a first attempt to provide evidence for a deferred strategy for varicose vein treatment. If proven durable, the advantage of this strategy is obvious. Bruising, hematoma, infection, and other wound-related morbidities associated with phlebectomy are eliminated as well as attendant pain. The need for general anesthesia and a formal operating room setting are eliminated. Normal activity is essentially uninterrupted. These benefits, along with our documented regression rate, might encourage a revision of current practice strategies if long-term recurrence parallels or improves upon historical standards.

1. Myers TT. Results and technique of stripping operation for varicose veins. JAMA 1957;163:89-92.

2. Flye MW. Venous disorders. In: Sabiston DC Jr, editor. Textbook of surgery: the biological basis of modern surgical practice. 15th ed. Philadelphia: WB Saunders; 1997. pp. 1581-93.

3. Lurie F, Creton D, Eklof B, Kabnick LS, Kistner RL, Pichot O, et al. Prospective randomized study of endovenous radiofrequency obliteration (Closure procedure) versus ligation and stripping in a selected patient population (EVOLVeS Study). J Vasc Surg 2003;38:207-14.

4. Homans J. The operative treatment of varicose veins and ulcers, based upon a classification of these lesions. Surg Gynecol Obstet 1916;22: 143-58.

5. Mayo CH. Treatment of varicose Veins. Surg Gynecol Obstet 1906;2: 385-8.

6. Homans, J. Varicose veins: natural history and treatment. In: Browse NL, Burnand KG, Irvine AT, Wilson NM, editors. Diseases of the veins. 2nd ed. London: Oxford University Press, 1999.

7. Lofgren EP, Lofgren KA. Recurrence of varicose veins after the stripping operation. Arch Surg 1971;102:111-4.

8. de Takats G. Ambulatory ligation of the saphenous vein. JAMA 1930; 94:1194-7.

9. Fegan WG. Continuous compression technique of injecting varicose veins. Lancet 1963;2:109-12.

10. Hobbs JT. The treatment of varicose veins: a random trial of injectioncompression therapy versus surgery. Br J Surg 1968;55:777-80.

11. Hobbs JT. Surgery and sclerotherapy in the treatment of varicose veins. Arch Surg 1974;109:793-6.

12. DePalma RG, Rose SS, Bergan JJ. Treatment of varicosities of saphenous origin: a dialogue. In: Goldman MP, Weiss RA, Bergan JJ, editors. Varicose veins and telangiectasias: diagnosis and treatment. 2nd edition. St. Louis: Quality Medical Publishing, Inc; 1999. pp. 197-216. Submitted Feb 9, 2005; accepted Aug 24. 2005.

From the Vein Surgery & Treatment Center of Northern California.
Competition of interest: none.
Presented at the Seventeenth Annual Meeting of the American Venous Forum, San Diego, Calif, Feb 9-13, 2005.
Reprint requests: Daniel L. Monahan, MD, Vein Surgery & Treatment Center of Northern California, 2 Medical Plaza Drive, # 275, Roseville, CA 95661 (e-mail:
0741-5214/$ Copyright � 2005 by The Society for Vascular doi:10.1016/j.jvs.2005.08.034

Monahan D. Can phlebectomy be deferred in the treatment of varicose veins? J Vasc Surg 2005;42(6):1145-1149 doi:10.1016/j.jvs.2005.08.034