Endovenous laser ablation: Venous outcomes and thrombotic complications are independent of the presence of deep venous insufficiency
Article Outline
- Abstract
- Methods
- Results
- Discussion
- Conclusion
- Author contributions
- Acknowledgment
- Appendix
- References
- Copyright
Objective
We hypothesize that endovenous laser ablation (EVA) therapy is equally successful in improving venous insufficiency symptoms in patients with or without deep venous insufficiency (DVI).
Methods
From January 2005 through August 2007, EVA of the great saphenous vein (GSV) was attempted in 364 patients (460 limbs) with symptomatic GSV reflux. The GSV was successfully cannulated and obliterated in all but 17 limbs. EVA was performed alone in 308 limbs (69.5%) and with phlebectomy or perforator ligation (EVAP) in 135 limbs (30.5%). Venous clinical severity scores (VCSS) were recorded preoperatively and at 30, 90, 180, and 360 days postoperatively. Patients were classified as those with or without DVI based on duplex imaging valve closure times at the common femoral vein (CFV) and popliteal vein (PV). In a subset of 181 patients undergoing EVA therapy in the operating room, perioperative thrombosis prophylaxis was administered based on a risk-stratification protocol. Patients were assessed with direct end points (VCSS) and indirect end points (vein occlusion rates).
Results
Successful performance of EVA led to complete saphenous vein ablation in 99.8% at 1 month and 95.9% at 1 year. Median VCSS preoperatively was 6 (interquartile range, 5-8), generally decreasing over all time points to 4 (interquartile range, 2-5) beyond 360 days (P < .001). Male gender was independently associated with greater improvement in scores with time (P = .019). Changes in VCSS and duration of vessel occlusion were equivalent regardless of DVI for both isolated EVA and EVAP. For EVAP, the true deep venous thrombosis (DVT) rate was 2.2%, whereas for isolated EVA, the rate was 0% (P = .028); the rate of saphenofemoral thrombus extension was 5.9% for EVAP vs 7.8% for isolated EVA (P = .554). The use of risk-adjusted heparin prophylaxis in patients undergoing EVAP did not have a significant effect on thrombotic complications. There were no differences in true DVT, thrombus extension, or superficial thrombophlebitis between patients with or without DVI. Performance of concomitant phlebectomy, DVI, gender, and age had no effect on the duration of vessel occlusion.
Conclusion
EVA produces successful ablation and is associated with sustained improvement in VCSS. These outcomes are independent of the presence of DVI. Finally, the use of a risk-adjusted thrombosis prevention protocol had no effect on the rate of superficial thrombus extension from EVA or EVAP in patients undergoing general anesthesia.
Superficial venous insufficiency is a major health problem leading to symptomatic lower extremity venous varicosities in 20% of the United States population.1 Left untreated, superficial venous reflux may progress to more severe disease, with signs and symptoms including swelling, aching, skin discoloration, and venous ulceration. The traditional management of superficial venous insufficiency has been ligation of the great saphenous vein (GSV) or the small saphenous vein (SSV), or both, with or without stripping of the vein and removal of the superficial venous varicosities. The interaction of superficial and deep venous insufficiency (DVI), however, has been evaluated in only a limited fashion in the literature,2 and to our knowledge, no evaluation has been done of the specific effect of DVI on the outcomes of saphenous ablation.
Endovenous laser ablation (EVA) of the saphenous vein has become an accepted alternative to ligation and stripping. This technique is equivalent to surgical ligation and vein stripping in terms of clinical symptom resolution and superior in terms of postoperative morbidity in several trials.3, 4, 5 In many centers endoluminal ablation occurs in an outpatient setting under local anesthesia only.
In this report we present our 2-year experience with EVA of the GSV. We hypothesize that EVA is effective in alleviating symptoms of superficial venous insufficiency, irrespective of the presence or absence of deep venous involvement.
Methods
Data source
Patients presenting with varicose veins or other manifestations of superficial venous insufficiency (with or without DVI) were offered treatment. Laser ablation of the GSV, which took place in the operating room and outpatient surgery center (810-nm Endovenous Laser Treatment [EVLT] system, Diomed Inc, Andover, Mass) or in the interventional radiology suite (810-nm Vari-Lase system, Vascular Solutions, Minneapolis, Minn), depending on patient demographics, was attempted in 364 patients and 460 limbs from January 2005 through August 2007. Procedures for patients who needed concomitant phlebectomy (EVAP) or who desired general anesthesia were performed in the operating room. Patients were prospectively entered into a quality assurance database and reviewed retrospectively using the venous reporting standards guidelines.6 This study was approved by the Institutional Review Board of the University of Michigan (IRB approval number HUM00005768).
Duplex testing
Patients underwent a full duplex ultrasound evaluation from groin to knee to rule out acute deep venous thrombosis (DVT). Valve closure times were measured at the level of the GSV 1 cm distal to the saphenofemoral junction, the common femoral vein (CFV) 2 cm distal to the saphenofemoral junction, and the popliteal vein (PV). At all levels, imaging was performed in the sagittal plane, and the angle was corrected.
Reverse flow in the GSV and CFV were measured during a Valsalva maneuver, and reverse flow in the PV was measured during above knee proximal compression. According to University of Michigan Diagnostic Vascular Laboratory protocol, valve closure times >500 milliseconds were considered abnormal (modified from Labropoulos et al).7
Reflux in either the CFV or the PV was considered DVI for the purposes of this study. Perforator incompetence was identified with proximal compression and distal augmentation related to the perforator, again using a valve-closure time of >500 milliseconds as abnormal. Postoperatively, patients underwent a full duplex ultrasound venous evaluation documenting flow and the presence or absence of thrombosis, but not reassessing reflux.
EVA procedure
Patients were treated in three settings: an outpatient clinic, an interventional radiology suite, or in the operating room under general anesthesia, usually with concomitant phlebectomy (EVAP). Approximately 200 mL of tumescent anesthesia (450 mL of normal saline mixed with 50 mL of 1% lidocaine with epinephrine and 15 mL sodium bicarbonate) was infused around each GSV by hand or in the latter part of the series by a Klein pump (HK Surgical, San Clemente, Calif).
The leg was cannulated just above the knee, if possible, or at a level where the saphenous vein was of an adequate size. After placement of the laser catheter, the vein was treated with 14 W of continuous energy as the laser catheter was withdrawn at 1 mm/second for the first 100 seconds, followed by 2.5 mm/second until the laser was 1 cm from the skin surface. Ablation was initiated either distal to the first tributary of the GSV in the groin or 2 cm from the saphenofemoral junction. External compression with the ultrasound probe was used, and the patient was placed in 30° Trendelenburg position during pullback.
If concomitant phlebectomy was performed, stab incisions were made and veins removed with specially designed crochet hooks (Cardinal Health, Ann Arbor, Mich). All wounds were closed with SteriStrips (3M Corp, St. Paul, Minn).
Postoperatively, patients' legs were dressed with a nonstretch Comprilan dressing (Beiersdorf Inc, Hamburg, Germany), a stretch ACE wrap (Hartman-Conco, Inc, Rock Hill, SC), or a combination of both. Early in the experience, patients were instructed not to ambulate the evening of the operation; later they were encouraged to ambulate postoperatively. Compression bandaging was left in place for 48 hours, followed by daily bandaging until approximately 1 week postoperatively. At this time, patients were placed back in thigh-length compression stockings, with 30 to 40 mm Hg pressure preferred.
Demographic and clinical information
We recorded gender, side of procedure, date of operation, and type of operation (EVA or EVAP). Preoperative basic CEAP assessment8 was determined from clinic notes and venous reflux measurements.
Complications
We evaluated the rates of true DVT, thrombus extension from the saphenofemoral junction into the deep system, pulmonary embolism, superficial thrombophlebitis, neovascularization, cellulitis, fluid collections requiring drainage procedures, and bruising. We specifically differentiated thrombus extension from the saphenofemoral junction into the deep system from true DVT based on the ultrasound report. Bruising was categorized as absent, mild, or moderate based on subjective assessment by the examining clinician; no cases of severe bruising were noted. No specific bruising scale was used.
Prevention of DVT
We instituted a policy in our main operating room of preoperative DVT risk factor assessment9, 10 and administration of preoperative and postoperative unfractionated or low-molecular-weight heparin according to on overall risk. Patients underwent general anesthesia in most cases. Patients with up to two risk factors received no prophylactic heparin. Those with three or four risk factors received one dose of 5000 U of unfractionated heparin or 30 mg of Lovenox (Sanofi-Aventis, Bridgewater, NJ) within 60 minutes of the operation. Those with five or more risk factors received the perioperative dose of prophylactic unfractionated heparin or Lovenox as well as prophylactic Lovenox for 1 week postoperatively (based on our own protocol). This change went into effect May 9, 2006. All patients whose procedures were performed in the operating room after this time received the DVT prevention protocol. We evaluated thrombotic complications in the subset performed in our operating room before and after this protocol was implemented.
Venous clinical severity scoring
At each clinic visit, the examiner performed a venous clinical severity scoring (VCSS) assessment11 (maximum, 30 points). These scores were assessed preoperatively as well as postoperatively at days 0 to 30, 30 to 90, 90 to 180, 180 to 360, and >360.
Imaging follow-up
Duplex ultrasound evaluation of the superficial and deep venous systems was performed at similar time points, assessing for DVT and successful saphenous vein ablation. We defined a vessel as being reopened and a treatment failure if there was recanalization from the laser insertion site to the saphenofemoral junction.
Statistical analysis
All categoric comparisons were evaluated using χ2 analysis, ordinal variables using nonparametric tests, and continuous variables using the t test. Duration of venous occlusion by ultrasound evaluation was assessed by life-table analysis. Analysis of factors predictive of late failure was performed using Kaplan-Meier and Cox proportional hazards modeling. Analysis of the interaction of specific variables with the rate of change of VCSS was performed by applying a fixed-effects model to analyze intravariable and intervariable contributions to variation. Statistical calculations were done using SPSS 11.0.1 (SPSS, Inc., Chicago, Ill) and STATA 10.0 (StataCorp, College Station, Tex) software.
Results
Demographic information
EVA was attempted in 364 patients and 460 limbs. The average length of follow-up was 7.4 ± 6.2 months (range, 1 week-35.6 months); two patients were lost to follow-up. Treatment occurred in the operating room for 181 limbs (39.4%), at an outpatient surgery center for 231 (50.2%) and in the radiology suite for 48 (10.4%).
There were no major differences between the patients with DVI and without DVI (Table I). Average age was 50.6 ± 12.2 years for patients with DVI compared with 51.1 ± 12.2 years for those without DVI (P = .671). Similar percentages were women (77.5% vs 78.0% for those with and without DVI, respectively, P < .99) and underwent bilateral treatment (10% vs 8.3%, P = .723). More patients with DVI had treatment of a right-sided saphenous vein than those without DVI (51.4% vs 40.9%, P = .048). The number of DVT risk factors was similar, with a median of four factors (interquartile range, 3-6) for patients with DVI vs four (interquartile range, 3-7; P = .578) for patients without DVI. In addition, the number of stab incisions for EVAP was similar, with a median of 30 (interquartile range, 24-37) for patients with DVI vs 26 (interquartile range, 21-31; P = .097) for patients without DVI.
Table I. Demographics of patients with and without deep venous insufficiency
| Variable | DVI | No DVI | P |
|---|---|---|---|
| Age, average ± SD (range), y | 50.6 ±12.2(16-81) | 51.1±12.2(27-90) | .671 |
| Female, No. (%) | 241(77.5) | 103(78.0) | >.99 |
| DVT risk factors, median (IQR) No. | 4(3-6) | 4(3-7) | .578 |
| Right side, No. (%) | 160(51.4) | 54(40.9%) | .048 |
| Bilateral EVLT,a same setting, No. (%) | 31(10) | 11(8.3%) | .723 |
| EVAP stabs, median (IQR) No.b | 30(24-37) | 26(21-31) | .097 |
aEVLT (Endovenous Laser Treatment) system, Diomed Inc, Andover, Mass. |
bData are for patients who underwent phlebectomy concomitantly with EVA. |
Seventy-four cases were performed in the operating room before the initiation of the DVT prevention protocol and 107 cases after initiation of the protocol. Two or fewer risk factors were present in 12 patients (11.2%), and therefore they received no prophylaxis; 42 (39.3%) had three or four risk factors and were treated with a single perioperative dose of unfractionated heparin or Lovenox; and 45 (42.1%) had five or more risk factors and were therefore treated with a perioperative dose of unfractionated heparin or Lovenox as well as a week postoperatively. Risk factor data were missing for eight patients (7.4%).
The distribution of the maximum CEAP clinical class per patient successfully treated was as follows: C1, 1.6%; C2, 52.6%; C3, 15.6%; C4, 20.1%; C5, 7.4%; and C6, 2.7%. A total of 436 limbs (98.4%) were treated for symptomatic disease; those without symptoms were treated either for severe varicosities that interfered with compression therapy compliance or for recurrent thrombophlebitis. Etiology was primary in 421 limbs (95.0%), secondary in 18 (4.1%), and congenital in four (0.9%). Reflux was isolated to the superficial system only in 132 limbs (29.8%) and involved the superficial and deep system in 311 (70.2%). Reflux of the common femoral vein valve was observed in 302 limbs (68.2%), the PV valve in 71 (16.0%), and in both the CFV and PV in 61 (13.8%).
The average energy delivered per ablation was 2201.4 ± 542.9 J. The average lineal endovenous energy density per limb was 80.7 ± 14.3 J/cm.
Overall clinical outcomes
EVA was performed successfully in all but 17 limbs, for a technical success rate of 96.3%. In seven limbs, the saphenous vein was in spasm and a catheter could not be advanced up to the saphenofemoral junction. In five limbs the micropuncture needle could not access the vein, in four the wire could not be introduced, and in one the catheter could not be placed due to venous webs. Four immediate technical failures occurred in the operating room, and all were immediately converted to open ligation and stripping of the saphenous vein. In the other venues, repeat procedures or referral for surgical approaches was necessary.
The overall rate of vessel occlusion was 99.8% at 1 month, 98.7% at 6 months, 95.9% at 12 months, 94.8% at 18 months, and 91.4% at 24 and 30 months.
Of the 11 recanalizations, two patients were asymptomatic and treated conservatively with compression stockings, six underwent open saphenous ligation and stripping, and three underwent repeat EVA. There was no difference in average lineal endovenous energy density between vessels that recanalized and those that did not (77.5 ± 14.2 vs 80.8 ± 14.3 J/cm, P = .511). However, the vein size at the saphenofemoral junction was larger in vessels that recanalized (0.99 ± 0.34 vs 0.75 ± 0.27 cm, P = .042), and the same trend was seen at the site of entry (0.67 ± 0.27 vs 0.52 ± 0.18 cm, P = .085). Cox proportional hazards modeling failed to identify any patient-specific variables related to the duration of occlusion.
VCSS scores improved significantly with time (P < .001 via Kruskal-Wallis test). Median VCSS decreased at all time points, from a preoperative score of 6 (interquartile range, 5-8) to 4 (interquartile range, 2-5) beyond 360 days (Table II). Men tended to have more rapid improvement in their VCSS than women (P = .019).
Table II. Venous clinical severity scores by component at the preoperative visit and postoperatively at 1 month and 1 year
| Variable | Initial screening (n = 322), No. (%) | Post-op screening, No. (%) | |
|---|---|---|---|
| 1 month (n = 319) | 1 year (n = 104) | ||
| Pain | |||
| None | 106(32.9) | 116(36.4) | 80(76.9) |
| Occasional | 87(27.0 | 77(24.1) | 15(14.4) |
| Daily | 97(30.1) | 58(18.2) | 6(5.8) |
| Daily with meds | 32(9.9) | 68(21.3) | 3(2.9) |
| Varicose veins | |||
| None | 5(1.6) | 72(22.6) | 35(33.7) |
| Few | 56(17.4) | 164(51.4) | 56(53.9) |
| Multiple | 151(46.9) | 64(20.1) | 8(7.7) |
| Extensive | 110(34.2) | 19(6.0) | 5(4.8) |
| Venous edema | |||
| None | 103(32.0) | 209(65.5) | 84(80.8) |
| Evening only | 111(34.5) | 59(18.5) | 13(12.5) |
| Afternoon | 56(17.4) | 21(6.6) | 4(3.8) |
| Morning | 52(16.1) | 30(9.4) | 3(2.9) |
| Skin pigmentation | |||
| None | 199(61.8) | 240(75.2) | 83(79.8) |
| Limited, old | 64(19.9) | 45(14.1) | 16(15.4) |
| Diffuse, recent | 48(14.9) | 24(7.5) | 3(2.9) |
| Wider, recent | 11(3.4) | 10(3.1) | 2(1.9) |
| Inflammation | |||
| None | 304(94.4) | 302(94.7) | 102(98.1) |
| Mild cellulitis | 10(3.1) | 12(3.8) | 1(1.0) |
| Moderate cellulitis | 7(2.2) | 4(1.3) | 1(1.0) |
| Severe cellulitis | 1(0.3) | 1(0.3) | 0(0) |
| Induration | |||
| None | 307(95.3) | 292(91.5) | 103(99.0) |
| Focal <5 cm | 5(1.6) | 19(6.0) | 1(1.0) |
| Medial leg | 8(2.5) | 6(1.9) | 0(0) |
| Diffuse leg | 2(0.6) | 2(0.6) | 0(0) |
| Active ulcers, No. | |||
| 0 | 314(97.5) | 314(98.4) | 102(98.1) |
| 1 | 7(2.2) | 4(1.3) | 2(1.9) |
| 2 | 0(0) | 0(0) | 0(0) |
| >2 | 1(0.3) | 1(0.3) | 0(0) |
| Ulcer duration, mon | |||
| 0 | 315(97.8) | 314(98.4) | 102(98.1) |
| <3 | 2(0.6) | 2(0.6) | 0(0) |
| 3-12 | 3(0.9) | 1(0.3) | 1(1.0) |
| >12 | 2(0.6) | 2(0.6) | 1(1.0) |
| Active ulcer size, cm | |||
| 0 | 316(97.8) | 314(98.4) | 102(98.1) |
| <2 | 5(1.5) | 4(1.3) | 1(1.0) |
| 2-6 | 2(0.6) | 1(0.3) | 1(1.0) |
| >6 | 0(0) | 0(0) | 0(0) |
| Compression therapy | |||
| None | 129(39.9) | 35(11.0) | 20(19.0) |
| Intermittent | 40(12.4) | 5(1.6) | 14(13.3) |
| Most days | 60(18.6) | 26(8.2) | 31(29.5) |
| Fully compliant | 94(29.1) | 252(79.2) | 40(38.1) |
| Median VCSSa | 6(5-8) | 6(4-7) | 3(2-5) |
aThese values are mean (IQR). |
In our series, there were 3 limbs with true DVT (0.7%), 32 with saphenofemoral thrombus extension (7.2%), 11 with superficial thrombophlebitis (2.5%), and 1 pulmonary embolism (0.2%) occurred that was associated with a saphenofemoral thrombus extension and was not a true DVT.
Bruising was assessed at the first postoperative visit (postoperative day 4 at the outpatient surgery center and postoperative day 7 at the main hospital and radiology center) by the clinical examiner at follow-up. Data were available for 303 limbs (67.5%). Bruising was minimal in 48 limbs (15.9%), moderate in 77 (25.4%), and no bruising was apparent in 178 (58.7%).
There were 9 cases (2.0%) of cellulitis in patients' limbs, 4 (0.9%) of postoperative fluid collections requiring further intervention, 4 (0.9%) of neovascularization and perivenous inflammation, and 2 (0.5%) of paresthesias.
The role of DVI
By Kaplan-Meier analysis, there was no significant difference in the duration of vessel occlusion in patients with or without DVI (P = .117, see Fig 1). Using a fixed-effects model, we found that DVI had no effect on the rate of change of the VCSS (P = .572; Fig 2).
Fig 1.
No difference was identified in duration of vessel occlusion in limbs with deep venous insufficiency (DVI) compared with limbs without DVI (P = .117).
Fig 2.
Box and whisker plots show that by fixed-effects modeling, the presence of deep venous insufficiency (DVI) had no effect on the rate of total venous clinical severity scores (VCSS) improvement (P = .572). The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively. The whiskers mark the 90th and 10th percentiles.
There were no significant differences in the rates of DVT, saphenofemoral thrombus extension, superficial thrombophlebitis, paresthesias, or bruising when comparing limbs in the presence or absence of DVI.
The role of concomitant phlebectomy
EVA was the sole procedure in 321 limbs; Concomitant procedures in 135 limbs (29.6%) were phlebectomy in 98 (21.4%) or perforator ligation in 37 (8.2%). The average number of stab incisions was 31 ± 12 (range, 7-73). There were no differences in the rate of change of VCSS or the duration of vessel occlusion based on performance of EVAP (data not shown).
When concomitant phlebectomy or perforator interruption was performed with EVA, the rate of true DVT increased (2.2% vs 0%, P = .028), but no difference was found in saphenofemoral thrombus extension (5.9% vs 7.8%, P = .554) or superficial thrombophlebitis (4.4% vs 1.6%, P = .097). Of note, in all cases of thrombophlebitis in which concomitant EVAP was performed, the thrombophlebitis was related to the phlebectomy sites, not to the ablated saphenous vein. The rates of paresthesias or cellulitis did not differ, but the rate of bruising was substantially higher in cases of EVAP (P < .001), related to the phlebectomy itself.
The role of the DVT prevention protocol
Before the initiation of the protocol, there was one case (1.4%) of DVT, two cases (2.8%) of saphenofemoral thrombus extension, and three cases (4.2%) of superficial thrombophlebitis. In procedures performed after the protocol went into effect, there were two cases (1.9%) of DVT, eight (7.5%) of saphenofemoral thrombus extension, and five (4.7%) of superficial thrombophlebitis. The rates of these complications were not statistically different between the two treatment groups.
Correlation of recannalization to clinical outcome
To assess the relevance of recanalization as a valid outcome, a fixed-effects model was applied to the VCSS vs recanalization. The rate of change of the VCSS was significantly correlated with recanalization, with less improvement noted in VCSS (P = .04) in the presence of recanalization (Fig 3).
Fig 3.
Box and whisker plots show that by using fixed-effects modeling, there was significantly less improvement (P = .04) of venous clinical severity scores (VCSS) in limbs with complete vessel recanalization compared with vessels that remained occluded. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively. The whiskers mark the 90th and 10th percentiles.
Discussion
Venous insufficiency is a common disorder leading to a significant health burden for millions of Americans. Risk factors for varicose veins include age, female gender, family history of varicosities, obesity, and occupational factors.12 Visible venous varicosities occur in 10% to 15% of men and 20% to 30% of women.1, 13 Yet, the symptoms of venous insufficiency, including leg heaviness, aching, cramping, and swelling, are often under-recognized by patients and primary care physicians. Treatment of superficial venous reflux is associated with significant improvements in symptomatic complaints as well as improved quality of life scores.4, 5, 14, 15, 16
Traditional management of superficial reflux has been open ligation and stripping of the GSV. The principles of management are the removal of the hydrostatic forces by ligation of the incompetent saphenofemoral junction and removal of the saphenous trunk. However, this procedure is associated with significant morbidity and postoperative pain. Attempts to limit the procedure to simple ligation of the saphenofemoral junction, ligation and removal of venous varicosities, or ligation and sclerotherapy have been characterized by unacceptably high recurrence rates.17, 18, 19 Recurrence is problematic even in patients undergoing saphenous stripping alone. In a 2007 study, Kostas et al20 documented 23% recurrence and a 16% saphenous nerve injury rate in patients undergoing stripping of the GSV. In addition, although the traditionally accepted rate of DVT from open surgical management of superficial venous insufficiency is 0.15% to 1.8%,21 van Rij et al22 documented a 5.3% rate of DVT in a prospective study.
In 1999 Boné23 first published the technique of EVA of the saphenous vein. In 2001 Navarro et al24 published the first report in the English literature of 40 saphenous veins treated with an 810-nm laser and documented no recanalizations with a mean follow-up of 4.2 months. In the same year, Min et al25 documented a 99% occlusion rate in 90 saphenous veins treated with EVA. A randomized trial comparing EVA with open ligation and stripping documented similar results in quality of life improvement and VCSS, and significantly decreased pain in patients treated with endoluminal ablation.5 In addition, the VCSS has been recently validated in a study using the technique of radiofrequency ablation.26
There has been concern that thromboembolic complications may be higher with endoluminal techniques because the practice of flush ligation in the open procedure cannot be duplicated.27 However, in the five largest studies of EVA in the literature to date, including a total of 3720 limbs, the incidence of DVT was 0.027% (1 in 3720 limbs), pulmonary embolism was 0.038% (1 in 2644 limbs), and superficial thrombophlebitis was 5.5% (91 in 1654 limbs).28, 29, 30, 31, 32 It is unclear from the descriptions of these studies whether there were no extensions of thrombus from the saphenofemoral junction, if the one recorded DVT was such a thrombus, or whether this entity was even systematically evaluated.
Smaller series that have focused specifically on saphenofemoral thrombus extension have documented rates of 2.3% to 7.7%.27, 33 In our study the true DVT rate was 0.7%, the rate of saphenofemoral thrombus extension was 7.2%, the pulmonary embolism rate was 0.2% (1 case), and the superficial thrombophlebitis rate was 2.5%. The institution of a risk-based prophylaxis protocol had no effect on these rates. Some investigators have reported use of 5 to 7 days of postoperative low-molecular-weight heparin,33 whereas others have reported no benefit to prophylaxis in high-risk patients, which would support our findings.34, 35, 36
At present there is no consensus on the management of thrombus extensions into the deep venous system. Johnson and McLafferty37 recommend low-molecular-weight heparin, followed by a 6-week course of Coumadin (Bristol-Myers Squibb, Princeton, NJ), with a target international normalized ratio of 2.0 to 3.0. It is our practice to treat these patients with 1 week of low-molecular-weight heparin and a repeat duplex ultrasound study. In almost all cases, the thrombus extension has resolved and no further treatment is required.
One of the principal findings in this study was that DVI was not a barrier to the safe and effective performance of EVA for the GSV. We found that VCSS improved, irrespective of the presence or absence of DVI. The VCSS has been suggested to be a good means of evaluating clinical improvement after saphenous ablation.26 We also recorded a revised VCSS, eliminating the compression component, because many patients were inadequately treated with compression preoperatively or were noncompliant. However, the findings were identical to those for the total VCSS, and therefore we have not included these data.
There have been suggestions in the literature that DVI may be secondary to superficial reflux, and therefore surgical correction of saphenous reflux should decrease the degree of deep venous reflux.38 We cannot support or refute these statements, because we did not routinely obtain repeat postablation reflux studies. Hach-Wunderle et al2 presented an analysis of open surgical vs endovenous management of superficial venous insufficiency and its role in the development of secondary deep insufficiency, suggesting that removal of the superficial varicosities is sufficient in mild forms of the disease, but crossectomy and stripping of the saphenous trunk were required for severe disease. Sharif et al39 documented excellent wound healing in patients with chronic venous insufficiency and C5 or C6 level disease after EVA, yet patients with DVI were specifically excluded from analysis.
We have documented effectiveness of EVA of the GSV in our patients, with vessel occlusion rates of 98.7% at 6 months and 91.4% at 30 months. This is comparable with other studies that have documented occlusion rates of 95.7% to 98.5% at 6 months and 93.4% to 97.8% at 24 months (Table III).28, 29, 30, 31, 32
Table III. Comparative table of studies in the literature for endovenous laser ablation
| First author, year | No. | Duration (months), % | ||||||
|---|---|---|---|---|---|---|---|---|
| 1 | 3 | 6 | 12 | 24 | 30 | 36 | ||
| Min,28 2003 | 499 | 98.0 | 99.3 | 98.5 | 97.5 | 93.4 | . . . | . . . |
| Agus,29 2006 | 1076 | . . . | . . . | . . . | . . . | . . . | . . . | 97.0 |
| Ravi,30 2006 | 990 | 96.7 | . . . | . . . | . . . | . . . | . . . | . . . |
| Desmyttère,31 2007 | 511 | 98.4 | . . . | 95.7 | 96.8 | 97.8 | . . . | 99.3 |
| Theivacumar,32 2008 | 644 | . . . | 96.0 | . . . | . . . | . . . | . . . | . . . |
| Present study | 449 | 99.8 | . . . | 98.7 | 95.9 | 91.4 | 91.4 | . . . |
One of the limitations of this study is that it was retrospective and nonrandomized. Patients were chosen for EVA or EVAP based on the clinician's assessment of vein size and whether saphenous ablation alone would be sufficient treatment for the varicosities.
Second, this database included three different operators at three different sites. However, although we did not include these data in the results, we found no significant differences in duration of vessel occlusion or complication rate based on site or operator.
Third, follow-up assessment was done by the participating clinicians who were not blinded to the procedure performed. However, VCSS were obtained by multiple providers who were not aware of preoperative or prior scores.
Finally, our definition of DVI was based on an isolated evaluation of the CFV and PV using a 500-millisecond cutoff for valve-closure time.
One potential criticism of the evaluation of recanalization rates as a clinical end point is that it is not clear that this event is associated with worse outcomes or recurrence of symptoms. In one study, greater than 40% of recanalized saphenous veins exhibited symptomatic resolution.40 Yet, when we applied a fixed-effects model to the rate of change of VCSS, we found that although symptomatic improvement was present, it was diminished compared with limbs without recanalization. Therefore, we consider recanalization to be a valid surrogate endpoint of efficacy.
Conclusion
In our experience the performance of EVA was associated with sustained vessel occlusion and significant improvements in clinical severity assessments regardless of DVI. We also found that the use of a prevention protocol based on risk factors had no effect on the rate of thrombotic complications.
Author contributions
We would like to acknowledge the assistance of Justin Dimick, MD, in the statistical analysis of this work.
Appendix
The Michigan Venous Study Group includes the authors and Gayle Adams, PA, Eric Ferguson, MD, Peter K. Henke, MD, Michael Hong, MD, and Farah Mansoor, BS, all from the Section of Vascular Surgery at the University of Michigan, Ann Arbor.
References
- Chronic venous disease in an ethnically diverse population: the San Diego Population Study. Am J Epidemiol. 2003;158:448–456
- . Invasive therapeutic options in truncal varicosity of the great saphenous vein. Vasa. 2006;35:157–166
- . Comparison of endovenous treatment with an 810 nm laser versus conventional stripping of the great saphenous vein in patients with primary varicose veins. Dermatol Surg. 2005;31:1685–1694
- . A nonrandomised controlled trial of endovenous laser therapy and surgery in the treatment of varicose veins. Ann Vasc Surg. 2006;20:451–457
- . Randomized trial comparing endovenous laser ablation of the great saphenous vein with high ligation and stripping in patients with varicose veins: short-term results. J Vasc Surg. 2007;46:308–315
- ; American Venous Forum; Society of Interventional Radiology (Recommended reporting standards for endovenous ablation for the treatment of venous insufficiency: joint statement of The American Venous Forum and The Society of Interventional Radiology). J Vasc Surg. 2007;46:582–589
- Definition of venous reflux in lower-extremity veins. J Vasc Surg. 2003;38:793–798
- Revision of the CEAP classification for chronic venous disorders: Consensus statement. J Vasc Surg. 2004;40:1248–1252
- A guide to venous thromboembolism risk factor assessment. J Thromb Thrombolysis. 2000;9:253–262
- Caprini risk factor assessment, updated November 2006. http://www.venousdisease.com/pad.docAccessed: May 13, 2008
- . Venous severity scoring: an adjunct to venous outcome assessment. J Vasc Surg. 2000;31:1307–1312
- . The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol. 2005;15:175–184
- . Epidemiology of varicose veins. Br J Surg. 1994;81:167–173
- . Long saphenous vein stripping and quality of life–a randomised trial. Eur J Vasc Endovasc Surg. 2001;21:545–549
- . The effect of long saphenous vein stripping on quality of life. J Vasc Surg. 2002;35:1197–1203
- Randomised clinical trial, observational study and assessment of cost-effectiveness of the treatment of varicose veins (REACTIV trial). Health Technol Assess. 2006;10:1–196iii-iv
- . Objective assessment of high ligation without stripping the long saphenous vein. Br J Surg. 1991;78:1139–1142
- . Recurrent varicose veins: a varicographic analysis leading to a new practical classification. Br J Surg. 1995;82:60–62
- . Randomized trial of stripping versus high ligation combined with sclerotherapy in the treatment of the incompetent greater saphenous vein. Am J Surg. 1994;168:311–315
- . The appropriate length of great saphenous vein stripping should be based on the extent of reflux and not on the intent to avoid saphenous nerve injury. J Vasc Surg. 2007;46:1234–1241
- . Is there an increased risk for DVT with the VNUS closure procedure?. J Vasc Surg. 2003;38:628
- . Incidence of deep vein thrombosis after varicose vein surgery. Br J Surg. 2004;91:1582–1585
- . Tratamiento endoluminal de las varices con laser de Diodo (Estudio preliminary). Rev Patol Vasc. 1999;5:35–46
- . Endovenous laser: a new minimally invasive method of treatment for varicose veins–preliminary observations using an 810 nm diode laser. Dermatol Surg. 2001;27:117–122
- . Endovenous laser treatment of the incompetent great saphenous vein. J Vasc Interv Radiol. 2001;12:1167–1171
- . The utility of the venous clinical severity score in 682 limbs treated by radiofrequency saphenous vein ablation. J Vasc Surg. 2007;45:1008–1015
- . Extension of saphenous thrombus into the femoral vein: a potential complication of new endovenous ablation techniques. J Vasc Surg. 2005;41:130–135
- . Endovenous laser treatment of saphenous vein reflux: long-term results. J Vasc Interv Radiol. 2003;14:991–996
- . The first 1000 cases of Italian Endovenous-laser Working Group (IEWG) (Rationale, and long-term outcomes for the 1999-2003 period). Int Angiol. 2006;25:209–215
- . Endovenous ablation of incompetent saphenous veins: a large single-center experience. J Endovasc Therapy. 2006;13:244–248
- . Endovenous 980-nm laser treatment of saphenous veins in a series of 500 patients. J Vasc Surg. 2007;46:1242–1247
- . Factors influencing the effectiveness of endovenous laser ablation (EVLA) in the treatment of great saphenous vein reflux. Eur J Vasc Endovasc Surg. 2008;35:119–123
- . Endovenous laser therapy and radiofrequency ablation of the great saphenous vein: analysis of early efficacy and complications. J Vasc Surg. 2005;42:488–493
- . Morbidity of varicose vein surgery: auditing the benefit of changing clinical practice. Ann R Coll Surg Engl. 1996;78:345–349
- . Thromboprophylaxis can be omitted in selected patients undergoing varicose vein surgery and hernia repair. Br J Surg. 2003;90:818–820
- Minimally invasive technique for ligation and stripping of the small saphenous vein guided by intra-operative duplex ultrasound. Surgeon. 2006;4:372–377
- . Endovenous laser ablation of varicose veins: review of current technologies and clinical outcome. Vascular. 2007;15:250–254
- . Femoral venous reflux abolished by greater saphenous vein stripping. Ann Vasc Surg. 1994;8:566–570
- . Role of endovenous laser treatment in the management of chronic venous insufficiency. Ann Vasc Surg. 2007;21:551–555
- . Long-term outcomes of endovenous radiofrequency obliteration of saphenous reflux as a treatment for superficial venous insufficiency. J Vasc Surg. 2005;42:502–509
Competition of interest: none.
PII: S0741-5214(08)01189-0
doi:10.1016/j.jvs.2008.07.052
© 2008 Published by Elsevier Inc.
Refers to erratum:
- Correction
