🔎

BİLİMSEL-TIBBI MAKALELER

Tüm makalelerime www.drgolbasi.com web sitemden ulaşabilirsiniz.

Endovenous laser with miniphlebectomy for treatment of varicose veins and effect of different levels of laser energy on recanalization. A single center experience

Ilhan Golbasi,MD

Lasers in Medical Science

ISSN 0268-8921

Lasers Med Sci
DOI 10.1007/s10103-014-1626-

Received: 24 February 2014 / Accepted: 25 June 2014 # Springer-Verlag London 2014

Abstract

Varicose veins, associated with great saphenous vein (GSV) incompetence, are traditionally treated with convention- al surgery. In recent years, minimally invasive alternatives to surgical treatment such as the endovenous laser ablation (EVLA) and radiofrequency (RF) ablation have been devel- oped with promising results. Residual varicose veins following EVLA, regress untouched, or phlebectomy or foam sclerother- apy can be concomitantly performed. The aim of the present study was to investigate the safety and efficacy of EVLA with different levels of laser energy in patients with varicose veins secondary to saphenous vein reflux. From February 2006 to August 2011, 740 EVLA, usually with concomitant miniphlebectomies, were performed in 552 patients. A total of 665 GSV, 53 small saphenous veins (SSV), and 22 both GSV and SSV were treated with EVLA under duplex USG. At 84 patients, bilateral intervention is made. In addition, miniphlebectomy was performed in 540 patients. A duplex ultrasound (US) is performed to patients preoccupying chronic venous insufficiency (with visible varicose veins, ankle edema, skin changes, or ulcer). Saphenous vein incompetence was diagnosed with saphenofemoral, saphenopopliteal, or truncal vein reflux in response to manual compression and release with patient standing. The procedures were performed under local anesthesia with light sedation or spinal anesthesia. Endovenous 980-nm diode laser source was used at a continuous mode. The mean energy applied per length of GSV during the treatment was 77.5 ± 17.0 J (range 60–100 J/cm). An US evaluation was performed at first week of the procedure. Follow-up evaluation and duplex US scanning were performed at 1 and 6 months, and at 1 and 2 years to assess treatment efficacy and adverse reactions. Average follow-up period was 32±4 months (3– 55 months). There were one patient with infection and two patients with thrombus extension into the femoral vein after EVLA. Overall occlusion rate was 95 %. No post-procedural deep venous thrombosis or pulmonary embolism occurred. Laser energy, less than 80 J/cm, was significantly associated with increased recanalization of saphenous vein, among the other energy levels. EVLA seems a good alternative to surgery by the application of energy of not less than 80 J/cm. It is both safe and effective. It is a well-tolerated procedure with rare and relatively minor complications.

Ilhan Golbasi (*) : C. Turkay : O. Erbasan : C. Kemaloğlu :
Ö. Bayezid
Cardiovascular Surgery, Akdeniz University, Antalya, Turkey e-mail: [email protected]

S. Sanli
Anesthesiology and Reanimation, Akdeniz University, Antalya, Turkey

M. Turkay
Public Health, Akdeniz University, Antalya, Turkey

Keywords Endovenous laser ablation . Venous insufficiency . Miniphlebectomy

Introduction

Chronic venous insufficiency and varicose veins affect a sig- nificant proportion of any population. Varicose veins, associ- ated with great saphenous vein (GSV) incompetence, have been traditionally treated with surgical saphenofemoral liga- tion and stripping as the standard of care for nearly a century. In recent years, minimally invasive alternatives such as the endovenous laser ablation (EVLA) and radiofrequency (RF) therapy have been developed with promising results [1–3].

Endovenous laser allows delivery of laser energy directly into the vein lumen to cause collagen contraction and denu- dation of endothelium. This stimulates vein-wall thickening with eventual fibrosis of the vein. Secondary treatment of

Endovenous laser with miniphlebectomy for treatment of varicose veins and effect of different levels of laser energy on recanalization. A single center experienceresidual varicosities after EVLA is controversial: Phlebectomy or sclerotherapy simultaneously can be done with vein ablation or residual varicosities can be observed for spontaneous regres- sion following vein ablation alone [4].

Aim of the study

The aim of the present study was to investigate the safety and efficacy of EVLA with different levels of laser energy in patients with varicose veins secondary to saphenous vein reflux. In the present study, we report the outcome of EVLA with different energy levels, combined with miniphlebectomy in a series of our patients with GSV and small saphenous vein (SSV) incompetence.

Patients and methods

From February 2006 to August 2011, 552 patients (740 lower limbs) with symptomatic varicose veins and venous insuffi- ciency underwent endovenous ablation and mostly with miniphlebectomy. Before the procedure, an informed written consent was obtained from all patients. Of the patients, 58 % were women (228 men and 324 women) and the mean age of the patients was 44.3±14.2 years. A total of 665 GSV, 53 SSV, and 22 both GSV and SSV were treated with EVLA. In addition, miniphlebectomy was performed for residual truncal varicosities in 540 patients. All patients underwent preopera- tive clinical and duplex ultrasonographic examination. A du- plex ultrasound (US) is performed to patients preoccupying chronic venous insufficiency (with visible varicose veins, ankle edema, skin changes, or ulcer). Linear probe frequency was 7.5 Hz. Both longitudinal and cross scannings are made. Patients with superficial venous insufficiency confirmed by duplex US scanning were considered as candidates for EVLA treatment. Saphenous vein incompetence was diagnosed with saphenofemoral, saphenopopliteal, or truncal vein reflux in response to manual compression and release with patient standing. Saphenofemoral junction (SFJ), saphenopopliteal junction (SPJ), GSV, and SSV were checked for reflux while patient is at supine position. Patients with reflux time more than 2 s (with Valsalva maneuver) were also candidates for EVLA treatment. Saphenous vein diameter is also taken to consideration: All saphenous veins were between 6 and 15 mm in diameter (health insurance system only pays the intervention cost if saphenous vein diameter is 6 mm or wider and supine reflux time is more than 0.5 s). We made venous mapping and marked the varicose veins, if needed, before the procedure at standing position. Exclusion criteria included continuous axial deep venous reflux, isolated perforator vein reflux, saphenous vein diameter less than 6 or more than 15 mm, non-palpable pedal pulses, deep vein thrombosis

(DVT), general poor health, pregnancy, previous treatment challenges, and extremely tortuous veins.

According to the clinical, etiology, anatomy, and patho- physiology (CEAP) clinical classification, there were 12 limbs with telangiectasia or reticular veins (c1), 597 legs with visible varicose veins (c2) and 51 with edema (c3), 46 patients had skin changes due to venous disease (c4), another 23 patients had a healed ulcer (c5), and 11 patients were suffering from an active ulcer (c6; Table 1).

Technique of EVLA

When we first started to perform EVLA in 2006, the proce- dure was performed under spinal anesthesia, supplemented with local tumescent anesthesia. In the process of time, local anesthesia with light sedation and tumescent anesthesia is preferred. Patients were placed in the supine position for GSV ablation and turned to the prone position for SSV to be ablated. The patient was placed in the reversed Trendelenburg position. The GSV is seen on B-Mode US longitudinally and a 18-G Seldinger needle is inserted into the GSV. The GSV was cannulated usually below the knee at the point where it is wide enough and not tortuous. A 0.035-in. J-tipped guide wire was inserted into the vein followed by placement of a 5-F sheath. For the GSV, the laser catheter was introduced through the sheath into the vein and advanced 2 cm distal of the SFJ. When the small saphenous was treated, the tip was positioned 2 to 3 cm below the SPJ. The 600-μm laser bare fiber was then

CEAP clinical, etiology, anatomy, and pathophysiology, GSV great sa- phenous vein, SSV small saphenous vein passed to the tip of the sheath. Tumescent local anesthetic solution was infiltrated along the whole length of the SV using a 0.9 mm × 180 mm needle under US. Tumescent anesthetic solution was prepared by diluting 30 ml of 2 % lidocaine with 1:200.000 adrenaline and 1.5 % Na-bicarbonate in 500 ml of saline [5]. During the laser ablation process, patients were maintained in the Trendelenburg position [6]. Endovenous 980-nm diode laser source (CHIROLAS surgical laser, Ger- many) was used at continuous mode. The mechanism of action at 980-nm wavelength is, combined effects of vein spasm, compression by perivenous tumescent anesthesia and ablation in the Trendelenburg position results in an empty vein and direct thermal damage to the vein wall. Heating of the vein wall causes collagen contraction and destruction of the endothelium. This stimulates vein-wall thickening, leading to luminal contraction, venous thrombosis, and vein fibrosis [7, 8]. Although the pulse mode and the continuous mode pro- duce similar vein occlusion rates, we used the continuous mode because it affects the entire vein, not causes the febrile effect at intermittent vein-wall portions [9]. Intermittent- pulsed laser fiber pullback has been reported, in a retrospec- tive review, to cause significantly greater levels of post- operative pain and bruising, compared with a continuous pullback protocol [10].

The EVLA catheter was pulled slowly, at a rate of 1 to 2 cm/s for the 60–100 J/cm using 10- to 15-W laser energy. The GSV was treated from below the knee to the SFJ and the SSV was ablated from mid-calf to the SPJ. In patients with varicose collaterals, miniphlebectomy was performed with 1- to 2-mm incisions over varicosities by using a hook following laser ablation. Incisions were closed with Steri-Strips. Elastic adhesive bandage was applied to the whole length of the treated limb after the procedure for 2–4 days, then it was changed to a class II (30–40 mmHg) full-length graduated support stocking that was worn for a further 3–5 weeks, except during sleep and baths. Some studies showed that compression significantly reduces post-operative pain [10]. All patients were asked to walk 4 h after the procedure and to return to normal daily activities as soon as they feel com- fortable. Non-steroidal anti-inflammatory drugs were pre- scribed for all patients with no contraindication to their use.

Follow-up protocol

All patients were followed up on an outpatient basis for physical examination and duplex ultrasonography. The pres- ence of residual varicosities, ecchymosis, pain, induration, skin burns, dysesthesia, superficial thrombophlebitis, and he- matoma, edema, ecchymosis, and dyschromia, and ulceration were recorded and classified as minor complications.

An US evaluation was performed at first week of the procedure to evaluate occlusion of the vein, wall thickness, and clot extension into the deep venous system. Follow-up

evaluation and duplex US scanning were performed at 1 and 6 months, and at 1 and 2 years after procedure to assess treatment efficacy and adverse reactions.

Statistical analysis

For statistical evaluation, χ2 test was used to investigate the difference between variables.

Results

Demographic and clinical data of our patients are shown in Table 1. The mean length of GSVs treated was 43.5±5.0 cm (range 18–61 cm). The mean length of SSVs treated was 13± 3.6 cm (range 6–19.5 cm). Of the varicose veins, 19 % was above-knee and 75 % was below-knee. GSVs are treated usually from below-the-knee level to the saphenofemoral junction (the punction of GSV is made as distally as possi- ble—where distal GSV is wide enough and non-tortuous) and SSVs are treated from mid-calf to parvopopliteal junction. The mean energy applied per length of GSV during the treatment was 77.5±17.0 J (range 60–100 J/cm; Table 2). Median duration of procedure was 32 min (range 16–65). Average follow-up was 32.0±4 months (3–55 months). All treated veins were occluded at first week US control except for one patient. This patient was treated with 60 J/cm of laser energy and had a saphenous vein diameter of 12 mm. The excellent early occlusion rates following EVLA have been maintained at 93 to 97 % on follow-up out to 2 years with the majority of recurrences occurring by the first 3 months. Longer follow-up to 4 years has demonstrated a 95 % occlusion rate with recurrences in reflux occurring secondary to recanalization and not to neovascularization. Total and partial recanalizations were noted in 8 (1.3 %) and 22 (3.7 %) cases, respectively, and saphenous vein occlusion rate was 95 % as assessed by duplexNumber of lower limbs Recanalization (partial/total) N (%)15 (10/5) (8.52) 7 (5/2) (12.50) 7 (6/1) (0.35)
1 (1/0) (0.63) 30 (22/8) (5.42)*p<0.01, less than 80 J was significantly associated with increased recanalization among the other energy levels

US (Table 2). We encountered all recanalizations in the first month and determined neither partial nor total recanalizations in other patients at following months. Complementary miniphlebectomy was performed in 540 patients. An average of 9 phlebectomies was performed per limb (range 4–21). Major complications after EVLA were infection in one patient and thrombus extension into the femoral vein in two patients. No post-procedural deep venous thrombosis or pulmonary embolism occurred. Minor complications were hematoma at the percutaneous venotomy site in seven patients, temporary paresthesia in three patients, and some degrees of the ecchy- mosis and dyschromia in all patients. Post-laser phlebitis is a frequent minor complication. We observed some degrees of phlebitis in 24 % of our patients. Post-procedure pain occurred in 36 % of the patients but did not affect their quality of life. Residual varicosities were found in 1 % of the patients at 6 months after procedure, but only 0.5 % of those required subsequent interventions.

Discussion

Majority of varicose veins arise from incompetence of the SFJ and GSV reflux. Saphenofemoral high flush ligation—strip- ping and modifications—have been used as the standard of care for nearly a century. However, it is associated with significant post-operative morbidities such as deep venous thrombosis, paresthesia, and saphenous nerve injury [1, 2, 11]. Endovenous ablation procedure has rapidly become pop- ular with clinicians who treat varicose veins due to its relative simplicity and high rate of patient satisfaction. The advantages of endovenous laser therapy are the lack of surgical wounds, infection, and scarring [4, 12–14]. We performed most of our cases under local anesthesia and only a few cases with spinal or general anesthesia. Those were the first cases, and our inexperience about punction of the GSV under US and the risk for turning to open surgery were the reasons for regional or general anesthesia. Both patients and surgeons were satis- fied with ablation and miniphlebectomy under local anesthe- sia, and we have not seen any major complication.

The 810-, 940-, 980-, 1,064-, 1,320-, 1,470-, and 1,500-nm wavelength lasers are effective in inducing vessel occlusion. EVLA causes permanent vein closure through a high- temperature photothermolytic process at the point of contact between the vein and the laser [4, 12, 15, 16]. All the wave- lengths produce a venous thrombosis which later on involves to a fibrotic process. Intravenous temperature monitoring studies during EVLA have confirmed that the peak tempera- tures at the fiber tip exceed 1,000 °C, and continuous temper- atures of at least 300 °C are maintained in the firing zone for the majority of the procedure. The main chromophore of 1,064- to 1,320-nm lasers, at least initially, is water, while other wavelengths used for EVLA primarily target

hemoglobin. Intravenous high temperature performs destruc- tion of the endothelium and contraction of collagen in the venous walls, and ultimately results in venous fibrosis. Less energy is necessary to close the target vein when higher wavelength lasers (1,470–1,500 nm) are used, as compared to the use of lower wavelength lasers [16].

Some of the laser systems use bare-tip laser fibers and others use radial fibers which are with more uniform energy distribution. Studies comparing two different fibers show no significant difference about the occlusion rates of saphenous vein but less bruising and pain with radial fibers [17]. Bare fibers can be used either in pulsed mode, in which the laser energy is applied intermittent, or in continuous mode. We have used the laser with a bare fiber and 980-nm wavelength at continuous mode in all patients.

EVLA causes permanent vein closure through a high- temperature photothermolytic process at the point of contact between the vein and the laser. Collagen has been noted to contract at about 50 °C, while necrosis occurs between 70 and 100 °C. The ranges of energies used to achieve durable ablation included 50 J/cm for veins ≤4.5 mm and 70–120 J/ cm for veins 5–10 mm in diameter. The success of EVLA has been shown to depend on the amount of laser energy delivered with non-occlusion and early reopening of the GSV seen more frequently with delivery of less than 70 J/cm [14, 15]. In our follow-up, the occlusion rate was 95 %, and partial or total recanalization rate was 5 % (22/8). Twenty-two of 30 recan- alization patients were the ones treated with 60 and 70 J/cm laser energy (15 recanalizations were seen after 60 J/cm and 7 were seen after 70 J/cm). This parameter corresponded to another study, in which high early recanalization occurred after low-energy EVLA. The recanalization after EVLA with 46.6 J/cm was 22.5 % [14].

Endovenous laser ablation of the SSV has excellent early and mid-term results. A multicenter prospective study evalu- ated the feasibility, safety, and efficacy of endovenous laser ablation to treat SSV [18]. The treated SSV was 14.36 % in our study population. Since the sural nerve is close to distal 1/3 of the SSV, laser ablation should not be used in this portion. It was reported that the prevalence of thrombosis and paresthesia was very low, and symptom relief was very good [19]. We performed EVLA of the SSV in 35 cases and observed paresthesia only at one patient.

Major complications following EVLA such as DVT, infec- tion, and pulmonary embolism have been reported rarely. There was one infection case in our series.

Rates of DVT, pooled from multiple series, are lower than 1 % [20–22]. With the growing popularity of endovenous thermoablation, there have been numerous reports of the complication endovenous heat-induced thrombosis (EHIT). EHIT is thrombus extending from the superficial venous system into the deep venous system at a site of recent thermoablation, most commonly thrombus extending from

the great saphenous vein (GSV) into the common femoral vein. One group reported a rate of EHIT as 7.7 % [23, 24]. We have not observed DVT in our patients but encountered two cases of EHIT.

Minor complications have been reported in 3 to 20 % of patients, including bruising around the puncture site, bleeding, transient paresthesias, superficial phlebitis, skin burns, or pig- mentation. Nerve injury following laser therapy occurs rarely. We have not observed saphenous nerve injury following laser therapy of the GSV.

Phlebectomy or foam sclerotherapy of residual varicose veins following EVLA is controversial. These options include simultaneous vein ablation or observation for spontaneous regression following vein ablation alone. Many branch vari- cosities do not require treatment after GSV reflux is eliminat- ed; they resolve completely or near completely.

Previous experience with EVLA indicates that some 60 % of the patients require delayed foam sclerotherapy for residual varicosities following ablation of the above-knee GSV. Com- plete varicose vein resolution was 41 % for above-knee and 25 % for below-knee [25, 26].

Residual varicosities cause superficial thrombophlebitis in 5 to 10 % of patients because of the stasis from altered venous drainage. The author’s treatment of choice is laser vein abla- tion with concurrent phlebectomy as this procedure greatly reduces the risk of developing secondary thrombophlebitis. If delayed treatment is selected and necessary, either phlebectomy or sclerotherapy may be chosen depending on physician’s preference. The remaining or persistent varicose veins can be removed with a small hook through tiny inci- sions. The procedure is ideal for treating veins that are too large to treat with sclerotherapy. This adds only 10 to 20 min to procedure time. The resulting improved cosmetic effect and quick recovery are the results of the latest advances in phle- bology. Both procedures can be done concomitantly without a significant increase in complications. Some studies show that secondary reintervention rates at 1 year were 66 % in the sequential group versus 0.04 % in the concomitant group. Quality of life was initially superior in the concomitant group [27, 28]. Our rate of post-operative complications such as pain, cellulitis, or paresthesias is similar with series where EVLA was the only procedure performed. Most individuals were able to return to their normal routines immediately following the procedure. Since the incisions made were very small, most patients experienced almost no scarring.

Conclusions

EVLA combined with miniphlebectomy for treatment of varicose and saphenous vein reflux seemed to be a very effective and well-tolerated procedure with rare and relatively minor complications. Concomitant ablation and microphlebectomy was also safe and associated with a low reintervention rate.

References

1. Caggiati A, Allegra C (2007) Historical introduction. In: Bergan JJ (ed) The vein book. Elsevier Academic Pres, Oxford, pp 1–14

2. Bergan JJ, Kumins NH, Owens EL, Sparks SR (2002) Surgical and endovascular treatment of lower extremity venous insufficiency. J Vasc Interv Radiol 13:563–568

3. Evans C, Fowkes FG, Ruckley CV, Lee AJ (1999) Prevalence of varicose veins and chronic venous insufficiency in men and women in the general population: Edinburgh Vein Study. J Epidemiol Community Health 53:149–153

4. Navarro L, Min RJ, Boné C (2001) Endovenous laser: a new mini- mally invasive method of treatment for varicose veins—preliminary observations using an 810 nm diode laser. Dermatol Surg 27:117– 122

5. Rasmussen LH, Bjoern L, Lawaetz M et al (2007) 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 46:308–315

6. Stirling M, Shortell CK (2006) Endovascular treatment of varicose veins. Semin Vasc Surg 19:109–115

7. Desmyttere J, Grard C, Mordon S (2005) A 2 years follow-up study of endovenous 980 nm laser treatment of the great saphenous vein: role of the blood content in the GSV. Med Laser Appl 20:283–289

8. Schmedt CG, Sroka R, Steckmeier S et al (2006) Investigation on radiofrequency and laser (980 nm) effects after endoluminal treat- ment of saphenous vein insufficiency in an ex vivo model. Eur J Vasc Surg 32:318–325

9. Satokawa H, Yokoyama H, Wakamatsu H et al (2010) Comparison of endovenous laser treatment for varicose veins with high ligation using pulse mode and without high ligation using continuous mode and lower energy. Ann Vasc Dis 3(1):46–51

10. Zimmet SE (2007) Endovenous laser ablation. Phlebolymphology 14:51–58

11. Weiss RA (2001) Endovenous techniques for elimination of saphe- nous reflux: a valuable treatment modality. Dermatol Surg 27:902– 905

12. Bone C (1999) Tratamiento endoluminal de las varices conlaser de Diodo. Estudio preliminary. Rev Patol Vasc 5:35–46

13. Proebstle TM, Krummenauer F, Gul D, Knop J (2004) Nonocclusion and early reopening of the great saphenous vein after endovenous laser treatment is fluence dependent. Dermatol Surg 30:174–178

14. Timperman PE, Sichlau M, Ryu RK (2004) Greater energy delivery improves treatment success of endovenous laser treatment of incom- petent saphenous veins. J Vasc Interv Radiol 15:1061–1063

15. Agus GB, Mancini S, Magi G (2006) The first 1000 cases of Italian Endovenous-laser Working Group (IEWG). Rationale, and long-term outcomes for the 1999–2003 period. Int Angiol 25:209–215

16. Vuylsteke ME, Vandekerckhove PJ, De Bo TH, Moons P, Mordon S (2010) Use of a new endovenous laser device: results of the 1500-nm laser. Ann Vasc Surg 24:205–211

17. Schwarz T et al (2010) Endovenous laser ablation of varicose veins with the 1470-nm diode laser. J Vasc Surg 51:1474–1478

18. Fan CM, Rox-Anderson R (2008) Endovenous laser ablation: mech- anism of action. Phlebology 23:206–213

19. Kalra M, Gloviczki P (2008) Endovenous ablation of the saphenous vein. Perspect Vasc Surg Endovasc Ther 20:371–380

20. Kontothanassis D, Di Mitri R, Ferrari Ruffino S, Zambrini E, Camporese G, Gerard JL, Labropoulos N (2009) Endovenous

Author's personal copy

laser treatment of the small saphenous vein. J Vasc Surg 49:

973–979
21. Marston WA, Brabham VW, Mendes R, Berndt D, Weiner M, Keagy

B (2008) The importance of deep venous reflux velocity as a deter- minant of outcome in patients with combined superficial and deep venous reflux treated with endovenous saphenous ablation. J Vasc Surg 48:400–406

22. Fernández CF, Roizental M, Carvallo J (2008) Combined endovenous laser therapy and microphlebectomy in the treatment of varicose veins: efficacy and complications of a large single-center experience. J Vasc Surg 48:947–952

23. Perkowski P, Ravi R, Gowda RC, Olsen D, Ramaiah V, Rodriguez- Lopez JA, Diethrich EB (2004) Endovenous laser ablation of the saphenous vein for treatment of venous insufficiency and varicose veins: early results from a large single-center experience. J Endovasc Ther 11:132–138

24. Mozes G, Kalra M, Carmo M, Swenson L, Gloviczki P (2005) Extension of saphenous thrombus into the femoral vein: a potential complication of new endovenous ablation techniques. J Vasc Surg 41:130–135

25. Theivacumar NS, Dellagarammaticas D, Beale R, Mavor A, Gough M (2006) The clinical significance of persistent below-knee great saphe- nous vein (BK-GSV) reflux following endovenous GSV laser ablation (EVLA): do we need to modify treatment? Phlebology 93:141–156

26. Schanzer H (2010) Endovenous ablation plus microphlebectomy/ sclerotherapy for the treatment of varicose veins: single or two- stage procedure? Vasc Endovasc Surg 44:545–549

27. Monahan DL (2005) Can phlebectomy be deferred in the treatment of varicose veins? J Vasc Surg 42:1145–1149

28. Carradice D, Mekako AI, Hatfield J, Chetter IC (2009) Randomized clinical trial of concomitant or sequential phlebectomy after endovenous laser therapy for varicose veins. Br J Surg 96(4):369–375

Author's personal copy

Lasers Med Sci