Evaluation of endovenous radiofrequency ablation and laser therapy with endoluminal optical coherence tomography in an ex vivo model

Article Outline

• Abstract
• Abstract
• Methods
  • Ex vivo model
  • Endovenous radiofrequency ablation
  • Endovenous laser therapy
  • Endovascular optical coherence tomography
  • Microscopy
  • Qualitative and quantitative evaluation before and after treatment
  • Statistical analysis
• Results
  • Qualitative analysis of thermal effects on tissue by endovascular optical coherence tomography
  • Quantitative analysis of media thickness and vessel lumen with endovascular optical coherence tomography and TapeMeasure
• Discussion
• Conclusions
• Author contributions
• Acknowledgment
• References
• Copyright

Background

This study evaluated the ability of endovascular optical coherence tomography (eOCT) to detect qualitative tissue alteration and quantitative changes of vein wall thickness and vein lumen diameter comparing endovenous radiofrequency ablation (RFA) and endovenous laser therapy (ELT) in an established ex vivo model.

Methods

Endoluminal eOCT was performed by means of a new prototype rotating system (System M1, LightLab Imaging Inc, Boston, Mass) with automatic pullback of 1 mm/s. In the course of an eOCT examination of a 50-mm vein segment, 264 electronic cross section images with a spatial resolution of 15 to 20 μm are acquired. The eOCT scans were performed before and after treatment of each of 13 treated vein segments and of six control vein segments. Thirteen subcutaneous cow foot veins were reperfused in situ, and the defined 50-mm vein segments in the study were treated with RFA (n = 2) and ELT (n = 11). RFA followed the clinical VNUS-Closure protocol (VNUS Medical Technologies, San Jose, Calif) using a 6F 60-mm Closure-Plus catheter. ELT was performed using light of λ = 980 nm with a laser power of 3 (n = 2), 5 (n = 2), and 7 W (n = 4) with a paced pullback protocol with laser irradiation for 1.5 seconds every 3 mm, resulting in linear endovenous energy densities (LEED) of 15, 25, and 35 J/cm. Using 11 W (n = 3) with a continuous pullback protocol at 3 mm/s resulted in a LEED of 36.5 J/cm. Ten histologic cross sections of each treated and control vein segment were correlated with the corresponding eOCT cross sections to evaluate qualitative representation of vein wall layers and tissue alterations such as ablation and vein wall perforation. In addition, 26 eOCT cross sections of every treated vein segment before and after treatment and every control vein segment were analyzed to calculate quantitative changes in media thickness and vein lumen diameter.

Results

In all specimens, qualitative analysis with eOCT demonstrated a clear match with histologic cross sections. A symmetrical, complete, circular disintegration of intima and media structures, without any transmural tissue defects, was shown after RFA. Pronounced semicircular tissue ablations (3 to 14 per 50 mm) and complete vessel wall perforations (0 to 16 per 50 mm) were detected after ELT. The quantitative analysis demonstrated a significant (P < .0001) increase in intima-media thickness after RFA (37.8% to 66.7%) and ELT (11.1% to 45.7%). A significant (P < .0001) reduction of vessel lumen diameter (36.3% to 42.2%) was found after RFA. Owing to the limited number of treated vein segments and inhomogeneous baseline vein lumen diameters, no linear correlation between laser energy level and effects on tissue such as ablation/perforation, media thickening, or vein lumen diameter could be identified.

Conclusions

In our ex vivo cow foot model, eOCT is able to reproduce normal vein wall structures and endovenous acute thermal alterations, such as tissue ablation and vessel wall perforations. Endovenous eOCT images can also be analyzed quantitatively to measure media thickness or vein lumen diameter. Endovascular OCT could become a valuable alternative tool for morphologic investigation of tissue alterations after endovenous thermal procedures.

Clinical Relevance

Clinical data indicate that endovenous radiofrequency ablation (RFA) and laser therapy (ELT) can be performed safely; however, reports of recanalization, phlebitis, ecchymosis, and paresthesia indicate a certain potential for improving clinical results. Parallel with clinical trials, experimental studies could also help to identify certain treatment protocols that lead to effective collagen denaturation, vein wall thickening, and reduction of vein lumen diameter with minimal perivascular injury. This article reports the first results, to our knowledge, of a standardized experimental evaluation of endovenous radiofrequency and laser therapy using a newly developed ex vivo model. A new prototype endovascular high-resolution optical imaging procedure, endovascular optical coherence tomography (eOCT), was also used. The analysis of effects on tissue showed a pronounced and more reproducible collagen denaturation after RFA compared with ELT with a bare-tipped optical fiber. This report extends the understanding of tissue effects and offers suggestions for technical improvements especially for ELT. In the context of the endoluminal vein treatment presented, eOCT will mainly be used in experimental settings. When technical problems in this prototype device have been completely solved, eOCT could also be used in blood-filled vessels. EOCT might be considered for clinical trials investigating effects on tissue after RFA and ELT as an additional instrument alongside transcutaneous duplex ultrasound, but not in clinical routine. This new method could be of interest for investigating saphenous veins in situ to evaluate the suitability for venous bypasses or for intraoperative quality control of bypass anastomoses. Beyond that, the monitoring of percutaneous intra-arterial or of intravenous interventions such as balloon angioplasty and stent implantation should be possible.

Endovenous radiofrequency ablation (RFA) and endovenous laser therapy (ELT) are established techniques in clinical routine for occluding incompetent saphenous veins. The aim of complete, irreversible occlusion without any perivenous injury cannot be achieved in all patients, however.¹, ², ³, ⁴ Various clinical data and experimental findings indicate a certain potential for developing and optimizing both procedures to achieve higher occlusion rates without adverse events such as ecchymosis, hematoma, phlebitis, nerve injury, or recanalization.⁵, ⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹ Parallel with clinical studies, theoretic and experimental models for endovenous thermal treatment could help to identify the most effective treatment protocols with minimal perivascular injury.

For the standardized experimental evaluation of immediate macroscopic and microscopic changes after endovenous thermal ablation with RFA and ELT, our group created cost-effective ex vivo cow foot model.¹² In addition to macroscopic and microscopic evaluation, a prototype endoluminal optical coherence tomography (eOCT) system was introduced allowing for a reproducible, high-resolution morphologic examination of vein wall structures with high sensitivity and specificity.¹³

This study evaluated the ability of eOCT to demonstrate immediate thermal changes after RFA and ELT using various treatment protocols. This investigation covered qualitative alterations such as the number and characteristics of ablations and perforations as well as quantitative changes such as the post-treatment increase of media thickness and the reduction of vessel lumen diameter. The study also investigated whether these tissue effects correlate with the applied laser energy level.

Back to Article Outline

Methods 

Ex vivo model 

The ex vivo model consists of the left hind foot of a cow (18 to 24 months old and weighing 550 to 650 kg) slaughtered 1 to 3 hours before the investigation. This foot contains a subcutaneous vein 20.0 to 25.0 cm in length and complete vessel diameter of 3.5 to 5.5 (V. saphena lateralis and V. digitalis dorsalis communis III), which is suitable for treatment with clinical systems for endovenous RFA and ELT (Fig 1.).¹²

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Cow foot model (left hind extremity) for standardized evaluation of endovenous thermal procedures. The 50-mm study vein segments of vena saphena lateralis and vena digitalis dorsalis communis II used in this experimental protocol are marked. Endovenous thermal treatment was performed after repositioning of skin flap without perivascular tumescence.

In this experimental protocol, which was approved by the Institutional Review Board of the University of Munich, the overlaying skin was removed and the V. digitalis dorsalis communis III was cannulated distally by a sheath system (6F for RFA, 5F for ELT, 4F for control veins). Superficial veins were reperfused with heparinized blood and an intraluminal pressure of 5.0 to 10.0 cm H₂O was established for 15 seconds. For endovenous thermal treatment, the overlaying dissected skin flap was repositioned, and blood flow through the sheath system was reduced to approximately 6.0 mL/min, resulting in no significant positive intraluminal pressure. Perivenous tumescence was not used in this series. A total of 19 ex vivo specimens were examined, each foot containing a 50-mm study vein length of V. saphena lateralis or V. digitalis dorsalis communis III. Thirteen blood-reperfused vein segments (SV1 to SV13) were treated either with RFA (n = 2) or ELT (n = 11) at various energy levels. Six control vein segments (CV1 to CV6) were examined without endoluminal treatment.

Endovenous radiofrequency ablation 

The Closure System (VNUS Medical Technologies, San Jose, Calif) was used for RFA. A 6F 60-mm catheter (Closure-Plus) was continuously pulled back manually. Because of the feedback system, a constant setpoint temperature of 85°C was maintained at the probe tip and the vessel wall, resulting in a mean pullback velocity of 0.5 mm/s.

Endovenous laser therapy 

ELT was performed using a diode laser (CERALAS D15 ELVeS, CeramOptec GmbH, Bonn, Germany) emitting light of the wavelength λ = 980 nm, transported through a 600-μm bare-tipped optical fiber to the endovenous treatment area through a 5F sheath system. In eight treated vein segments, laser energy was applied by interval irradiation. Laser power of 3, 5, or 7 W was applied for an irradiation time of 1.5 second per spot. A paced pullback of the laser fiber together with the sheath resulted in laser irradiation spots at 3-mm intervals. In three treated vein segments (SV7 to SV9), laser energy was applied by continuous irradiation. Laser power was set to 11W with a continuous pullback velocity of 3.0 mm/s of the laser fiber in all three segments.

To examine changes in effects on tissue using different laser energy levels, the linear endovenous energy density (LEED)¹⁴ was calculated according to equation: LEED (J/cm) = LP (W) × [IT (sec)/VL (cm)], where LP is laser power, IT is irradiation time, and VL is the treated vein length.

The IT value for a complete treated vein segment was 25 seconds with paced pullback and 16.7 seconds with continuous pullback treatment. The LEED with continuous pullback was 36.5 J/cm using 11 W. Using paced pullback, LEED values were 35.0 J/cm using 7 W, 25 J/cm with 5 W, and 15 J/cm with 3 W.

The endovenous fluence equivalent (EFE)¹⁴ was also calculated for every ELT-treated vein segment using the equation: EFE (J/cm²) = LP (W) × IT (sec)/VS (cm²), which describes the energy level applied to inner vein surface (VS), with VS (cm²) = 2 π × LR (cm) × VL (cm), with LR the vein lumen radius.

A detailed list of treatment protocols of all treated vein-segments is given in Table I, Table II. LEED and EFE were calculated according to laser power, pullback protocol, and vein-lumen radius, respectively, for all ELT-treated vein segments.

Table I. Treatment parameters and qualitative results of tissue ablation and vein wall perforation before and after treatment in 13 study vein segments

Code†	Treatment	Power (W)	Type	Pullback Velocity (mm/s)⁎	LEED (J/cm)	EFE (J/cm2)	ABL (n) (mm)	PER (n) (mm)
SV1	RFA	0-6	C	0.5	RFA	RFA	0	0
SV2	RFA	0-6	C	0.5	RFA	RFA	0	0
SV3	ELT	7.0	P	P	35.0	48.9	14	0
SV4	ELT	7.0	P	P	35.0	47.2	12	0
SV5	ELT	7.0	P	P	35.0	40.5	0	16
SV6	ELT	7.0	P	P	35.0	26.1	14	0
SV7	ELT	11.0	C	3.0	36.5	88.0	25mm	0mm
SV8	ELT	11.0	C	3.0	36.5	52.3	10mm	10mm
SV9	ELT	11.0	C	3.0	36.5	42.6	20mm	0mm
SV10	ELT	5.0	P	P	25.0	64.7	7	6
SV11	ELT	5.0	P	P	25.0	47.1	6	0
SV12	ELT	3.0	P	P	15.0	24.9	7	0
SV13	ELT	3.0	P	P	15.0	15.4	3	0

LEED, Linear endovenous energy density; EFE, endovenous fluence equivalent; ABL, ablation; PER, perforation; RFA, radiofrequency ablation; C, continuous pullback of treatment device; ELT, endovenous laser therapy at 980 with 600 bare fiber; P, paced pullback of treatment device; SV, study vein.

SV7-SV9 experiments were performed with continuous pull back. Thermal effects (ABL-Ablations, PER-Perforations) result in longitudinal lesions, which were given in mm complete length. All other experiments were performed with stepwise pullback which results in multiple, very localized thermal lesions, which were given as the total number of all lesions of a segment (n-number).

Table II. Quantitative results of media thickness and lumen diameter before and after treatment in 13 study vein segments

Diff, Difference between values before and after intervention; LD, Lumen diameter; MT, media thickness.

Endovascular optical coherence tomography 

High-resolution optical in situ examination was performed by means of a prototype eOCT system (System M1, LightLab Imaging Inc, Boston, Mass). The eOCT-system was initially developed for endovascular inspection of the vessels wall and has been evaluated in arteries¹⁵ and other hollow structures.¹⁶ Previous studies in our group proved a morphologic correlation between histologic cross sections and eOCT cross sections of cow foot veins using this ex vivo model.¹³ As an example, Fig 2 demonstrates similar layered structures identified as intima (with lamina elastica interna), media, and adventitia.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Correlation of (A) histologic cross section (hematoxylin-eosin stain, original magnification ×40) and (B) endovascular optical coherence tomography (eOCT) cross section of a untreated control cow’s foot vein. A, In histologic cross section, circumferentially intact intima (I) with lamina elastica interna, media (M), and adventitia (A) is represented. B, In eOCT cross section intima (I) and lamina elastica interna is represented by a bright signal. Media is represented by signals with changing intensity; collagenous fibers and elastic fibers may increase signal intensity. There is a sharp boundary (marked by white arrows) between media and adventitia, which may be caused by lamina elastica externa. Owing to limited penetration depth, the transition from adventitia to perivascular tissue is less sharply defined. P, eOCT probe.

In this study, before and immediately after endovenous thermal treatment, the vein segments were examined by eOCT using this prototype rotating endoluminal device. Before introducing the eOCT-device and during eOCT inspection, veins were flushed with normal saline, establishing a constant intraluminal pressure of 5.0 to 10.0 cm H₂O. The eOCT-light source was a superluminescence diode-emitting light at a centered wavelength of λ = 1300 nm, with a coherence length of 10 to 15 μm.

For cylindrical measurements, the light is fed through an appropriate optical fiber that rotates in a silicon sheath (d_p = 400 μm; d = diameter, p = probe). To obtain a cylindrical cross section, the light is deflected out vertically from the lumen axis by means of a mirror system on the tip of the fiber. The light is remitted from structures of the tissue, detected by the same fiber, and transported to the measuring interferometer system, which allows a spatial resolution of 15 to 20 μm. Depending on the optical properties of the vessel wall, the maximum depth from which a signal could be detected was limited to 1.5 to 2.0 mm.

To receive a measurement of a certain vein segment, the fiber was withdrawn within the catheter by a motorized pullback system at a rate of 1.0 mm/s. During this study, the picture frame rate was set to 5 Hz. A total of 264 eOCT cross-section pictures were obtained for each examined 50-mm-long vein segment.

Microscopy 

After in situ examination with eOCT, vein segments were harvested and prepared for histologic investigation by perfusion fixation with 4% formaldehyde. Ten defined 7-μm histologic cross sections (hematoxylin and eosin [HE] stained) of each treated vein segment and each nontreated control vein segment were obtained at 5-mm intervals. During light microscopy and descriptive evaluation, HE cross sections were correlated with corresponding pretreatment and post-treatment eOCT cross sections, both for position as well observed structures (eg, tissue layers, thermal effects).

Qualitative and quantitative evaluation before and after treatment 

All 264 post-treatment eOCT cross sections of every treated vein segment were analyzed by counting the number of significant tissue ablations (>25% of media thickness) and complete vein wall perforations. The number of ablations and perforations was listed for each vein segment.

After the identification of morphologic structures of the vein walls, a quantitative evaluation of thermal changes in media thickness and lumen diameter, based on correlating pretreatment and post-treatment eOCT cross sections was performed electronically by means of the TapeMeasure software (INDEC Systems, Mountain View, Calif). This procedure was used to analyze 26 of 264 eOCT cross sections (ie, 2-mm intervals within each) for each treated vein segment (before and after treatment) as well as for each control vein segment (Fig 3). TapeMeasure was used to assist in analyzing the eOCT images by marking the vessel lumen and then the identifiable media-adventitia borders.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Endoluminal optical coherence tomography (eOCT) and TapeMeasure for quantitative evaluation of thermal effects such as heat-induced collagen shrinkage with increase of media thickness and reduction of vessel lumen. A. demonstrates eOCT cross section of a nontreated cow foot vein with marked lumen-intima border and media-adventitia border for computer-based calculation of vessel lumen diameter and intima-media thickness. P, eOCT probe. B. is a schematic representation of vessel wall layers, which can be identified by eOCT and parameters calculated by TapeMeasure. L, vessel lumen; M, media; A, adventitia; L, lumen diameter; MT, media thickness.

After the calibration of the TapeMeasure software, the media thickness and the lumen diameter could be calculated for each of the 26 pretreatment and post-treatment cross section images per vein segment to be studied. Pretreatment and post-treatment changes were calculated in percentages for RFA-treated and ELT-treated vein segments and correlated with applied energy level (LEED, EFE) for ELT vein segments.

Statistical analysis 

Statistical analysis was performed with the software JMP 3.2.6 (SAS Institute Inc, Cary, NC). The mean media thickness and mean lumen diameter of a particular 50-mm treated vein segment was calculated as the arithmetic mean value of 26 TapeMeasure measurements taken at 2-mm intervals. The nonparametric Wilcoxon test was used to compare 26 values of media thickness and lumen diameter before treatment with 26 values after treatment. A P < .05 was considered to be significant.

Back to Article Outline

Results 

All eOCT cross sections after treatment demonstrate similar thermal tissue alteration when compared with corresponding, standardized histologic cross sections after treatment. As an example, Fig 4 demonstrates thermal changes of vessel wall structures after RFA in the area of a vein valve (study vein SV2). Thermal alterations are evenly distributed around the circumference in all cross sections examined. The corresponding histologic cross section confirms the circular disintegration of intima and media with delamination and circular fissures. The representation of thermal tissue alteration in post-ELT cross sections matches perfectly with corresponding standardized post-ELT histologic cross sections.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Endovenous radiofrequency ablation (RFA) (panels A-C). Corresponding cross sections of a cow’s foot study vein (SV2) with valve structures. A, Endoluminal optical coherence tomography (eOCT) cross section of a cow foot vein before RFA treatment representing normal intima, media and adventitia as well as valve structures. B, Corresponding eOCT cross section after RFA with reduced vessel lumen diameter and distinct media thickening of complete circumference. Valve structures seem to be thickened also. Media shows circular homogenization of signal intensity. No localized loss of tissue (ablation) is visible. C, Corresponding post RFA histologic (hematoxylin-eosin stain, original magnification ×40) cross section represents circular destruction of intima. Media appears to be thickened, comprising the complete circumference, Nuclear rarefication, loss of cell contours, and delamination with circular fissures demonstrate thermal injury of media. No areas with carbonization or transmural thermal lesions including adventitia are visible.

Fig 5 shows the corresponding pretreatment eOCT, post-treatment eOCT, and post-treatment histologic cross section after ELT with paced pullback using 5 W in study vein segment SV11 (LEED, 25.0 J/cm). The corresponding postinterventional histologic cross section confirms the spot-like localized tissue ablation in the area of increased thermal alteration and only minor tissue changes in the vein wall opposite.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 5. 

  Endovenous laser therapy (ELT) with paced pullback at λ = 980 nm, 5W, and linear endovenous energy density at 25 J/cm. Panels A-C, Corresponding cross sections of a cow foot study vein (SV11). A, Endoluminal optical coherence tomography (eOCT) cross section before representing normal vein wall layers. B, Corresponding post ELT cross section with circular homogenization of media signal intensity. Arrow shows localized loss of tissue (ablation) including intima and 50% of media thickness. C, Corresponding histologic cross section (hematoxylin-eosin stain, original magnification ×40) represents localized thermal injury (arrow) with carbonization and tissue ablation including intima and 50% of media thickness, which corresponds with localized increased thermal injury. Media in this quadrant of the vein circumference additionally shows delamination with circular fissures. Intima and media in other quadrants of the circumference seems to be intact. No thermal injury has occurred to the adventitia.

Fig 6 shows a further example of the findings after ELT with 7 W (LEED, 35.0 J/cm) in SV5. Here massive thermal damage and total wall perforation can be seen in eOCT to match the corresponding histological cross section.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 6. 

  Endovenous laser therapy (ELT) with paced pullback at λ = 980 nm, 7 W, and laser endovenous energy density at 35 J/cm. Panels A-C, Corresponding cross sections of a cow’s foot study vein (SV5). A, Endoluminal optical coherence tomography (eOCT) cross section before ELT. B, Corresponding eOCT cross section after ELT representing localized massive thermal alteration with loss of tissue including all vein-wall layers leading to perforation (arrow). Circular homogenization and thickening of media. C, Corresponding histologic cross section (hematoxylin-eosin stain, original magnification ×40) representing massive tissue destruction (arrow), including perivascular tissue leading to complete vessel wall perforation. Carbonization of intima, media, and adventitia is visible. Circular thermal alteration is represented by destruction of intima and repetitive delamination of media with circular fissures.

Qualitative analysis of thermal effects on tissue by endovascular optical coherence tomography 

In eOCT cross sections after RFA, the transitions between the different vein wall layers become more diffuse, resulting in a less layered signal pattern from the vein wall structures. The boundary between media and adventitia is less sharply defined. No significant tissue ablation or complete vein wall perforation is visible in any of the post-RFA eOCT cross sections.

The eOCT cross sections after ELT demonstrate repetitive spot-like intima-media tissue defects (ablations) or complete vein-wall perforations every 3 mm, corresponding to paced laser irradiation pullback. By contrast, ablations are semicircular and longitudinal after continuous pullback treatment. The number of tissue ablations (>25% of intima-media thickness) and perforations detected per 50 mm of treated vein segment is listed in Table I. The number of ablations or perforations seems to be higher after paced, pullback treatment with 7 W (LEED, 35.0 J/cm) and 5 W (LEED, 25.0 J/cm) than with 3 W (LEED, 15.0 J/cm).

By analyzing eOCT cross sections, it was possible to exactly measure the length of longitudinal ablations and perforations after ELT with continuous pullback. This was not possible using analysis of histologic cross sections. The length in millimeters of longitudinal ablations and perforations after ELT with 11 W and continuous irradiation and pullback with LEED at 36.5 J/cm is also listed in Table I.

In general, the size of tissue defects after ELT is variable in all the vein segments, whether after paced pullback or continuous pullback treatment. Effects range from limited tissue ablations to transmural vessel wall defects with complete perforations. Vein wall structures opposite tissue ablation or perforation often show no or only minor thermal alteration. Fewer histologically detected spot-like tissue ablations or perforations were found after ELT with paced pullback (10 cross sections per treated vein segment) than in eOCT cross sections (264 cross sections per treated vein segment).

Quantitative analysis of media thickness and vessel lumen with endovascular optical coherence tomography and TapeMeasure 

After RFA, a statistically significant increase of the mean media thickness and a reduction of mean vessel lumen diameter could be determined in all post-treatment eOCT cross sections compared with pretreatment cross sections (Table II). In Fig 7, A, the mean media thickness values along SV2 are represented before and after RFA, as measured with eOCT and TapeMeasure, and a significant increase is demonstrated (+66.7%, P < .0001). In Fig 7, B, the mean lumen diameter along the same treated vein segment showed a significant reduction after RFA (−41.2%, P < .0001).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 7. 

  A, Example of media thickness (MT) subject to the position of the vein length treated with radio frequency ablation (RFA) standardized TapeMeasure measurement in study vein segment SV2. B, Example of lumen diameter (LD) subject to the position of the RFA-treated vein length standardized TapeMeasure measurement in study vein-segment SV2.

In the same way, a significant increase of mean media thickness compared with pretreatment baseline values can be demonstrated after ELT in all treated vein segments (SV3 to SV13). Fig 8, A, demonstrates the media thickness values along SV6 before and after ELT with 7 W. Mean values (see also Table II) confirm a significant increase after treatment (+13.0%, P < .0001). Regular localized tissue ablations are visible and marked on the curve corresponding to points affected by paced pullback. In Fig 8, B, a significant reduction was found of the mean lumen diameter along the same study vein segment after ELT (−41.5%, P < .0001).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 8. 

  A, Example of media thickness (MT) subject to the position of the vein length treated after paced pullback treatment with endovenous laser therapy (ELT) at λ = 980 nm, laser endovenous energy density at 35 J/cm, and endovenous fluence equivalent at 26.1 J/cm². Standardized TapeMeasure measurement in study vein-segment SV6. Thin diamond, tissue ablation, 25% to 50% of MT; thick diamond, tissue ablation >50% of MT. B, Example of lumen diameter (LD) subject to the position of the ELT-treated vein length. Standardized TapeMeasure measurement in study vein-segment SV6.

Values of media thickness before and after ELT with 7 W of SV5 are shown in Fig 9. A significant increase of mean media thickness (+45.7%, P < .0001) and a significant reduction of mean lumen diameter (−45.1%, P < .0001) were achieved; however, repetitive complete vessel wall perforations were detected corresponding to points affected by paced pullback.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 9. 

  Example of media thickness (MT) subject to the position of the vein length treated with paced pullback endovenous laser therapy (ELT) at λ = 980, laser endovenous energy density at 35 J/cm, and endovenous fluence equivalent at 40.5 J/cm². Standardized TapeMeasure™ measurement in study vein segment SV5. Solid diamond shows the localization of complete vessel wall perforations.

Fig 10 shows values of media thickness before and after ELT with 11 W and continuous pullback in the treated vein segment SV9. Again a significant increase of mean medial thickness (+14.8%, P < .0001) and a significant reduction of mean lumen diameter (−13.6%, P < .0001) were calculated. As a result of continuous irradiation pullback, eOCT detects longitudinal tissue ablations but no localized repetitive ablations.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 10. 

  Example of media thickness (MT) subject to the position of the vein length treated with continuous pullback endovenous laser therapy (ELT) at λ = 980 nm, laser endovenous energy density at 36.5 J/cm, and endovenous fluence equivalent at 42.6 J/cm²) vein length. Standardized TapeMeasure measurement in study vein-segment SV9. Thick black line shows tissue ablation >50% of media thickness.

The extent of pretreatment and post-treatment changes of mean media thickness or mean vein lumen diameter shows no direct relation to the LEED or EFE applied to the particular vein segment. The large variations in EFE that appeared in our calculations although the same LEED applied were related to the different vein lumen diameters (see Table I).

The results of measurements of control veins (CV1 to CV6), which are listed in Table III, demonstrated certain interindividual variations. The mean baseline value of media thickness of all examined vein segments (SV1 to SV13 and CV1 to CV6) is 0.53 mm for V. saphena lateralis (VSL) and 0.39 mm for V. digitalis dorsalis communis III (VDD). The mean baseline value of lumen diameter is 2.81 mm for VSL and 1.84 mm for VDD.

Table III. Mean media-thickness and mean vein lumen-diameter in six control vein segments^⁎ using standardized endoluminal optical coherence tomography and TapeMeasure measurement

CV, Control vein segment; MT, media thickness; LD, lumen diameter.

Back to Article Outline

Discussion 

Endoluminal procedures for treating saphenous vein insufficiency offer an attractive, minimally invasive alternative to conventional surgical procedures such as high ligation (crossectomy) and stripping. Although endovenous RFA has become a largely standardized procedure,¹, ¹⁷, ¹⁸, ¹⁹ ELT is characterized by a great variety of treatment modifications that use different laser wavelengths, energy levels, and pullback protocols.⁸, ⁹, ¹⁴, ²⁰, ²¹ Most of the treatment modifications have been used in clinical practice without previous systematic experimental evaluation. Publications of experimental investigations under standardized conditions that achieve reproducible data for the existing different endovenous thermal procedures and treatment protocols are rare.

To investigate the thermal effects on tissue of RFA and different ELT treatment protocols and technical modifications, we used the cow foot to develop an appropriate ex vivo model¹² that can investigate acute thermal effects on tissue. The major objective of endovenous therapy of insufficient saphenous veins remains the irreversible occlusion of the insufficient saphenous vein that results from post-traumatic healing processes in the weeks after thermal alteration. This process can only be investigated in living animal models or clinically in humans. To date, it is unclear how results obtained from this ex vivo model, which treats the veins in the foot of a healthy young cow, can be extrapolated into the clinical situation. However, identifying a certain treatment protocol that leads to a reproducible, effective complete circular vessel wall thickening (collagen contraction) and avoids complete transmural or perivascular thermal injury in an ex vivo model could be useful in reducing the number of experiments with living animals or the number of clinical trials.

Histologic cross sections offer good, high-resolution visualization of thermally affected tissue and cell alterations²²; however, they always represent only a limited section of a complex thermal damage that spreads in a circular and longitudinal direction along the wall of a treated vein segment. To obtain a complete histologic analysis of a longer venous segment, a large number of histologic sections are required. This consumes time and many other resources.

Once the morphologic vessel wall structures are identified, the dynamic, spiraling examination procedure of eOCT offers the opportunity to obtain high-resolution optical images and analyze complete vessel segments simultaneously in the circular and longitudinal direction. Beyond that, computer-based electronic processing of eOCT pictures allows quantitative analyses. It is not necessary to remove the vein from the surrounding tissue for eOCT evaluation. Thermal vein wall injury can be investigated without destroying the biologic model, which may in some cases be live models.

Compared with intravascular ultrasound as the standard for endovascular optical diagnostic imaging, eOCT provides a substantially higher resolution that enables the investigator to distinguish different vessel wall layers. In the context of endoluminal vein treatment, presented eOCT will mainly be used in experimental settings. The eOCT device that was used in our experiments is still a prototype. When technical problems have been completely solved, eOCT could also be used in blood-filled vessels. It might be considered for clinical trials investigating effects on tissue after RFA and ELT as an additional instrument alongside transcutaneous duplex ultrasound imaging, but not in clinical routine. Further indications for the use of eOCT in vascular surgery might include investigating saphenous veins in situ to evaluate the suitability for venous bypasses or for intraoperative quality control of bypass anastomoses. Beyond that, the monitoring of percutaneous intraarterial or of intravenous interventions such as balloon angioplasty and stent implantation should be possible.

Endovenous thermal treatment induces protein denaturation and destruction of cell structures and leads to collagen contraction resulting in thickening and induration of the vein wall.¹², ²³, ²⁴ These alterations can be confirmed in duplex sonography and histologic studies.²², ²³, ²⁵, ²⁶ This study demonstrates the ability of eOCT to detect tissue ablation and perforation and offers the potential for quantifying immediate thermal alteration such as increased media thickness and reduced vessel-lumen diameter. Histologic and eOCT cross sections after RFA showed reproducible, complete circular thermal alteration. Thermal alteration of vein wall layers after ELT varied, ranging from localized tissue ablation in a certain quadrant of the vein circumference, with minor changes in other quadrants, up to complete transmural ablations and perforations with circular thermal alteration. All treated vein segments showed a statistically significant increase of media thickness after ELT. The amount of media thickening seems to be more pronounced, completely circular, and reproducible after RFA than after ELT. A direct relation between the energy level applied (LEED, EFE) and the number of tissue ablations/perforations as well as the extent of media thickening and vein lumen reduction after ELT could not be found. This might be due to the limited number of treated vein segments investigated and the varying baseline parameters, especially pretreatment vein lumen diameter.

Calculations show very clearly how sensitively EFE reacts to vein lumen diameter. A mathematic model showed that knowledge of exact vessel diameter is the most important parameter in calculating the optimal laser energy dosage to achieve adequate tissue alteration without adverse effects.²⁷ In this series, eOCT measurements of vein lumen diameters showed large intraindividual and interindividual variations. This might be due to real differences and also to changing intravascular pressure during measurement. It was not always possible to maintain exactly the targeted intraluminal pressure of 5.0 to 10.0 cm/H₂0 during eOCT measurements. This shows that standardization of eOCT measurement for the exact calculation of EFE is needed.

In general, vein lumen diameters in this ex vivo cow foot model are smaller than usually reported in clinical trials in which insufficient saphenous veins were treated. For this reason, findings have to be interpreted with caution even when the laser power used and the calculated LEED were adapted to small vein diameters and when the extrapolated EFE might be similar to the clinical situation. We point out, however, that human saphenous veins can lose diameter by spasmic contraction after cannulating with probes for RFA or ELT, which can be clearly observed with transcutaneous ultrasound. Moreover, three further factors can reduce lumen diameter down to RFA and ELT probe diameter, especially distally of the saphenofemoral junction: head-down position of the patient, perivascular tumescence, and external manual compression. To our knowledge, no systematic (intraluminal) measurement of vein diameters in the complete course of a human saphenous vein immediately before or after the endovenous thermal treatment has been published to date. The small vein diameters used in our series therefore simulate the clinical situation when the fiber tip is located very near the vein wall or touches the vein wall.

In addition to vein lumen diameter, the vein length treated also seems to be an important variable in evaluating endovenous procedures. Thermal tissue alteration may change during treatment of a vein segment 40 to 50 cm long owing to increasing carbonization of tissue at the electrodes (RFA) or bare fiber tip (ELT).¹²

The large variation of thermal effects on tissue after ELT without a direct inter-relationship to used energy levels, which we observed in our series, might be due to very complex mechanisms of endovenous laser-induced thermal alteration, which are not yet completely understood. These mechanisms include varying amounts of photon absorption in erythrocytes (hemoglobin) and other blood components as well as induction of a plasma “steam bubble” that conducts heat to the vein wall.²⁸ Direct photon absorption in vein wall cells and structures has also been postulated.²⁹ In addition, photon absorption has been observed in carbonized tissue at the fiber tip, with subsequent significant heating of the fiber tip¹⁰ and direct conduction of heat energy into the blood and vessel wall. Energy level does not seem to be the single variable influencing effective circular collagen shrinkage, vein wall thickening, and contraction of vessel lumen in ELT. Vein content (blood or saline), vein lumen diameter, vein wall spasm, the position of the bare fiber tip in the lumen, and the distance of the fiber tip from the vessel wall may influence the effect on tissue more than laser wavelength or laser power, which was set at the generator.³⁰

Our experience with these experiments leads us to the conclusion that technical modifications need to be developed to standardize ELT-induced thermal alteration and subsequent effects on tissue. We envision, for example, the creation of a device to position the laser fiber in the center of vein lumen or at a defined distance from the vessel wall, mechanisms to avoid carbonization at the fiber tip, new fiber tip designs with circular irradiation, and automatic, computer-controlled pullback devices with feedback mechanisms that account for vein lumen diameter and temperature in the vein wall.

We believe that to effectively occlude an insufficient saphenous vein, endoluminal thermal procedures should damage the complete circumference and wall thickness of intima, internal elastic membrane, media, external elastic membrane, and adventitia. Semicircular thermal alteration that leaves parts of the vein circumference uninjured may lead to regeneration and recanalization. Perforations are a sign for massive full-thickness thermal alteration with the potential for ecchymosis, hematoma, or perivascular injury such as nerve lesions.

Efforts should be made to further improve existing biologic and mathematic models to investigate and understand the mechanisms of tissue alteration of endovenous thermal procedures. Existing treatment protocols and new technologic developments could then be reproducibly evaluated in a standardized manner, and the most effective endovenous procedures avoiding perivenous injury could be identified without ethical limitations. The results of these experiments may reduce the number of experiments using live animals and also the number of clinical trials with humans.

Back to Article Outline

Conclusions 

In our ex vivo cow foot model, eOCT is able to reproduce with high sensitivity and specificity normal vein wall structures as well as endovenous acute thermal alterations, such as tissue ablation and vessel wall perforations, and thus complements histologic examinations. The use of dedicated software permits eOCT images to be analyzed quantitatively to measure media thickness or vein lumen diameter.

After endovenous thermal therapy with RFA and ELT, a reproducible and statistically significant increase of media thickness can be detected. This might be the consequence of heat-induced collagen contraction. Compared with ELT with different treatment protocols, the amount of media thickening and consecutive vein lumen reduction is more pronounced after RFA.

Ex vivo models using macroscopic, histologic and high-resolution optical investigations may be useful for standardized evaluation of the acute effects on tissue of different treatment protocols and of new technical developments in endovenous thermal procedures.

Back to Article Outline

Author contributions 

Back to Article Outline

We should like to thank Dr Stephanie Steckmeier of the Clinic for Dermatology (Director, Prof Dr Dr H. C. T. Ruzicka) at the Ludwig-Maximilians-University of Munich for her great support in conducting the experiments. We also wish to thank Dr Johannes Rieber of the Cardiology Division (Head, Prof Dr K. Theisen) at the Ludwig-Maximilians-University of Munich, and also Holger Hetterich for support in processing the eOCT pictures. Our thanks also to Annie Hollins and Harriet Hasenclever for help with the editing of the English manuscript.

Back to Article Outline

References 

1. Merchant RF, Pichot O Closure Study Group. Long-term outcomes of endovenous radiofrequncy obliteration of saphenous reflux as a treatment for superficial venous insufficiency. J Vasc Surg. 2005;42:502–509
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (355 KB)
   • CrossRef
2. Puggioni A, Kalra M, Carmo M, Mozes G, Gloviczki P. Endovenous laser therapy and radiofrequency ablation of the greater saphenous vein: Analysis of early efficacy and complications. J Vasc Surg. 2005;42:488–493
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (206 KB)
   • CrossRef
3. Ravi R, Rodriguez-Lopez JA, Trayler EA, Barrett DA, Ramaiah V, Dietherich EB. Endovenous ablation of incompetent saphenous veins: A large single-center experience. J Endovasc Ther. 2006;13:244–248
   • View In Article
   • MEDLINE
   • CrossRef
4. Sharif MA, Soong CV, Lau LL, Corvan R, Lee B, Hannon RJ. Endovenous laser treatment for long saphenous vein incompetence. Br J Surg. 2006;93:831–835
   • View In Article
   • MEDLINE
   • CrossRef
5. Disselhoff BCVM, der Kinderen DJ, Moll FL. Is there recanalization of the great saphenous vein 2 years after endovenous laser treatment. J Endovasc Ther. 2005;12:731–738
   • View In Article
   • MEDLINE
   • CrossRef
6. Kim HS, Nwankwo IJ, Hong K, McElgunn SJ. Lower energy endovenous laser ablation of the great saphenous vein with 980nm diode laser in continuous mode. Cardiovasc Intervent Radiol. 2006;29:64–69
   • View In Article
   • MEDLINE
   • CrossRef
7. Proebstle TM, Gül D, Lehr HA, Kargl A, Knop J. Infrequent early recanalization of greater saphenous vein after endovenous laser treatment. J Vasc Surg. 2003;38:511–516
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (416 KB)
   • CrossRef
8. Proebstle TM, Krummenauer F, Gul D, Knop J. Nonocclusion and early reopening of the great saphenous vein after endovenous laser treatment is fluence dependent. Dermatol Surg. 2004;30:174–178
   • View In Article
   • MEDLINE
   • CrossRef
9. Timperman PE, Sichlau M, Ryu RK. Greater energy delivery improves treatment success of endovenous laser treatment of incompetent saphenous veins. J Vasc Interv Radiol. 2004;15:1061–1063
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (54 KB)
   • CrossRef
10. Weiss RA. Comparison of endovenous radiofrequency versus 810 nm diode laser occlusion of large veins in an animal model. Dermatol Surg. 2002;28:56–61
    • View In Article
    • MEDLINE
    • CrossRef
11. Zikorus AW, Mirizzi MS. Evaluation of setpoint temperature and pullback speed on vein adventitial Temperature during endovenous radiofrequency energy delivery in an in-vitro model. Vasc Endovasc Surg. 2004;38:167–174
    • View In Article
12. Schmedt CG, Sroka R, Steckmeier S, Meissner OM, Babaryka G, Hunger K, et al. Investigation on radiofrequency and laser (980 nm) effects after endoluminal treatment of saphenous vein insufficiency in an ex-vivo model. Eur J Vasc Endovasc Surg. 2006;32:318–325
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (4373 KB)
    • CrossRef
13. Meissner OA, Schmedt CG, Hunger K, Hetterich H, Sroka R, Rieber J, et al. Endovascular optical coherence tomography ex vivo: venous wall anatomy and tissue alterations after endovenous therapy. Eur Radiol. 2007;Feb 8; [epub ahead of print].
    • View In Article
14. Proebstle TM, Moehler T, Gül D, Herdemann S. Endovenous treatment of the great saphenous vein using a 1340nm Nd:YAG Laser causes fewer side effects than using a 940nm diode lser. Dermatol Surg. 2005;31:1678–1683
    • View In Article
    • MEDLINE
    • CrossRef
15. Meissner OA, Rieber J, Babaryka G, Oswald M, Reim S, Siebert U, et al. Intravascular optical coherence tomography: comparison with histopathology in arteriosclerotic peripheral artery specimens. J Vasc Interv Radiol. 2006;17:343–349
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (2124 KB)
    • CrossRef
16. Mueller-Lisse UL, Meissner OA, Babaryka G, Bauer M, Eibel R, Stief CG, et al. Catheter-based intraluminal optical coherence tomography (OCT) of the ureter: ex-vivo correlation with histology in porcine specimens. Eur Radiol. 2006;10:2259–2264
    • View In Article
17. Chandler JG, Pichot O, Sessa C, Schuller-Pétrovic S, Kabnick LS, Bergan JJ. reatment of primary venous insufficiency by endovenous saphenous vein obliteration. Vascular Surgery. 2000;34:201–214
    • View In Article
18. 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;38:207–214
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (93 KB)
    • CrossRef
19. Lurie F, Creton D, Eklof B, Kabnick LS, Kistner RL, Pichot O, et al. Prospective randomised study of endovenous radiofrequency obliteration (closure) versus ligation and vein stripping (EVOLVeS): two-year follow-up. Eur J Vasc Endovasc Surg. 2005;29:67–73
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (156 KB)
    • CrossRef
20. Huang Y, Jiang M, Li W, Lu X, Huang X, Lu M. Endovenous laser treatment combined with a surgical strategy for treatment of venous insufficiency in lower extremity: A report of 208 cases. J Vasc Surg. 2005;42:494–501
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (243 KB)
    • CrossRef
21. Kabnick LS. Outcome of different endovenous laser wavelengths for great saphenous vein ablation. J Vasc Surg. 2006;43:88–93
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (161 KB)
    • CrossRef
22. Corcos L, Dini S, De Anna D, Marangoni O, Ferlaino E, Procacci T, et al. The immediate effects of endovenous diode 808-nm laser in greater saphenous vein: mrphologic study and clinical implications. J Vasc Surg. 2005;41:1018–1025
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (302 KB)
    • CrossRef
23. Manfrini S, Gasbarro V, Danielsson G, Norgren L, Chandler JG, Lennox AF, et al. Endovenous Reflux Management Study Group Endovenous management of saphenous vein reflux. J Vasc Surg. 2000;32:330–342
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (279 KB)
    • CrossRef
24. Min RJ, Khilnani N, Zimmet SE. Endovenous laser treatment of saphenous vein reflux: long-term results. J Vasc Interv Radiol. 2003;14:991–996
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1208 KB)
    • CrossRef
25. Pichot O, Kabnick LS, Creton D, Merchant RF, Schuller-Pétrovic S, Chandler JG. Duplex ultrasound scan findings two years after great saphenous vein radiofrequency endovenous obliteration. J Vasc Surg. 2004;39:189–195
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (255 KB)
    • CrossRef
26. Salles-Cunha SX, Rajasinghe H, Dosick SM, Gale SS, Seiwert A, Jones L, et al. Fate of great saphenous vein after radio-frequency ablation: detailed ultrasound imaging. Vascular and Endovascular Surgery. 2004;38:339–344
    • View In Article
    • MEDLINE
    • CrossRef
27. Mordon SR, Wassmer B, Zemmouri J. Mathematical modeling of endovenous laser treatment (ELT). Biomed Eng Online. 2006;5:26;Available at: www.biomedical-engineering-online.com/content/5/1/26. Last accessed Dec 19, 2006.
    • View In Article
    • MEDLINE
    • CrossRef
28. Proebstle TM, Lehr HA, Kargl A, Espinola-Klein C, Rother W, Bethge S, et al. Endovenous treatment of the greater saphenous vein with a 940-nm diode laser: thrombotic occlusion after endoluminal thermal damage by laser-generated steam bubbles. J Vasc Surg. 2002;35:729–736
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1106 KB)
    • CrossRef
29. Min RJ, Khilnani NM. Endovenous laser ablation of varicose veins. J Cardiovasc Surg. 2005;46:395–405
    • View In Article
30. Sroka R, Schmedt CG, Steckmeier S, Meissner OA, Beyer W, Babaryka G, et al. Ex-vivo investigation of endoluminal vein treatment by means of radiofrequency and laser irradiation. Med Laser App. 2006;21:15–22
    • View In Article