Here we compared effect of serum from varicose patients undergoing endovenous laser ablation (EVLA) and classic vein stripping (CVS) on biological properties of endothelial cells and on the local and systemic profiles of proinflammatory agents. Results showed that serum from EVLA patients improved proliferation and reduced senescence and oxidative stress in the endothelial cells, as compared with the serum from CVS patients. These effects were related to a suppressed activity of TGF-
Varicose veins belong to the most frequent lifestyle diseases, as they affect up to 40% of industrialized countries’ citizens in the age between 30 and 70 [
Currently there are two ways for varicose vein management: lifestyle modifications and medical procedures. Lifestyle-related recommendations include the avoidance of a prolonged standing and sitting, an intensification of physical exercise, a loosening of restrictive clothes, and losing weight by obese people [
Until the past few years, classic surgical methods of varicose vein removal—mainly vein stripping—were considered as the most radical and effective ways to cope with the pathology. On the other hand, traumatizing nature of these methods yielded several side effects [
In this paper we joined the debate and compared patients treated with CVS and EVLA in terms of an effect of their serum on such critical, functional features of endothelial cells, like proliferation, senescence, production of reactive oxygen species (ROS), and secretory properties. These in vitro investigations were followed by a comparative ex vivo analysis of both groups of sera with respect to a level of eight arbitrarily selected mediators of vascular inflammation. Last but not least, the postsurgery changes in the serum concentration of tested mediators were evaluated.
Unless otherwise stated, all chemicals and culture plastics were purchased from Sigma (St. Louis, MO). Exogenous, recombinant forms of human GRO-1 and TGF-
The study included 40 patients undergoing surgical treatment for primary varicose veins. All subjects had symptomatic disease and the inclusion criteria for the study were based on clinical examination. Half of patients received CVS and the second half was treated with EVLA. Both procedures were performed under spinal anesthesia. The exclusion criteria included ankle brachial pressure index (ABPI) < 0.8, previous deep vein thrombosis or pulmonary embolism, coagulopathy, malignancy, and pregnancy. Demographic data regarding patients are presented in Table
Characteristics of patients undergoing surgical varicose vein removal using CVS and EVLA.
Parameter | CVS | EVLA |
---|---|---|
| 20 | 20 |
Sex (male/female; | 9/11 | 6/14 |
Age (mean ± SD/range; | | |
Comorbidities (number) | ||
No comorbidities | 10 | 8 |
Hypertension | 3 | 4 |
Diabetes mellitus | 1 | 2 |
Hypothyroidism | 2 | 1 |
Asthma | 1 | 1 |
Obesity | 0 | 2 |
Psoriasis | 1 | 2 |
Allergy | 1 | 0 |
Adrenal insufficiency | 1 | 0 |
During CVS, two incisions were made: one at the groin and one at the ankle. The saphenous vein was then opened at each end and a flexible wire was inserted into the entire vein and then pulled out to remove the vein. During EVLA, an access to the distal part of the vein was achieved through percutaneous puncture under ultrasound control. The working device (810 nm wavelength axial diode laser; EVLA810) was introduced and advanced towards the saphenofemoral junction. After confirmation of the position of the tip of the working device (1-2 cm from the saphenofemoral junction) in ultrasonography, the ablation was performed with the aim to deliver 60–100 J for 1 cm of the treated vein. The study was approved by the institutional ethics committee (consent number 441/13).
The samples of serum were collected from patients undergoing surgery due to varicose veins of lower extremities 1 h before surgery (point A) and 6 h after surgery (point B). Blood was centrifuged immediately after collection and serum samples were stored in aliquots at −80°C until required.
Human umbilical vein endothelial cells (HUVECs) were purchased in the ATCC (Rockville, MD, USA). The cells were cultured in DMEM with 20% FBS, L-glutamine (2 mM), HEPES (20 mM), EGF (10
Concentration of proinflammatory agents was measured using appropriate DuoSet® Immunoassay Development kits (R&D Systems), according to manufacturer’s instructions.
Endothelial cells were seeded into culture dishes at low density (5 × 104 cells per well), allowed to attach for 2 h, and then growth synchronized by serum deprivation for 4 h. Afterwards, the cells were exposed to 20% serum obtained from patients undergoing CVS and EVLA (point B) for 72 h. After the incubation, cells were detached and counted, and the number of population doublings (the measure of cell proliferation) was calculated using the following formula: population doublings =
The activity of senescence-associated
ROS production was assessed in endothelial cells exposed to 20% serum (point B) from patients treated with CVS and EVLA for 72 h. In brief, 1 × 105 cells were incubated in the presence of 5
Statistical analysis was performed using GraphPad Prism™ v.5.00 software (GraphPad Software, San Diego, USA). The means were compared with Mann–Whitney
Low-density HUVECs were exposed to 20% serum from patients treated with CVS and EVLA for 72 h and then three parameters being the measures of cellular stress, that is, cell proliferation, activity of SA-
Effect of serum from patients treated with CVS and EVLA on the number of population doublings (a), the activity of SA-
Both groups of sera were compared with respect to a concentration of four arbitrarily selected cytokines (TGF-
Concentration of TGF-
In order to verify if TGF-
Effect of exogenous, recombinant TGF-
Conditioned medium from HUVECs exposed to sera (point B) from the EVLA patients contained decreased concentration of three out of eight investigated mediators of vascular inflammation, that is, ICAM-1, VCAM-1, and E-selectin, as compared with the cells exposed to sera from the CVS patients. In addition, these media contained increased level of uPA. Concentrations of the remaining agents (P-selectin, PAI-1, ET-1, and TFPI) in both groups of sera were comparable (Figure
Concentration of proinflammatory agents in conditioned medium generated by endothelial cells subjected to serum from patients with varicose veins treated with classic vein stripping (CVS) and endovenous laser ablation (EVLA). Endothelial cells were subjected to 20% serum collected 6 h after surgery for 72 h and then they were washed and exposed for the next 72 h to serum-free medium to generate conditioned medium in which proinflammatory agents were quantified. Results (expressed as means ± SEM) derive from experiments performed with the sera from 12 patients per group. Asterisks indicate significant differences as compared with CVS.
Ex vivo studies in which concentration of the same molecules was measured directly in the sera obtained from both groups of varicose patient showed that the samples from EVLA patients contained lower concentration of ICAM-1, E-selectin, and P-selectin and higher concentration of uPA, PAI-1, and TFPI, as compared with the serum from CVS patients. Concentration of two remaining agents, that is, VCAM-1 and ET-1, in both groups was similar (Figure
Concentration of proinflammatory agents in serum from the patients with varicose veins treated with the classic vein stripping (CVS) and the endovenous laser ablation (EVLA).The measurements were made using the appropriate ELISA kits using the sera collected 6 h after surgery. Results (expressed as means ± SEM) derive from the analysis of sera obtained from 20 patients per group. Asterisks indicate significant differences as compared with CVS.
In order to analyze results of CVS and EVLA in terms of changes in the concentration of tested proinflammatory agents, a level of a given agent in the serum obtained after surgery (point B) was compared with its level before the procedure (point A). This study revealed that in case of every mediator tested there is a group of patients in whom both types of surgery lead to an increase in certain agents’ level and a second group of patients in whom the level of those agents declines.
This heterogenous reaction prompted us to divide the patients into two groups (marked as “decreases” and “increases”) and to make some assumptions regarding desired changes. Namely, we assumed that if EVLA is truly less proinflammatory than CVS, it should lead to the decreases in the level of tested agents in a higher percentage of patients. Per analogy, any increases in the level of the tested agents upon EVLA should occur in a lower fraction of patients, as compared with CVS. Analysis performed using such an algorithm followed by a proportion analysis with Fisher’s exact test showed that abovementioned assumptions were met in case of two agents, that is, ICAM-1 and ET-1. For the rest of studied mediators, a surgery-related changes in their serum level had no statistical significance (Table
Surgery-related changes in serum concentration of proinflammatory agents.
Agent | Decreases | Increases | | ||||||
---|---|---|---|---|---|---|---|---|---|
CVS | EVLA | CVS | EVLA | ||||||
| % | | % | | % | | % | ||
ICAM-1 | 9/20 | 45 | 17/20 | 85 | 11/20 | 55 | 3/20 | 15 | Yes |
VCAM-1 | 11/19 | 58 | 5/19 | 26 | 8/19 | 42 | 14/19 | 74 | No |
E-selectin | 14/19 | 74 | 13/20 | 65 | 5/19 | 26 | 7/20 | 35 | No |
P-selectin | 11/20 | 55 | 9/19 | 47 | 9/20 | 45 | 10/19 | 53 | No |
uPA | 14/19 | 74 | 13/20 | 65 | 5/19 | 26 | 7/20 | 35 | No |
PAI-1 | 8/19 | 42 | 14/20 | 70 | 11/19 | 58 | 6/20 | 30 | No |
ET-1 | 8/20 | 40 | 15/20 | 75 | 12/20 | 60 | 5/20 | 25 | Yes |
TFPI | 13/19 | 68 | 9/19 | 47 | 6/19 | 32 | 10/19 | 53 | No |
Although the presence of varicose veins predisposes to the development of certain complications, medical treatment of the disease with a classic surgery (CVS) or new endovascular techniques (EVLA) may also result in various procedure-related morbidities [
Based on this we designed a project to compare the local (endothelium-related) and systemic (serum-related) outcomes of CVS and EVLA. Experiments showed that EVLA is less stressful for endothelial cells than CVS, as the serum obtained from patients undergoing this procedure fueled their proliferation and reduced senescence and oxidative stress. This improved functionality of endothelial cells upon EVLA, in particular decreased senescence, is of special importance with respect to the spreading of the pathology within normal vessels, of which conclusion stems from the histopathological analysis by Aunapuu and Arend showing that endothelial cells from varicose patients display specific discontinuity and denudation that may be a manifestation of cellular senescence [
Comparative analysis of sera from the CVS and EVLA groups showed that the former contained increased level of TGF-
Another indicator of endothelial cell dysfunction that was tested in the project was the cells’ ability to secrete eight arbitrarily selected mediators of vascular inflammation, that is, adhesion molecules (ICAM-1, VCAM-1, E-selectin, and P-selectin), mediators of coagulation and fibrinolysis (uPA, PAI-1, and TFPI), and a vasoconstrictive agent, ET-1. Interestingly, cells exposed from serum from the EVLA patients adopted clearly less proinflammatory phenotype, as they secreted decreased amounts of ICAM-1, VCAM-1, and E-selectin. This finding implying that EVLA is, indeed, less traumatizing to endothelial cells than CVS agrees with other reports, for example, those describing that endothelium experiencing stressful conditions (elevated shear stress or expanded application of carbon nanotubes) exhibits increased expression of ICAM-1 and VCAM-1 [
Having established that EVLA exerts more favorable effects than CVS in local, cell-related conditions, we went to analyze the differences in the serum level of the same proinflammatory molecules that has been tested in vitro. Also in this case, EVLA suppressed the inflammatory serum phenotype, as the samples from patients treated with this procedure contained lower levels of ICAM-1, E-selectin, and P-selectin. The similarity of effects observed on cellular (conditioned medium) and systemic (serum) levels with respect to ICAM-1 and E-selectin may suggest that reduced serum content of these agents may directly result from their downregulated secretion by vascular endothelium. Taking into account the fact that both these molecules are typically shed to environment from the surface of activated endothelial cells [
As per a direct outcome of a surgery (point B versus point A), the results obtained appeared to be very intriguing, as they showed a diversified patients’ reaction. In fact, both the decreases and the increases in proinflammatory agent level upon both procedures were observed, confirming findings by Nesbitt et al. about high variability of varicose patient responses to a surgery [
Collectively, our findings showing that EVLA is less traumatizing to vascular endothelium and generates lower level of proinflammatory mediators than CVS may suggest that cellular and systemic changes resulting from this procedure are safer for varicose patients from the perspective of the development of further endothelium- and inflammation-dependent morbidities. At the same time, it should be emphasized that that our findings point on the differences between these two procedures at the cellular level. To a significant extent, different outcomes of CVS and EVLA are also associated with different degree of tissue destruction. During the stripping the entire vein is removed, side branches are disrupted, and bleeding occurs into the stripping channel. During the laser technique, the vessel is left in place and damaged by heat from inside without concomitant injury of the surrounding tissue and bleeding. For this reason, either factors associated with altered cell functionality or technique characteristics should be taken into account during the final selection of a method of varicose vein surgery.
The authors declare that there are no competing interests related to the publication of this paper.
Paweł Uruski and Krzysztof Aniukiewicz contributed equally to this paper.
This work was supported by a grant from the National Science Centre, Poland (decision no. DEC-2011/03/B/NZ3/01214).