Genetic polymorphisms of vein wall remodeling in chronic venous disease: a narrative and systematic review

Chronic venous disease (CVD) refers to a spectrum of overlapping diseases involving abnormalities of the venous system, both structural and functional. Chronic venous insufficiency (CVI) is a term that describes functional abnormalities of the venous system, but is often used to describe the full range of CVD manifestations such as varicose veins, venous ulceration, lipodermatosclerosis (LDS), and postthrombotic syndrome (PTS) (essentially a secondary form of CVI).^1-3  Almost a quarter of the adult population in the Western world has some form of CVD, but treatment is often delayed or deferred because of an underestimation of the prevalence and burden of the condition, leading to significant disability. Treatment is multifactorial, and often includes mechanical and/or surgical measures such as compression devices, leg elevation, ablations, and vein stripping.¹ 

CVI is a multifactorial disease, and although most commonly caused by valvular incompetence and venous hypertension, the exact pathogenesis remains unclear, although various studies have suggested a potential genetic contribution.⁴  This condition may be primary (abnormalities of vein walls or valves) or secondary (after venous thrombosis; ie, PTS). Risk factors include age, female gender, pregnancy, family history, obesity, and prolonged orthostasis. Clinical manifestations include leg pain, lower extremity edema, skin changes, varicose veins, and venous ulceration.⁵  Venous ulcers are a severe complication of CVI, although the underlying mechanisms are not fully known.

Varicose veins, a form of CVD and the most common manifestation of CVI, are caused by a loss of vessel wall homeostasis; venous hypertension leads to vein dilation, distortion, leakage, and inflammation, causing valve and wall disease and reflux.^6-8  Although sometimes caused by PTS, varicose veins are usually primary in nature. They are often hereditary and result from extensive extracellular matrix (ECM) remodeling, leading to vein wall weakening or dysfunction.^9-11 

PTS, a type of CVD, is an important and frequent chronic complication of deep venous thrombosis (DVT) that develops in 20% to 50% of patients (severe in 5% to 10%) after DVT despite appropriate anticoagulation.¹²  Risk factors include incomplete DVT symptom resolution, proximal or previous ipsilateral DVT, obesity, and increased age.¹²  Although the pathophysiology of PTS is not well understood, it involves venous hypertension caused by persistent venous obstruction and/or valvular reflux from valve destruction, and recent evidence also supports a role for inflammation.¹³  Symptoms are similar to CVI and can be debilitating, often including constant or intermittent limb swelling, aching, cramps, or numbness/tingling (improved with rest or recumbency). Treatment involves symptomatic relief using graduated elastic compression stockings or compression devices, leg elevation, or a trial of horse chestnut seed extract as a last resort.^12,14,15 

Recently, there has been some research into the genetic contributors and risk factors of CVD, such as varicose veins and venous ulceration, mainly relating to certain noted polymorphisms in genes associated with vein wall remodeling. In fact, genetic risk factors are already known to affect wound progression and healing, and screening in this regard may aid in the planning of appropriate individualized treatment and prophylaxis.¹⁶  Also, various thrombophilic single nucleotide polymorphisms (SNPs) may contribute to CVI and ulcers by increasing the risk of DVT.

More recently, however, other SNPs have been studied in genes relating specifically to vein wall remodeling and proinflammatory or angiogenesis-regulating factors and receptors. Although studies in this area do support a moderate to strong genetic predisposition toward CVI and varicose veins,¹⁶  the exact genes are still unknown—though they likely involve quantitative or qualitative defects in proteins associated with the vein wall, ECM, and cell organization/regulation.¹⁶  These polymorphisms, in turn, could plausibly increase the likelihood of CVD, including PTS after DVT, but evidence in this area is lacking.

Though various candidate genes have been identified as contributors to CVD, the evidence has not previously been compiled and critically reviewed. In this review, we use a systematic approach to explore the current evidence regarding genetic polymorphisms in vein wall structure and healing/remodeling as potential contributors to CVD, including PTS. Understanding these genetic associations may help clarify the underlying pathophysiology, identify patients at risk, and suggest novel therapeutic options.

To identify potential genes associated with CVD, we conducted a broad computerized literature search. We considered for inclusion any studies, either basic or clinical, describing or potentially describing an association between any gene, gene mutation, or genetic polymorphism and any of the following conditions: varicose veins, chronic leg ulcers, CVI, lipodermatosclerosis, PTS, or any combination of these. The search was conducted in September 2013 from the inception of each of the following databases: EMBASE, Medline (through the OVID interface), and PubMed. The search strategy used was: [“varicose veins” OR “chronic venous insufficiency” OR “leg ulcer” OR “post-thrombotic syndrome” OR “post thrombotic syndrome” OR “post-phlebitic syndrome” OR “post phlebitic syndrome”] AND [“genetics” OR “gene” OR “genes” OR “mutation” OR “polymorphism”]. We restricted the search to studies published in English. The retrieved references were initially screened for eligibility based on titles by 1 author and confirmed by another author. A preliminary list of potentially relevant studies was then independently screened by 2 authors based on titles and abstracts and a final list of studies to review in full was generated by consensus. Any studies whose relevance was unclear were reviewed in full.

Quality of the studies was not assessed because of the lack of validated scales for the type of studies included. We did not plan a meta-analysis because we anticipated substantial heterogeneity among studies. Our results are presented descriptively by CVI subtype. Finally, although no standards exist for reporting systematic reviews of basic science studies, whenever possible we attempted to adhere to the available reporting standards.^17,18 

The search process is summarized in Figure 1. The initial search generated 682 results. The initial screening by title alone resulted in 142 potentially relevant studies. The final selection for review in full included 41 studies, of which 3 were excluded and 38 articles were included in the final review.

Our literature search yielded various candidate genes that have evidence-based associations with a spectrum of CVD including CVI, varicose veins, LDS, venous ulceration, and PTS (Table 1). There was significant overlap in that these genes were often associated with more than 1 CVD manifestation (Figure 2), but they are presented here according to their most significant associations with a given subtype. The matrix metalloproteinases (MMPs) are discussed separately, given their consistent strong associations with various forms of CVD.

The ECM is a complex and dynamic framework of collagen, proteoglycans, elastin, glycoproteins, and cellular components. Degradation or destruction of the ECM disrupts the homeostasis of the vein (much of which is maintained by the MMPs, discussed later), thereby leading to varicosity. Vein wall weakness is a key player in the pathophysiology of varicose veins, and collagen is an important matrix component that provides strength; however, collagen dysregulation leads to vein wall abnormalities.⁴  In varicose veins, the total elastin content is decreased, type I collagen is upregulated, and type III collagen is downregulated.^7,19  Jin et al⁴  conducted a recent study that examined a 7-base pair insertion/deletion polymorphism (rs3917) in the COL1A2 (α-2 type I collagen) gene. They found that this polymorphism upregulated COL1A2 expression and produced a 1.6-fold increase in CVI risk and speculated that genetic variations in this gene alter transcriptional activity, affect messenger RNA (mRNA) structure, and ultimately allow expressional upregulation.⁴ 

Kowalewski et al²⁰  demonstrated increased expression of the cytokine vascular endothelial growth factor A (VEGF-A) and receptor VEGF R2 in the walls of varicose veins as compared with normal saphenous veins, particularly if complicated by thrombophlebitis. Similarly, Hollingsworth et al²¹  observed increased transcription of VEGF and its receptors in varicose veins, reflecting a potential early role in varicogenesis. Of note, VEGF-A is involved in the maintenance of vessel wall integrity. Increased expression of VEGF-A or its receptor R2 leads to increased activation of nitric oxide synthase, which then leads to vessel wall damage mediated by oxygen free radicals as well as decreased vessel tone that predisposes to venous stasis.²⁰  VEGF also plays a key role in angiogenesis, so SNPs in the VEGF gene (C936T and −1780 T/C have been described) can be considered risk factors for impaired wound healing and venous ulceration.²² 

Chang et al⁶  conducted a study using microarray bioinformatics that systematically explored various genes involved in biological pathways that may contribute to varicosity. The results showed that 32 genes were upregulated and 74 genes were downregulated in varicose veins, most of which were related to apoptosis and angiogenesis. Important examples of upregulated genes included: HSP90, ILK, and TGF-β1.⁶  In fact, Saito et al²³  earlier noted that TGF-β1 stimulates collagen synthesis and alters levels of MMPs (discussed later).

Mutations in the FOXC2 gene have also been associated with varicose veins. FOXC2 encodes a transcription factor involved in lymphatic and vascular development, and mutations in FOXC2 are seen in lymphoedema distichiasis, which is characterized by lymphoedema and varicose veins.^7,24  Ng et al²⁴  conducted an early twin linkage study that strongly implicated FOXC2 in the development of varicose veins as a heritable condition. Similarly, knowing that FOXC2 had previously been implicated in primary venous valve failure, Al-Batayneh et al²⁵  identified 3 specific SNPs in the FOXC2 gene that may contribute to varicose vein and hemorrhoid development. More recently, Mellor et al²⁶  conducted a small study of 18 patients and noted a strong association between FOXC2 and primary venous valve failure in both superficial and deep veins of the lower limb.

Other, lesser-known genes have also been examined in preliminary studies and must be mentioned. Jeong et al²⁷  performed large scale mRNA screens among normal varicose veins and found the greatest differential expression in the octamer-binding transcription factor-1 gene (Oct-1), which was upregulated in primary varicose veins. Cario-Toumaniantz et al⁸  found overexpressed vitamin K–dependent matrix gla protein (MGP) in varicose veins, and speculated that its role in wall remodeling involved smooth muscle proliferation and mineralization processes.

Grzela et al²²  reviewed 4 genes that may play a role in venous ulceration: tumor necrosis factor (TNF), fibroblast growth factor-R (FGF-R), estrogen receptor, and HFE. Levels of TNF-α and interleukin-1 (IL-1) are higher in venous leg ulcers compared with normal acute wounds, with certain SNPs (such as the −308A variant in TNF-α) conferring a higher risk of venous ulceration compared with wild-type.^5,22,28  Along with demonstrating the previously mentioned association with TNF-α, Wallace et al²⁸  also identified a polymorphism in intron 10 of the BAT1 gene (human leukocyte antigen B–associated transcript-1) as being a significant risk factor for venous ulceration.

Similarly, FGFs and their receptors (FGF-R) are cytokines that are imperative for the control of connective tissue regeneration in wound healing. Kowalewski et al⁹  found increased acidic fibroblast growth factor (a-FGF) expression in the walls of varicose veins. They noted that these walls contained extensive ECM remodeling and altered collagen and glycosaminoglycan differentiation; a-FGF likely influences the expression of certain enzymes through various pathways (eg, mitogen-activated protein kinase pathway) involved in ECM metabolism and varicosity. Furthermore, a-FGF synthesis is enhanced by hypoxia, which may be a consequence of venous stasis.⁹  Previous studies indicate that certain SNPs in FGF-R2, most frequently the polymorphism A2451G, are more commonly observed in CVI or ulceration. Though the mechanism is unclear, these SNPs likely result in lower expression of FGF-R2, causing impaired regeneration of connective tissue, longer vein wall reepithelialization, and eventually CVI or ulceration.²²  Nagy et al²⁹  conducted an association study and identified an SNP in the FGF-R2 gene that was more prevalent CVI patients with nonhealing ulcers; as mentioned previously, this likely led to abnormal reepithelialization and angiogenesis.

Estrogen is a known contributor to ECM metabolism; in fact, hormone replacement therapy prevents CVI and topical estrogen promotes wound healing in the elderly by decreasing the inflammatory response.³⁰  Although convincing associations have not been elucidated, several SNPs in the ER-β receptor have been associated with venous ulcers in the elderly.^22,30  Ashworth et al³⁰  found that polymorphisms in the upstream regulatory regions of the ER-β gene were significantly associated with venous ulceration, but did not conduct further functional studies to determine the precise mechanism.

Hemochromatosis is an inherited iron-overload disease caused by mutations in the HFE gene (most commonly allele C282Y), a major histocompatibility complex class 1-type membrane protein associated with β2-microglobulin. The mutated gene results in abnormalities in the regulation of iron absorption related to interactions between transferrin and its receptor. In patients with CVI, the HFE C282Y allele increases the risk of ulceration by almost 7-fold, likely because of accumulation of iron in tissues surrounding blood vessels, increased free radicals and oxidative stress, and upregulation of the inflammatory response that finally leads to tissue destruction.²²  Moreover, Gemmati et al³¹  found that the −8CG polymorphism in the FPN1 gene (ferroportin; involved in exporting iron out of the cell) increases susceptibility to leg ulcers.

Finally, Sam et al³²  suggested from preliminary data that mild to moderate hyperhomocysteinemia is common in patients with CVI (approximately 65%, especially with ulceration), and is associated in one-third of patients with an underlying methylenetetrahydrofolate reductase C677T homozygous polymorphism.

LDS is a consequence of CVI and is characterized by severe skin changes including hardening, atrophy, dark pigmentation, and edema. It often progresses to skin breakdown, venous ulceration, and delayed healing.^33,34  deGiorgio-Miller et al³³  analyzed leg skin biopsies from patients with LDS and found enhanced cell proliferation and procollagen gene expression as well as significant fibrotic changes which correlated directly with ulcer formation and healing time. Furthermore, imbalances in the MMPs and their inhibitors (discussed later) have also been implicated in LDS through the generation of epidermal and dermal skin defects.³⁴ 

PTS is a frequent yet poorly understood complication of DVT. Previous studies have explored genes involved in vein wall remodeling in relation to CVI and its manifestations (eg, varicose veins, venous ulcers). These are also manifestations of PTS, and it is logical to infer that polymorphisms in these very genes may potentially increase the risk of PTS as well. In fact, Deatrick et al³⁵  noted ongoing vein wall remodeling 6 months after an acute DVT that was associated with biomarkers such as MMP-9, which directly correlate with resolution and predict PTS. There is, however, a dearth of literature that directly associates the above genes with PTS.

An early study by Dahi et al³⁶  demonstrated increased MMP-2 and MT1-MMP activity (potentially mediated by thrombin) during DVT resolution, which, in turn, increased the risk of PTS. Wojcik et al³⁷  used a mouse model to conclude that IL-6 is associated with reduced monocyte recruitment, leading to reduced vein wall thickness and fibrosis. Given that PTS involves extensive perivenous and mural fibrosis, IL-6 may serve as a therapeutic target to prevent these fibrotic complications. Another recent mouse model study by Deatrick et al³⁸  reveals that MMP-9 modulates vein wall collagen content and contributes to inflammation and fibrosis, thus implicating it as a potential target to reduce the fibrotic complications of PTS.

To our knowledge, there has been no study to date that has analyzed polymorphisms in the previously mentioned candidate genes to uncover a direct association with PTS. Some of the studies presented in Table 1 involve tissue blocks from veins complicated by thrombophlebitis, and some of the patients were noted to have a history of venous thrombosis (however, scant patient data were provided), but none of the studies specifically included PTS as a subset.

Perhaps the most studied genes in venous disease are the MMPs and tissue inhibitors of metalloproteinases (TIMPs). A systematic review by Lim et al⁷  revealed that MMP-1, MMP-2, MMP-3, MMP-7, MMP-9, TIMP-1, and TIMP-3 were all upregulated in varicose veins. Similarly, Raffetto et al³⁹  noted that MMPs are involved in ECM degradation, which subsequently leads to venous remodeling, structural wall changes, and ultimately venous dilation and valve dysfunction; they are thus heavily involved in the pathogenesis of CVD. Though the molecular mechanisms of these genetic anomalies have not been fully elucidated, it appears that MMPs regulate or degrade the ECM through hydrolysis, whereas TIMPs are tissue inhibitors of MMPs that influence vascular remodeling. Hence, an imbalance between these proteins may lead to abnormalities in the vessel wall and, ultimately, vascular disease such as varicosity and ulceration.^7,35,40  Moreover, misregulation of MMP activity and TIMP counterregulation contributes to impaired ulcer healing.³⁴  An early study by Saito et al²³  suggested that increased MMP-2 levels affect tissue remodeling and contribute to a proulcer environment. Subsequently, Herouy et al⁴¹  examined stasis dermatitis, a common consequence of impaired venous drainage characterized by dermal neovascularization, and noted elevated MMP-1, −2, and −13 and diminished TIMP-1 and −2 in the skin lesions. More recently, Xu et al⁴⁰  showed that specific polymorphisms in the MMP-9 and TIMP-2 genes potentially put patients at a higher risk for developing varicose veins. Deatrick et al³⁵  examined vein remodeling and associated gene expressions, but did not correlate results with the development and severity of PTS. However, their preliminary data did show increased MMP-9 and decreased Toll-like receptor 9 expression in acute DVT. MMP-9 levels have been shown to be increased in varicose veins and venous ulcers, and some data suggest a potential role in DVT resolution, with gene deletions being associated with less collagen deposition. In fact, Beidler et al⁴²  used multiplexed protein analysis to show that all MMPs except MMP-7 were highly expressed in venous leg ulcers, especially MMP-8 and MMP-9. In addition, Singh et al⁴³  analyzed various SNPs associated with venous leg ulcers and found that MMP-12 gene polymorphisms may have a role in ulcer progression.

Interestingly, Kurzawski et al⁴⁴  demonstrated that polymorphisms in MMP-1 and MMP-3 did not predict susceptibility to varicose veins, but this study likely did not have the power to uncover the milder effects of these genes on an otherwise multifactorial disorder.

In this review, we examined different candidate genes and their polymorphisms (mostly SNPs) that have been linked to various forms of CVD.

In assessing these studies, a few candidate genes should be highlighted as having the strongest link to the development of CVD (because we could not perform a meta-analysis, these assessments are qualitative and based on the available evidence). Three studies agree that SNPs in FOXC2, a transcription factor involved in lymphatic and vascular development, provide a strong link toward venous varicosity, valve failure, and hemorrhoids. On the other hand, a polymorphism in the HFE gene increases the risk of venous ulceration by almost 7-fold, while also causing the known dysregulation in iron absorption. The MMPs (and, to a lesser extent, their inhibitors, TIMPs) are perhaps the best studied in venous disease. MMPs are involved in ECM degradation and structural vein wall changes, ultimately contributing to venous remodeling, dilation, and valve dysfunction. They are thus heavily involved in the pathogenesis of CVD. The exact mechanisms have yet to be elucidated and further research is required in this area.

Although genetic markers in CVD have been examined in some detail, there is a lack of evidence correlating these genes with the development and severity of PTS. If these polymorphisms are potentially associated with CVI, LDS, varicose veins, venous ulcers, and DVT resolution, it is plausible that they could also provide an underlying substrate for the development of PTS. The MMPs have once again been implicated in this respect, but data are scarce and further research is warranted.

In general, a limitation of all of the studies included in this review relates to the fact that they were conducted either in animal models, in very limited numbers of patients, or using surgical samples (eg, saphenectomy). It is difficult to assess the quality of each study because no standardized tools exist in this regard—in contrast to the quality assessment scales used for clinical studies. We believe that the included studies were methodologically sound; however, we cannot totally rule out the possibility of biased results.

Clearly, the genetic predisposition to CVD, particularly PTS, is controversial and unclear at best, and more studies are needed to clarify these associations. Earlier studies have promoted the use of a genome-wide approach; this allows for the identification of previously unknown markers, and may be a consideration for the future. Clinically, this would be a novel way to reexamine a patient’s propensity toward developing CVD, predict those who go on to develop it, and provide a better understanding of the underlying mechanisms to ultimately improve treatment options.

Herein, we examine the role of various candidate genes and their polymorphisms in the development of CVD, including the lesser studied PTS. Our observations support a genetic predisposition to CVD related to vein wall remodeling. Genes of significance include FOXC2, HFE, and the MMPs, all of which show strong associations with varicose veins, CVI, or venous ulceration. The data surrounding PTS are more limited, but given that it is a type of CVD with similar manifestations, we may infer that some of the above genes may be implicated. Thus, further studies are needed to examine these associations more directly. Most importantly, given the burden of this disease worldwide and the paucity of treatment options currently available, studying these polymorphisms could potentially allow us to better identify patients at higher risk of developing CVD, and also provide novel therapeutic targets.

The researching, writing, and editing were performed solely by the authors and no other individuals or organizations were involved.

Contribution: A.L.L. and S.K. conceived the idea; V.B. and A.L.L. performed the systematic review; V.B. wrote an initial draft of the article; V.B., S.K., and A.L.L. reviewed and edited the manuscript; and all authors reviewed and approved the final manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Alejandro Lazo-Langner, Hematology Division, London Health Sciences Centre, 800 Commissioners Rd E, Rm E6-216A, London, ON, N6A 5W9, Canada; e-mail: alejandro.lazolangner@lhsc.on.ca.