Dyslipidemia is a risk factor for atherosclerosis and PAD. This chapter will address (1) the significance of dyslipidemia in the pathogenesis of PAD, (2) the evidence for dyslipidemia as a risk factor for existence and progression of PAD, and (3) the evidence regarding lipid-modifying therapy in PAD.

Dyslipidemia is a key pathogenic factor predisposing to atherosclerosis. Lipoproteins are instrumental in the initiation of atherosclerotic plaques and their progression to hemodynamically significant lesions that cause arterial insufficiency.

Low-density lipoprotein cholesterol (LDL) is a key driver of the atherosclerotic process. Fatty streaks, the earliest visible sign of atherosclerosis, consist mainly of macrophage-containing foam cells that are full of oxidized LDL.¹ Multiple factors including increased endothelial permeability, retention of lipoproteins in the intima, and sluggish removal of lipoproteins result in accumulation of LDL in the vessel wall.^2,3,4 Oxidized LDL and its products cause a vigorous inflammatory and proatherogenic response through chemotactic signaling for monocytes, smooth muscle cells, and T lymphocytes.^5,6 Oxidized LDL leads to increased expression of a host of inflammatory factors including vascular cell adhesion molecule-1, monocyte chemoattractant protein-1, intracellular adhesion molecule-1, and macrophage colony stimulating factor.^{7,8,9,10,11,12,13,14,15} The net effect of this is to attract, trap, and tether leukocytes to the endothelium, initiate transformation of monocytes into macrophage foam cells, and enhance smooth muscle cell proliferation in the intima to form an atheroma. Oxidized modification of LDL may also contribute to so-called plaque vulnerability through induction of type 1 metalloproteinase expression and an increase in tissue factor activity.⁵ The oxidation of LDL, then, is an initial insult that leads to a cascade of immunologic and vascular events that cause endothelial dysfunction as well as initiation and progression of atherosclerotic plaques (Figure 3-1).

Further supporting the pivotal role of LDL in atherosclerosis, atherosclerotic lesion formation in mice has been inhibited through immunization with products of oxidized LDL.^16,17,18,19 Not surprisingly, in numerous human trials, lowering LDL through pharmacological means has been shown to reduce progression of atherosclerosis as well as the incidence of cardiovascular events.^{20,21,22,23,24,25,26,27,28}

Low levels of high-density lipoprotein cholesterol (HDL) also strongly predict cardiovascular events. Low levels of HDL are associated with insufficient reverse cholesterol transport.^29,30 When human ApoA-I, the major protein in HDL, was increased in mice by using liver-directed gene transfer, reverse cholesterol transport was promoted and regression of atherosclerotic lesions was seen.^31,32 Additionally, HDL has antioxidant properties at least partially mediated by paraoxonase—an enzyme associated with HDL that degrades oganophsosphates.^33,34 This presumably allows for protection against oxidation of LDL.

While the above information provides a framework for understanding the mechanisms of atherosclerosis in the vasculature in general, facets specific to the peripheral arteries have not been well studied. Unlike in the coronary arteries, in situ thrombosis in the peripheral arteries resulting in acute limb ischemia is a rare event. This maybe due to differences in the mechanical forces of blood flow through the peripheral arteries when compared to their coronary counterparts, vascular cell heterogeneity, and variation in plaque composition in distinct vascular beds.^{35,36,37,38,39,40,41,42} Additionally, recent studies support different pathogenic mechanisms and risk factors for large-vessel, or proximal, PAD versus small-vessel, or distal, PAD.^43,44 However, the latter has not been well studied to this point. For this reason, PAD atherosclerosis is currently hypothesized to be progressive stenosis of the larger peripheral arteries. This has been measured in various ways including symptom indices, Doppler wave forms, and angiography. The current accepted standard test for diagnosis of PAD is the easy, inexpensive ankle-brachial index (ABI). An ABI < 0.9 denotes PAD, and the ABI is a sensitive and specific marker for diagnosis and progression of the disease.⁴⁵

It is important to note that vascular calcification is a common component of peripheral atherosclerotic plaques. The calcification of atherosclerotic plaque is localized to the intimal layer. However, calcification may also occur, especially in patients with diabetes mellitus, in the medial layer causing supranormal ABI. The pathophysiology of intimal vascular calcification is an active process involving extracellular bone matrix proteins in the setting of inflammatory cells.⁴⁶ A high calcification content relative to lipid content is indicative of a more stable plaque.⁴⁷ More research is needed to determine if plaque rupture or hemorrhage is a significant part of the progression of atherosclerotic lesions in the lower extremities.

There is an abundance of data identifying elevated LDL levels and decreased HDL levels as independent risk factors for coronary atherosclerosis.^{20,21,22,23,24,25,26,27,28,29,30} Hypertriglyceridemia has also been shown to confer a more modest amount of increased cardiovascular risk in large epidemiological studies.⁴⁸ However, in contrast to studies targeted to reducing LDL, substantial “proof-of-concept” data do not exist regarding treatment of hypertriglyceridemia as a means for decreasing cardiovascular events.

Early studies of patients with PAD revealed positive associations with triglycerides, mixed associations with LDL, and negative associations with HDL.^{49,50,51,52,53,54,55,56,57} Unfortunately, these studies were limited by small sample sizes and inconsistency in the criteria used to define PAD. Large epidemiologic studies using more rigorous statistical methods, including multivariate regression, have helped better define the role of lipid moieties as independent risk factors for the presence of PAD.

An initial analysis of data from the Framingham cohort revealed that there was a 20% increase in risk of intermittent claudication seen for every 40-point increase in total cholesterol.⁵⁸ This study was limited by a lack of data on fractionated cholesterol and an insensitive measure of PAD, intermittent claudication. An analysis of data from the Systolic Hypertension in the Elderly Trial revealed univariate positive correlations between LDL, total cholesterol, and triglycerides with PAD as well as a negative correlation with HDL level and PAD.⁵⁹ Later studies continued to use the more sensitive measure of ABI <0.9 to define PAD and also used multivariate statistical models to more specifically identify independent risk factors for PAD. For example, a subsequent more detailed analysis of the Framingham Offspring cohort showed a 10% increase in PAD for every five-point decrease in HDL by using multivariate regression.⁶⁰ LDL, total cholesterol levels, and triglycerides were not significantly associated with the presence of PAD in this model. In a large epidemiological study of Asian men older than 70 years, multivariate regression revealed an odds ratio for PAD of 1.36 when comparing patients in the highest quintile of total cholesterol to those in the lowest quintile of total cholesterol.⁶¹ With this method, HDL showed a significant inverse association with PAD with an odds ratio of 0.68 in this cohort. LDL and triglycerides were not specifically examined. In the Strong Heart Study examining a diverse population of American Indians, LDL was the only lipid moiety shown to be significantly associated with PAD.⁶² A 19% increase in the risk of PAD was seen for every 30-point increase in LDL in this group. In data analyzed from the Cardiovascular Health Study, a population of more than 5000 patients older than 65 years, it was found that HDL decreased with decreasing ABI in all patients.⁶³ In this cohort, LDL and total cholesterol were significantly associated with PAD in women, but not in their male counterparts. Triglycerides were not shown to be significantly associated with ABI in multivariate analyses. In the Edinburgh Artery Study, multiple regressions of risk factors for PAD revealed inverse associations with HDL, but only univariate associations with triglycerides.⁶⁴ Finally, data from a large cohort of physicians showed total cholesterol/HDL ratio to be a stronger predictor of PAD than any individual lipid or inflammatory marker.⁶⁵

In summary, existing data seem to most strongly support decreased HDL levels as an independent risk factor for PAD. Multiple studies also reveal an association between total cholesterol levels and PAD. The data on LDL as an independent PAD risk factor are mixed in large epidemiological studies. However, patients with Fredrickson class IIa dyslipidemia (familial hypercholesterolemia) with an LDL receptor deficiency have a dramatically increased incidence of PAD.⁶⁶ Also, patients with Fredrickson class III dyslipidemia, a genetic defect in apoE synthesis leading to increased LDL, have an increased risk of PAD.⁶⁷ Lastly, while multiple studies have shown univariate associations between increased triglyceride levels and the presence of PAD, these associations largely disappear with multivariate analysis decreasing the strength of the evidence for hypertriglyceridemia as an independent risk factor for PAD.

Most studies examining risk factors for the progression of PAD have not focused on lipids and instead examined more novel inflammatory markers. A study of 381 patients revealed baseline hypertriglyceridemia to be associated with the progression of PAD for more than 3-year follow-up.⁶⁸ Another study examining a group of elderly patients showed LDL levels greater than 147 mg/dL to be associated with declining ABI.⁶⁹ In this analysis, association with triglycerides and HDL was not significant. Overall, studies directly examining lipid levels and progression of PAD are scant. However, there have been a number of studies examining the effects of lipid-modifying therapy on functional outcomes in patients with PAD; these are detailed next.

Lipoprotein(a), or Lp(a), is a modified form of LDL that has been implicated as an independent risk factor for coronary heart disease in patients with hypercholesterolemia.⁷⁰ There is also evidence that Lp(a) is associated with acute coronary syndromes and strokes.^71,72,73

A number of studies have attempted to establish whether Lp(a) levels are significantly associated with PAD. In a small study comparing 17 patients younger than 45 years with PAD to a control group without PAD, the patients with PAD were 3.9 times more likely to have a high Lp(a) level (greater than 30 mg/dL) than those without PAD.⁷⁴ In a post-hoc analysis of the Systolic Hypertension in Elderly Trial, 36% of patients with Lp(a) > 20 mg/dL had an ABI < 0.9 versus only 14% with Lp(a) < 20 mg/dL.⁷⁵ In this analysis, levels of Lp(a) were also correlated with disease severity as measured by worsening ABI. A study of Chinese patients with diabetes confirmed these results and noted that diabetic patients with an Lp(a) level > 13.3 mg/dL had a 2.7-fold increased risk of PAD.⁷⁶ A study examining novel risk factors for PAD did not find an association between PAD and Lp(a) levels.⁶⁵ This study, however, used self-reported claudication as the measure of PAD. This is now known to be a quite insensitive marker for the disease. Finally, a recent trial has shown that baseline Lp(a) levels are correlated with progression of PAD.⁴³ Overall, existing data have established an inverse correlation between ABI and Lp(a) levels. Lp(a) may also be a risk factor for PAD progression.