We sought to determine if oxLDL levels were heightened in patients with DM-mediated DPN and related to the presence of neuropathic pain in DPN-P patients. Our results indicate that we were unable to determine any relationship between plasma oxLDL levels and the presence of DPN or the severity of DPN. There was also no significant increase in plasma oxLDL levels in the presence of DM as compared to a non-diabetic control state. Furthermore, there was no detected worsening in clinical severity of neuropathy with higher levels of oxLDL, HbA1C, total cholesterol, triglycerides, or LDL levels. Recently reported studies have also demonstrated a lack of elevated oxLDL levels in patients with DPN or DM without DPN when compared to control subjects . Although prior studies indicate a positive relationship between any diabetic microvascular complication and abnormalities in LDL, triglycerides, and non-HDL cholesterol  with similar overall cholesterol results, it remains unclear regarding any association between lipid status and presence of peripheral neuropathy. Thus, although preclinical studies have identified potential mechanisms by which oxLDL may contribute to neurodegeneration, particularly in the presence of hyperglycemia related to DM, our clinical study did not identify any particular relationships.
The present findings should not be seen to discourage against further investigations of other methods by which hyperlipidemia could contribute to neurodegeneration in DM patient populations. There is clinical evidence that triglyceridemia contributes to progression of already existing DPN , while each of total cholesterol, low-density lipoprotein and triglyceride levels have been positively associated with increasing cumulative incidences  or greater severity  of diabetic peripheral neuropathy. Furthermore, a low high density lipoprotein level may also relate to presence of DPN . The mechanisms by which these forms of cholesterol may impact upon peripheral nerve function and structure are unclear. It is also difficult to simulate these human relationships in animal models; for example, mice have quite low levels of low density lipoproteins, while most cholesterol is transported by high density lipoproteins , distinct from the human situation. Despite this issue, wildtype mice fed high fat diets develop elevated plasma oxLDL levels . Peripheral nerve studies in transgenic mice featuring highly increased plasma cholesterol are untenable due to excessive short term mortality . In wildtype mice, a high fat diet plays an important role in the severity of experimental DM [26, 27], and it is even possible that a high fat diet may lead to peripheral neuropathy in the absence of DM . Thus, there is sufficient preclinical and clinical data to suggest that an association exists between lipid disorders and peripheral neuropathy, but the potential mechanism has remained elusive.
Further complicating our ability to understand relationships between lipid disorders and peripheral neuropathy is the controversial relationship of frequently prescribed statin medications used for lipid disorder management. Statins lower cholesterol through inhibition of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Statin use is extremely common [29, 30], particularly in diabetic subjects; in our study, over 50% of DM subjects were taking statins. There are a number of case series and case-control studies [6, 31–33] along with a recent cross-sectional study  suggesting a relationship between statin use and peripheral neuropathy presence. However, there are also some contrasting reports, suggesting that statins may have a beneficial effect upon peripheral neuropathy in humans (limited to hypotheses) [34, 35] and rodents [36–38]. Nevertheless, given the absence of a plausible mechanism by which statins could induce peripheral neuropathy, the effect of statins upon peripheral nerves remains unclear. In our study, the use of statins was significantly more common in the diabetic cohorts, which likely contributed to relatively depressed LDL levels relative to Control Subjects and Idiopathic Neuropathy Subjects. This is a confounder that certainly impacts upon our results as statins certainly will decrease oxLDL levels [39, 40] in our diabetic cohorts. This obstacle is difficult to overcome in clinical studies, as the great majority of diabetic patients now receive statin therapies. The washout of statin medications would likely be required in future studies of experienced DM patients to determine natural oxLDL levels; alternatively, investigation of newly diagnosed DM patients prior to interventions may assist in future determinations.
We chose to examine oxLDL due to its potential for neurotoxicity. Known to have a critical function in the development of atherosclerosis [7, 41, 42] along with its receptor, the lectin-like oxLDL receptor (LOX)-1 , its impact upon non-neural tissues has great substantiation. The receptor LOX-1 has non-constitutive but dynamically inducible expression  driven by both oxLDLs and hyperglycemia [11, 44]. Along with oxLDL, LOX-1 contributes to intracellular oxidative stress and inflammation injury [9, 10]. In vitro, oxLDLs contribute to oxidative stress and injury in DRG neurons via a LOX-1-mediated mechanism related to activation of NAD(P)H oxidase . These data not only suggest that oxLDL/LOX-1 may contribute to DPN, but that they may also lead to development of peripheral neuropathy even in the absence of DM. Dietary alterations may also play roles: high-fat feeding in murine models significantly increases plasma oxLDL as well as contributing to morphological and functional evidence of peripheral neuropathy prior to development of hyperglycemia . Therefore, although our study failed to identify higher levels of oxLDL in human subjects with DPN (or even only with DM), there is evidence to suggest that elevated oxLDL levels can be neurotoxic.
Our findings are presented with limitations to be considered. Although we used a standard ELISA method to measure oxLDL levels, there are different methods that may be favored in some circumstancses . Also, other methods of analyses including in vivo measurement  and ex vivo oxidation assays  exist. In some cases, serum oxLDL levels [48, 49] are measured instead of plasma oxLDL levels [42, 50]–it remains unclear which technique may be most accurate. Studied subjects were volunteers and were not randomly selected from a population with type 2 DM with or without DPN and with or without accompanying NeP. Our sample size was not based upon a pre-determined power analysis. Patients with type 1 DM were excluded due to potential differences in the pathophysiology of neurodegeneration and to maintain a more homogeneic patient population. As stated, gender matching between groups was not complete as female subjects were overrepresented in the control group. Also, statin use was much more common in DM patients, although its use was not associated with significant changes in measured oxLDL levels. The use of pharmacotherapies for management of DM and chronic pain related to DPN may have impacted upon oxLDL measurements in uncertain ways. Although total cholesterol and low density lipoprotein levels were positively associated, there was no association found between lipid measurements and plasma oxLDL levels. Clinical severity of neuropathy failed to correlate with types of lipids measured or with HbA1C levels. Although higher HbA1C levels are associated with risk of subclinical neuropathy  and electrophysiological parameters , HbA1C is not necessarily an independent predictor of diabetic neuropathy although it is clearly associated with other microvascular complications [53, 54]. However, in prior studies examining DPN populations, the clinical severity of DPN based upon the UENS was positively associated with age, duration of diabetes, height, weight and HbA1C . Our study population was small, but the lipid status and HbA1C values are similar to similar populations of DPN patients studied . We did not perform measurement of other markers for oxidative stress such as glutathione, malondialdehyde, or homocysteine, which may be elevated in DPN patients  and could have provided comparison to measured oxLDL and HbA1C levels. Finally, the individual group sizes were small after categorization, and we cannot rule out the possibility of type II errors contributing to the lack of statistical associations determined–future larger studies may be more forthright. However, there was a distinct lack of relationship between oxLDL levels and DPN manifestations found amongst studies performed to date.
Our results demonstrated no visible associations between oxLDL levels in DM subjects and the presence or severity of DPN. The results of our study do not exclude the possibility of elevated levels of oxLDL at the level of the peripheral nerve itself contributing to DPN [13, 57]; we did also not study the role of the LOX-1 receptor in our population. As a result, further more invasive studies may determine other potential findings related to oxLDL. Future studies may examine other aspects of lipid disorders for relationship with complications of DM, including DPN. Although our hypothesis of plasma oxLDL mediating greater presence and severity of DPN in subjects with DM was not verified, it is possible that our results may be confounded by some of the factors discussed above. Factors other than plasma oxLDL levels should be sought to determine causative or contributory factors in the development of DPN.