In the present study, we found that patients in the highest quartile of serum albumin had lower prevalence of DPN compared with subjects in the lowest quartile (P <0.01). There was an 50.1% decreased risk of DPN (95% confidence interval [CI] 0.404-0.544; P <0.01) per 1 SD increase of serum and 62.7% decreased risk of DPN in quartile 4 of serum albumin versus quartiles 1, 2, and 3 (95% CI 0.195-0.714; P=0.003) after multivariate adjustment. Serum albumin could predict DPN with 65.88% sensitivity and 66.7% specificity for the best cutoff value of 39.95 g/L. These findings suggest that lower serum albumin level may be associated with an increased risk of DPN.
Albumin, a multifunctional plasma protein, possesses an important antioxidant property by binding metal ions, and efficiently scavenging free oxygen and nitrogen radicals [3]. It has been reported that serum albumin has three to seven times the total antioxidant capacity compared to vitamin E, vitamin C, and bilirubin as a chain-breaking antioxidant in vitro [18]. Changes of plasma concentration and structural modification of albumin induced by high glucose or free radicals impair its antioxidant properties and aggravate oxidative stress [19]. Several previous studies have explored the association of serum albumin with the development of DPN, and provided inconsistent results [12-15]. Iwasaki et al. reported that serum albumin was independently related to the median motor nerve conduction velocity and minimum F-wave latency in 130 Japanese patients with T2DM, and was significantly inversely associated with the presence of DPN [12]. Data from a cross-sectional study of 409 Chinese patients with T2DM underwent measurement of nerve conduction (NC) showed that serum albumin was independently associated with peripheral nerve function as reflected by composite Z scores of all NC parametersin T2DM patients, especially in those with albuminuria [13]. Recently reported data from the Saudi study, which included 2,906 hospitalized adult diabetics, demonstrated that DPN cases had a significantly lower serum albumin levels compared with those without DPN [14]. However, no statistically significant association between serum albumin and DPN in US patients undergoing foot and ankle surgery with a history of Type 1 and 2 diabetes mellitus over a 13 month period [15]. In the present study, we observed that patients in the highest quartile of serum albumin had lower VPT, an useful and reliable method for early screening DPN and reflecting the clinical severity of DPN [16], and prevalence of DPN compared with subjects in the lowest quartile, and serum albumin was negatively associated with VPT and prevalence of DPN after adjustments for age, gender, BMI, and diabetic duration, demonstrating the potential neuroprotective actions of serum albumin in the pathogenesis of DPN. Moreover, there was an 50.1% decreased risk of DPN per 1 SD increase of serum and 62.7% decreased risk of DPN in quartile 4 of serum albumin versus quartiles 1, 2, and 3 after multivariate adjustment. Further analysis of ROC curves revealed that the serum albumin concentration of 39.95 g/L could predict DPN. Collectively, these data demonstrate that decreased serum albumin may play an important role in the development of DPN in a Chinese population with T2DM, which needed to be verified by further studies.
Numerous experimental and clinical studies have indicated that low-grade inflammation and oxidative stress induced by chronic hyperglycemia play a major role in the onset and development of DPN [13, 20]. Our study findings provided further evidence that patients with DPN had higher FBG, HbA1c, CysC, NLR, and lower GGT than those without DPN, mostly consistent with the findings of previous studies [20, 21, 22], supporting a role of inflammation and oxidative stress in the development of DPN. HbA1c represents blood glucose control conditions for recent 8-12 weeks [20]. UA, the end product of purine catabolism, has recently been regarded as an important endogenous antioxidant and anti-inflammatory [23]. GGT, the principal enzyme responsible for extracellular catabolism of the antioxidant glutathione, has been reported to be involved in oxidative stress and lipid peroxidation, and elevated under oxidative stress conditions [24]. CysC, a 13-kDa endogenous cysteine proteinase inhibitor, has been found to be positively associated with proinflammatory cytokines c-reactive protein (CRP), interleukin (IL)-6, and tumor necrosis factor-α (TNF-α) and marker of oxidative stress homocysteine levels, and negatively with antioxidant paraoxonase 1 activity across a variety of populations [25, 26]. NLR, defined as the neutrophil count divided by lymphocyte count, has recently been increasingly recognized as a marker of systemic inflammation [21]. Our data suggest that the T2DM subjects in the higher serum albumin quartiles exhibited higher levels of UA, and lower FBG, HbA1c, GGT, NLR, and CysC compared with subjects in the lowest quartile, and serum albumin levels were positively associated with UA, and negatively with HbA1c, GGT, NLR and CysC after adjustments for age, gender, BMI, and diabetic duration, indicating that serum albumin as a negative acute phase protein may have anti-inflammatory and antioxidant activity. These findings are consistent with the published literature [27-29]. Yeh et al. found that serum albumin level was inversely correlated with CRP, white blood cell, and NLR., and change in serum albumin level was inversely correlated with change in NLR in adult surgical intensive care unit patients receiving enteral nutrition [27]. Cantin and co-workers demonstrated that albumin augments intracellular glutathione (GSH) levels in lymphocytes and modulates activation of nuclear factor-κB (NF-κB), whereas a thiol-mediated reduction in serum albumin decreases glutathione levels, thus allowing NF-κB activation, and stimulate the inflammatory response [28]. Zhang and collaborators demonstrated that physiological concentrations of albumin selectively inhibit TNF-α-induced upregulation of vascular cell adhesion molecule-1, expression and monocyte adhesion in human aortic endothelial cells, most likely by inhibiting NF-kB activation in a GSH- independent manner [29]. These findings, together with data from our study, suggest that albumin could modulate cellular GSH as well as transcription of inflammatory genes and apoptosis, and the inflammatory state and oxidative stress associated with low serum albumin may be the main mechanism of DPN in these individuals, although further studies are needed to clarify the underlying mechanism.
There is growing evidence that diabetic vasculopathy caused by low-grade inflammation and oxidative stress, which could affect the nutrition supply of neuronal and Schwann cells and contribute to adequate blood supply and nerve degeneration, is believed to play an important role in the pathogenesis of DPN [17, 21]. Our study provided further evidence that supported the potential role of diabetic vasculopathy in the development of DPN, since we found that patients with DPN had higher CysC, Cr, urinary ACR, and lower DBP, eGFR, ABI than those without DPN. Diabetic nephropathy (DN), a complication characterized by a decreased eGFR, increased urinary ACR, and high levels of Cr [30], is one of the most feared diabetic chronic microvascular complications, and is independently associated with DPN [31]. The urinary ACR, a marker of DN, is used as a proxy for damage to the systemic microcirculation and predicts future progression of renal dysfunction and various cardiovascular events in several populations [32]. CysC, a new and more reliable surrogate marker than Cr and Scr-based eGFR for early detection of renal impairment in patients with diabetes, has been reported to be was significantly associated with DPN [22, 33]. ABI, a useful marker for detection of peripheral artery disease (PAD) or atherosclerosis at other vascular sites [34], has been demonstrated to be independently associated with DPN [35]. In the present study, we have revealed that the T2DM subjects in the higher serum albumin quartiles exhibited higher levels of DBP, eGFR, ABI, and lower Cr, CysC, and ACR compared with subjects in the lowest quartile, and serum albumin levels were positively associated with DBP, and negatively with Cr, CysC, and ACR after the adjustments, indicating that serum albumin may be associated with DN and PAD as diabetic micro- and macrovascular complications, respectively, which are consistent with those of many previous reports [15, 36-38]. Zhang et al. conducted a study of 188 patients with T2DM and biopsy-proven DN followed up for at least one year, and found that serum albumin level was significantly associated with proteinuria, renal function, and glomerular lesions, and hypoalbuminemia was associated with a poorer renal prognosis [36]. Another cross-sectional study of Japanese reported that patients with DN had a decrease in serum albumin concentration compared with those with non-DN [37]. Recently, Greenhagen and collaborators showed that US subjects with PAD had lower levels of serum albumin [15]. More recently reported data from the cross-sectional study of 10,900 Chinese hypertensive patients aged ≥18 years demonstrated that serum albumin levels were significantly inversely associated with the prevalence of PAD in men [38]. Together, our results and those findings of above previous studies may provide evidence that a negative association between serum albumin and DPN appears to be due to the systemic circulation damage, mainly caused by DN and PAD and accordingly adequate nutrition and blood supply of neurons and nerves in these individuals. However, the mechanism may warrant further investigation in larger databases and interventional studies.
Our study has several limitations that must be acknowledged. First, the cross-sectional design of our study likely does not reflect cause and effect. Thus, prospective studies are needed to confirm our findings. Second, we only evaluated UA, GGT, and CysC as parameters of oxidative stress, and NLR as marker of inflammation. Data about TNF-α, IL-6, GSH and 8-iso-prostaglandin F2α as classical inflammatory and oxidative stress markers is lacking in our study, which makes it difficult to draw definite and consistent conclusion that inflammation and oxidative stress may mediate the association of serum albumin with DPN. Third, although many confounding factors were adjusted in our study, it is possible that there are still some residual confounding and unmeasured factors, which may affect the true association between serum albumin and DPN. In spite of some limitations, our study has some strength, including a relatively large sample size and a thorough clinical and laboratory assessment and appropriate adjustment, which can raise the reliability of our findings.