1. Introduction
Retinopathy is a well-known microvasculature disease and one of the complications of diabetes and hypertension [
1,
2]. Diabetic retinopathy (DR) is suggested to be a leading cause of blindness, because its prevalence is approximately 33% in western countries [
3,
4]. In the 2005–2008 survey years of the National Health and Nutrition Examination Survey (NHANES), data reveal that approximately 30% of people with diabetes mellitus (DM) were at risk of DR [
5]. The precipitating factors related to DR have been reported in previous studies [
6,
7]. Poor control of diabetes and hypertension and long DM duration are strongly associated with DR [
8].
Iron plays an important role in catalyzing enzymatic reactions. It is a critical component of the electron transport chain [
9]. Iron represented a vital cofactor for guanylate cyclase, an enzyme that synthesizes cGMP, the second messenger in the retinal phototransduction [
10,
11]. Although an important micronutrient for the physiological processes of several proteins, excess iron can be harmful. Various types of retinal degeneration are suspected to result from iron-mediated retinal cell death [
12]. Increased iron accumulation is reported to accelerate the progression of DR in mouse models of DM [
13]. The primary focus of this paper was on the correlation of serum iron with DR among these US individuals.
2. Methods
2.1. Study Population
The NHANES is executed by the National Center for Health Statistics (NCHS), which contains these US noninstitutionalized civilian participants. All participants underwent comprehensive measurements, such as physical and laboratory examinations and standardized interview questionnaires, including socioeconomic, demographic, and health-related questions. From survey years 2005–2008 of the NHANES, 5321 eligible participants obtained were included in our study. According to the Declaration of Helsinki, this design was approved by the institutional review board of the NCHS. Before examinations, all participants completed informed consents.
2.2. Definition of DR and Ophthalmic Examination
DR was defined by the presence of hemorrhages, hard exudates (HE), cotton wool spots (CWS), microaneurysms (MA), venous beading, intraretinal microvascular abnormalities (IRMA), and retinal new vessels based on the severity scale of the Early Treatment for Diabetic Retinopathy Study (ETDRS). Nonmydriatic fundus photography (TRC-NW6S; Topcon, Tokyo, Japan) was applied for measuring the level of retinopathy in the worse eye. The grades were categorized into no DR, nonproliferative DR, and proliferative DR. Detailed information is listed in the Digital Grading Protocol of the NHANES.
2.3. Measurement of Serum Iron, Lead, Cadmium, and Mercury
The serum iron level was measured by the DcX800 method. The system detects the change of ferrous ion absorbance, complexed with the ferro zine iron reagent, at 560 nm at a fixed-time interval. The concentration of iron in the sample is directly proportional to the change in absorbance. The concentration of ferritin was measured by the Roche/Hitachi immunoturbidity assay. The function of a transferrin receptor is to transport iron into cells and maintain iron metabolism. The concentration of a transferrin receptor was also measured by the Roche/Hitachi immunoturbidity assay. The concentration of hemoglobin was determined using the Coulter HMX Hematology Analyzer. The levels of serum lead and cadmium were evaluated by atomic absorption spectrophotometry with Zeeman background correction (SIMAA 6000 model; Perkin-Elmer, Norwalk, CT, USA). Serum inorganic mercury was measured using stannous chloride as a reductant, and we also utilized microwave digestion.
2.4. Covariates Assessment
There are five races/ethnicities in the NHANES database, including Mexican American, non-Hispanic white, non-Hispanic black, other Hispanic, and Others. Current cigarette smoking was defined as having ever smoked and smoking at the time of the survey. Before the determination of biochemical data, blood samples were obtained from participants with 8 h fasting.
2.5. Statistical Analysis
The Statistical Package for the Social Sciences (version 18.0, SPSS Inc., Chicago, IL, USA) was applied for analysis. The differences of continuous and categorical variables were investigated using the independent t-test and the chi-squared test, respectively. These logistic regression models were used to determine the relationship of serum iron with the presence of DR. Model 1 was unadjusted. Model 2 was adjusted by age, gender, and race. Model 3 = Model 2 and adjusted by alanine aminotransferase (ALT), fasting plasma glucose (FPG), hemoglobin, and cigarette smoking. To assess the optimal cutoff points of serum iron, a receiver operating characteristic (ROC) curve analysis was conducted. Two-sided p values of less than 0.05 were considered to indicate significance.
4. Discussion
In the present study, the occurrence of DR had a significant association with serum iron rather than other toxic metals. Participants with DR had lower levels of serum iron than control groups. Compared with the ferritin and transferrin receptors, we highlighted that only serum iron had a significantly inverse relationship with the presence of DR. Notably, the present study was the first to examine the association of serum iron with the presence of DR.
Our results were contrary to a previous study, which indicated that excessive iron accelerates the progression of DR in mice [
13]. Compared with this animal research, we measured the iron level from serum samples rather than directly from retinal samples. Iron overload in the retina was suggested to result in cell death by generating reactive oxygen species and releasing proinflammatory cytokines [
14]. Our findings implied that the development of DR might have different potential mechanisms related to serum iron.
DR is considered a microcirculatory disease of the retina triggered by hyperglycemia. Myriad risk factors have been reported for prediction of the development and severity of DR, including high blood pressure, proteinuria, high serum lipids, the duration of DM, and renal disease. Iron deficiency is suggested to be associated with increased hemoglobin Alc levels in patients with diabetes [
15,
16]. Ohira et al. proposed that decreased iron levels contributed to elevated gluconeogenic enzymes and serum glucose levels [
17]. In a population-based study, lower levels of serum iron were related to glucose dysregulation [
18]. In a case-series study of three patients who had DM for an average of 17 years, the background DR of the patients rapidly progressed from a mild-to-moderate grade to a severe proliferative phase, after the development of severe iron deficiency anemia [
19]. Consistent with our findings, we found that low serum iron was significantly associated with an increased likelihood of DR. Collectively, iron homeostasis might be a vital step in the progression of DR.
Maintenance of oxygen supply plays a critical role for retinal function [
20]. A long duration of hypoxia caused by hyperglycemia or low hemoglobin can lead to retinopathy [
21]. Most important of all, iron is incorporated into the heme for transport of oxygen in the blood [
22,
23]. Nagababu et al. reported that low iron levels enhanced red blood cell oxidative stress [
24]. Several studies have described the role of oxidative stress in the development of DR, in which excessive superoxide accelerates the production of advanced glycation end-product (AGE), expression of the AGE receptor, activation of protein kinase C isoforms, and overactivity of the hexosamine pathway [
25,
26]. Anemia is suggested to accelerate the progression of retinal ischemia in subjects with DR [
27]. Andrews et al. also demonstrated that DM patients with anemia exhibited higher expressions of IL-6 and C-reactive protein, compared to diabetic patients without anemia [
28]. These inflammatory mediators insulted vascular permeability and functions that harbored a predisposing milieu for DR formation.
There are some limitations to this study. First, even though a relationship between serum iron and the presence of DR was established, the causal relation could not be addressed due to the study’s cross-sectional design. Second, the NHANES did not include institutionalized citizens; thus, the prevalence of DR might be underestimated. Third, the presence of DM was obtained from a self-reported questionnaire and did not categorize diabetes into Type 1 and Type 2. Finally, we measured serum iron using a single point method; however, the concentration of serum iron varies with time and diurnal variation of serum iron was observed.