Urinary orosomucoid and retinol binding protein levels as early diagnostic markers for diabetic nephropathy

Background: Diagnosing diabetic nephropathy is important to prevent long-term kidney damage and determine the prognosis of patients with diabetes. Since some kidney injury biomarkers increase in the early stages of diabetic nephropathy, this study investigated the clinical signicance of combined detection of urine orosomucoid and retinol binding protein for early diagnosis of diabetic nephropathy. Methods: We recruited 72 patients with type 2 diabetes and 34 healthy persons from August 2016 to July 2018 at our hospital. Using Mogensen grading criteria, participants were classied as diabetes or diabetic nephropathy, and healthy persons constituted the control group. Urine orosomucoid and retinol binding protein were measured and correlated with other variables. Results: Increase in renal damage raised urinary orosomucoid level gradually (P < 0.05). Urinary retinol binding protein and microalbumin levels were signicantly higher in the diabetes group than in control and nephropathy groups. Orosomucoid and retinol binding protein might be independent risk factors for diabetes and diabetic nephropathy. Urinary orosomucoid signicantly correlated with retinol binding protein and microalbumin (r = 0.489 and 0.513, respectively) in the diabetic nephropathy group. The receiver operating characteristic curve yielded a sensitivity, specicity, and correction index of 0.941, 0.842, and 0.783, respectively, while analysis for retinol binding protein yielded a sensitivity of 0.942, specicity of 1.000, and a correction index of 0.941. Conclusion: The increase in the levels of urine orosomucoid and retinol binding protein can be detected in the early stages of type 2 diabetic nephropathy. Both these markers are important for diabetic nephropathy detection and early treatment.

Urinary orosomucoid and retinol binding protein levels as early diagnostic markers for diabetic nephropathy Xue

Introduction
Diabetic nephropathy (DN) is the most common and severe chronic vascular complication among patients with type 2 diabetes mellitus (T2DM) [1]. It leads to chronic renal failure and is the leading cause of death due to diabetes. However, DN often occurs with no obvious symptoms in the early stage.
Although renal biopsy is the most important diagnostic method for DN, it is associated with some trauma; as most patients nd it di cult to endure, it cannot be readily employed for routine examination.
Common diagnostic indicators for DN include 24-hour urine microalbumin (MAL), urea nitrogen, and serum creatinine. However, they can be affected by many factors, such as urinary tract or systemic infections, strenuous exercise, bleeding, or drugs that affect the kidneys [2]. The accuracy and speci city of these indicators are not high, and they have limitations [3]; thus, more research is needed to identify newer, more accurate, and speci c early diagnostic markers of DN. At present, preliminary progress has been made in kidney disease research using proteomics technology. Diabetic urine proteome research has shown that some protein markers have a predictive effect for early DN [4,5]. In this study, the urine orosomucoid and retinol binding protein (RBP) levels were measured in healthy people, patients with T2DM, and patients with early DN. The differences between the three groups were compared. We also assessed their clinical signi cance in the diagnosis of early type 2 DN and their clinical value in the progression of nephropathy.

Data Collection
We recorded the medical history of all patients, and measured their diastolic blood pressure (DBP), systolic blood pressure (SBP), height (cm), and weight (kg). The levels of total cholesterol (CHOL), lowdensity lipoprotein cholesterol (LDL), triglycerides (TG), fasting blood glucose (FBG), 2-hour blood glucose, glycosylated hemoglobin (HbA1c) and serum creatinine (SCr) were measured. Urine was collected from 10 pm to 6 am the following morning, and the total urine volume (mL) was recorded after mixing. The urine MAL level was assessed for 24 hours. The urine orosomucoid and RBP levels were measured by the ELISA method. The MDRD equation [7] was used to estimate the glomerular ltration rate (eGFR).

Statistical analysis
For normally distributed data as determined by the Shapiro-Wilk's test, the indicators in each group were expressed as the mean ± standard deviation. Chi-squared tests were used to compare quantitative data between groups. The mean values for each of the three groups were compared using one-way analysis of variance. If there were signi cant differences between the groups, intra-group comparisons were performed using least signi cant difference. A binary logistic regression model was used to determine the factors associated with T2DM and T2DN, and correlation analyses were performed using Spearman's rank correlation. The receiver operating characteristic (ROC) curve was used to analyze the diagnostic points and diagnostic value of orosomucoid and RBP in DN. All statistical analyses were performed using SPSS version 23.0 (IBM Corp., Armonk, NY, USA). A two-tailed test with P < 0.05 was considered statistically signi cant.

Results
There were no signi cant differences in age and gender among the three groups (P > 0.05).There were no statistically signi cant differences in DBP, BMI, CHOL, TG, and LDL levels among the three groups (P > 0.05) ( Table 1). There were signi cant differences in SBP, HbA1c, and FBG between the NC and the other two groups (P < 0.05). However, there was no statistical difference in the general clinical data between the T2DM and T2DN groups (P > 0.05). As renal damage increased in patients, urine orosomucoid levels gradually increased as well (P < 0.05) ( Table 2). Urine RBP and MAL levels in the T2DN group were signi cantly higher than those in the NC and T2DN groups (P < 0.001). The eGFR levels in the T2DN group were signi cantly lower than in the NC and T2DM groups (P < 0.001). There were no signi cant differences in RBP, eGFR, and MAL levels between the NC and T2DM groups (P > 0.05).
For the NC group and the T2DM group, the dependent variable was whether T2DM had occurred (Yes = 1, No = 0), and the independent variables were the four variables with differences between the two groups, shown in Tables 1 and 2, including SBP, HbA1c, FBG, and orosomucoid. A binary logistic regression model was established and used to determine the in uence of these four variables on T2DM (Table 3), and all were shown to be risk factors (all OR > 1, p < 0.05). For the T2DM group and the T2DN group, the dependent variable was whether T2DN had occurred (Yes = 1, No = 0), and the independent variables were the ve variables with differences between the two groups shown in Tables 1 and 2. A binary logistic regression model was established for analysis (Table 4). Abbreviations: RBF, renal blood ow; eGFR, estimated glomerular ltration rate; MAL, microalbumin; SBP, systolic blood pressure; OR, odds ration; CI, con dence interval.
Of the ve factors that were included in the regression model (p < 0.05), SBP, orosomucoid, RBP, and MAL were all determined to be risk factors (OR > 1) and eGFR was shown to be a protective factor (OR = 0.948 > 1). Correlation analysis showed that in the T2DN group, urinary orosomucoid level was signi cantly positively correlated with RBP (r = 0.489) and MAL (r = 0.513). RBP and MAL were signi cantly positively correlated with a correlation coe cient of 0.468. eGFR and urine orosomucoid, RBP, and MAL were signi cantly negatively correlated (r = -0.577, -0.474, and − 0.466, respectively).
ROC curve analysis was used to assess the diagnostic points and diagnostic value of orosomucoid and that of RBP to predict DN. Figure 1 and Table 5 show the areas under the ROC curves for orosomucoid and RBP with the respective standard error values. Abbreviations: RBP, renal blood ow; ROC, receiver operating characteristic; -LR, negative likelihood ratio; LR, positive likelihood ratio.
The diagnostic value of DN had improved; however, no statistical signi cance was observed (Z = 0.598, P = 0.550 > 0.05). The diagnostic point of orosomucoid was 22.43, sensitivity was 0.941, speci city was 0.842, and Youden index was 0.783. The diagnostic point of RBP was 0.53, sensitivity was 0.942, speci city was 1.000, and Youden index was 0.941.

Discussion
DN has become the leading indication for dialysis due to end-stage renal disease (ESRD) [8,9]. Recent ndings suggest that immune-mediated in ammatory processes play a crucial role in DN. Many prein ammatory cells, growth regulators, and adhesion factors interact with each other and cross-link, resulting in an expansion of the corresponding cascade of in ammation [10]. In recent years, the rapid development of proteomics technology has provided us with new methods and ideas for identifying early diagnostic markers of DN. Proteomics techniques have been used to identify disease-speci c biomarkers and other related proteins in urine. Differential proteins have been identi ed, and some protein markers were found to have predictive effects for glomerular diseases.
Orosomucoid protein is a non-speci c acute phase reaction protein that is mainly synthesized and secreted by the liver, is low in healthy body uids, but is signi cantly increased in a state of in ammation or for people with tumors. Orosomucoid can act in damaged areas, be released into the circulation and intercellular uid, and become involved in the induction and regulation of body damage, immune, and in ammatory responses [11]. Elevated urine orosomucoid levels in T2DM patients have predictive effects on cardiovascular complications and mortality [12]. El-Beblawy et al. [13] pointed out that orosomucoid is an independent factor of diabetic microvascular complications and can be considered as an early marker of kidney damage. Fandiño-Vaquero et al. [14] also found that orosomucoid levels increased in patients with T2DM, and that it might mirror local endothelial dysfunction or in ammatory processes. Although there exists research and knowledge about orosomucoid, research on changes in orosomucoid concentration in urine during the early stage of DN is lacking. In this study, we found that as the disease progressed, urine orosomucoid levels gradually increased (P < 0.001). Orosomucoid might be an independent risk factor for T2DM and T2DN, and it had a signi cant positive correlation with MAL (r = 0.489) and a signi cant negative correlation with eGFR (r = -0.577). The results also revealed an increase in orosomucoid in the early stage of DN, suggesting that this increase may promote the occurrence and development of DN.
RBP is ltered through the glomerulus and absorbed and degraded by proximal tubular epithelial cells.
Therefore, it is generally stable in urine, di cult to decompose, and has a low excretion rate. An increase in RBP excretion may re ect the extent of damage [15,16]. Studies have shown that urine RBP levels in patients with T2DM are closely related to DN [17,18]. Some studies have shown that urine RBP can be used to assess the degree of renal interstitial brosis due to various causes, progress with ESRD dialysis, and even diabetes related to increased risk of death [19,20]. In this study, we found that urine RBP and MAL levels in the T2DN group were signi cantly increased (P < 0.001). Therefore, RBP might be an independent risk factor for T2DN, as it had a signi cant positive correlation with MAL (r = 0.468) and a signi cant negative correlation with eGFR (r = -0.474). Urine RBP may re ect early renal damage in DN. In the area under the ROC curve for predicting DN by orosomucoid and RBP both factors have high sensitivity and speci city. Therefore, both orosomucoid and RBP can be used to diagnose DN.
The main limitation of this study is that all the participants were residents of Henan Province, China, and the sample size was small. Further veri cation is needed through large sample sizes and multicenter studies.

Conclusions
In summary, urine orosomucoid and RBP may be new markers for the early diagnosis of DN, which is of great signi cance for the early detection and treatment of DN. However, the underlying molecular mechanisms and the clinically important levels of these potential biomarkers need to be studied further. The study protocol was approved by the ethics committee of The First A liated Hospital of Henan Polytechnic University (Jiaozuo Second People's Hospital) (IRB number: 2016010), and all patients provided written informed consent to participate in the study.

Consent for publication
Consent publication was provided.

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request and with the permission of The First A liated Hospital of Henan Polytechnic University ethics committee. Receiver operating characteristic curve of two indicators for predicting diabetic nephropathy RBP: retinol binding protein