Increased neutrophil count Is associated with the development of chronic kidney disease in patients with diabetes

Abstract Background This study aims to investigate the potential association of peripheral inflammatory blood cell parameters with the incidence and progression of chronic kidney disease (CKD) in patients with diabetes. Methods The cross‐sectional study included 1192 subjects with diabetes derived from one center. The cohort study included 2060 subjects with diabetes derived from another two centers followed up for 4 years. Logistic regression and Cox proportional hazards models were used to evaluate the association of peripheral inflammatory blood cell with CKD. Results In the cross‐sectional study, neutrophil count performed best as an independent risk factor for CKD (odds ratio 2.556 [95% confidence interval 1.111, 5.879]) even after 1:1 case–control matching for age, gender, history of high blood pressure and duration of diabetes. Spline regression revealed a significant linear association of CKD incidence with continuous neutrophil count in excess of 3.6 × 109/L. In the cohort study, subjects were grouped based on tertile of neutrophil count and neutrophil‐to‐lymphocyte ratio. Cox regression analysis results showed that only neutrophil count was independently associated with CKD progression (the highest group vs. the lowest group, hazard ratio 2.293 [95% confidence interval 1.260, 4.171]) after fully adjusting for potential confounders. The cumulative incidence of CKD progression in patients with diabetes gradually increased with increasing neutrophil count (53 (7.7%) subjects in the lowest group vs. 60 (8.2%) in the middle group vs. 78 (12.2%) in the highest group). Conclusions This study suggested that neutrophil count is an independent risk factor for progression of CKD in patients with diabetes.


| INTRODUCTION
The global prevalence of diabetes mellitus has increased enormously over the past few decades. Approximately half of all affected patients will develop chronic kidney disease (CKD). 1 Of the microvascular and macrovascular complications of diabetes, CKD imposes a high financial burden. Cardiovascular mortality and progression to endstage renal disease are the two major adverse health outcomes in patients with diabetes and CKD. 2 Consequently, prevention or management of CKD is the primary aim of management in patients with diabetes. Nonetheless current management that consists of controlling risk factors such as hyperglycemia, hypertension, hyperuricemia, and lipid regulation do not prevent the progression of nephropathy. Therefore, exploring the pathophysiology of CKD in diabetes is crucial to its management.
An increasing number of studies has indicated that the pathogenesis of CKD in diabetes is multifactorial, and lowgrade chronic inflammation plays a crucial role. [3][4][5][6] Routine inflammatory blood cell parameters such as white blood cell (WBC) count, neutrophil count, and neutrophil-tolymphocyte ratio (NLR) have been investigated for an association with and ability to predict diabetes and CKD. [7][8][9][10] Nonetheless results have been inconsistent, possibly due to small sample sizes and a lack of well-characterized study cohorts.
This study aimed to investigate the potential association of peripheral inflammatory blood cell parameters with the incidence and progression of CKD in patients with diabetes by cross-sectional study of expanded sample size and cohort study. First, in the cross-sectional study, WBC count, neutrophil count and NLR were all increased and lymphocyte count was decreased in patients with CKD. All were also associated with parameters of nephropathy including creatinine (Cr), estimated glomerular filtration rate (eGFR) and urine albumin/creatinine ratio (UACR). Nonetheless neutrophil count was the best independent risk factor for and predictor of CKD. In addition, significant linear associations between continuous neutrophil count, NLR and the incidence of CKD were analyzed by spline regression. Finally, our cohort study further confirmed that an increased neutrophil count, not NLR, predicted onset of CKD in diabetes. Moreover, there was a linear relationship between increasing neutrophil count and cumulative incidence of CKD. Hospital of Shanghai, Fudan University (2010-2024). Consent has been obtained from each patient or subject after full explanation of the purpose and nature of all procedures used. The study procedure is described in Figure 1. In cross-sectional study, definition of CKD is based on the presence of decreased kidney function (ie, GFR <60 ml/min per 1Á73 m 2 ) or UACR ≥ 30 mg/g (two of three specimens of UACR collected within a 3-to 6-month period). 11 In the cohort study, CKD progression was defined in the cohort as 30% decline in eGFR. Participants were excluded for any of the following reasons: (1) presence of any infectious disease excluding by medical history collection and routine examination such as chest X-ray, electrocardiogram, and biochemistry test; (2) presence of viral infection or positive carrier status (hepatitis B virus, syphilis or HIV); (3) previous diagnosis of urinary tract infection, urolithiasis, or liver cirrhosis; and (4) other known renal disease: IgA nephropathy, nephrotic syndrome. All study subjects were of Han Chinese origin and lived in the same region at the time of the study.

| Data collection and laboratory assessments
All subjects underwent blood and urine sample analyses. After an overnight fast for 12 h, blood samples were collected for blood cell counts (Automatic Blood cell analyzer, Sysmex XN9000, Japan) and measurement of biochemical parameters (Automatic biochemical analyzer, Roche Cobas 8000, Switzerland). A urine sample was collected for assessment of UACR. Anthropometric parameters were used to calculate body mass index (BMI), and blood pressure was measured with a digital automatic blood pressure monitor (model HEM-907; Omron, Tokyo, Japan). A patient questionnaire was completed and included information about history of high blood pressure (HBP) and duration of diabetes.

| Statistical analysis
Data are shown as mean ± SD, unless otherwise noted. T-test was used for between-group comparisons of continuous variables and the χ 2 test for categorical variables.
Skewed variables are expressed as median (interquartile range), and rank sum test was used to compare the difference between two groups. To avoid potential bias due to uneven distribution of covariates between type 2 patients with or without CKD, a case-control matching method was employed to match variables that included gender, age, history of HBP and duration of diabetes. Matching tolerance was 0, 2, 0, and 0 respectively. Logistic regression analysis was performed to detect the predictability of WBC, neutrophil and lymphocyte count, and NLR for CKD. The covariates were prespecified for inclusion in multivariable models based on clinical and biologic importance. A continuous F I G U R E 2 Association of WBC, neutrophil, and lymphocyte count and NLR with Cr, eGFR, and UACR. Association of WBC, neutrophil, and lymphocyte count and NLR with Cr, eGFR, and UACR. WBC count was positively associated with Cr (r = 0.124, p < 0.001) (A) and UACR (r = 0.129, p = 0.001) (C) and negatively associated with eGFR (r = À0.072, p = 0.014) (B). There was also a strong association between neutrophil count, Cr (r = 0.202, p < 0.001) (D) and UACR (r = 0.206, p < 0.001) (F) and a negative association with eGFR (r = À0.169, p < 0.001) (E). Additionally, NLR exhibited a significant and positive association with Cr (r = 0.220, p < 0.001) (J) and UACR (r = 0.237, p < 0.001) (L), but a negative association with eGFR (r = À0.222, p < 0.001) (K). Cr, creatinine; eGFR, estimated glomerular filtration rate; NLR, neutrophil-to-lymphocyte ratio; UA, uric acid; UACR, urinary albumin-to-creatinine ratio; WBC, white blood cell. e association of neutrophil count and NLR with CKD incidence was determined by spline regression analysis.
In the retrospective cohort study, patients were divided into three groups based on tertile of neutrophil count: lowest group (<3.19 Â 10 9 /L), middle group (3.19-4.10 Â 10 9 /L), and highest group (>4.10 Â 10 9 /L); and NLR: lowest group (<1.58), middle group (1.58-2.16), and highest group (>2.16). Analysis of variance (ANOVA) and Student's t-test were used to identify difference in mean between groups. Variables with a skewed distribution were analyzed by Wilcoxon tests. The cumulative incidence of CKD in relation to neutrophil tertile was assessed by Kaplan-Meier plots.

| Increased WBC and neutrophil count and NLR were associated with Cr, eGFR, and UACR
Simple linear regression analyses were performed to investigate the correlation between CKD and biochemical and inflammatory blood cell parameters in patients with diabetes. Level of creatinine was positively associated with age, BMI, TG, and UA (all p < 0.05) and negatively associated with HDL (p < 0.05). There was also a strong association between UACR and age, SBP, UA, TC, TG, and LDL (all p < 0.05), whereas eGFR was negatively associated with age, SBP, DBP, BMI, and UA (all p < 0.001) ( Table 2). There was a significant and positive association of WBC count, neutrophil count, and NLR with Cr and UACR (all p < 0.001), but a negative association with eGFR (all p < 0.001). Additionally, lymphocyte count was negatively associated with Cr and positively associated with eGFR (all p < 0.05) (Table 2, Figure 2). F I G U R E 3 Continuous association of neutrophil count and NLR with the incidence of CKD in diabetes. Adjusted for BMI and gender. A for neutrophil count; B for NLR. The risk of developing CKD in diabetic patients increased when neutrophil count exceeded 3.6 Â 10 9 /L and NLR exceeded 2.8. BMI, body mass index; CKD, chronic kidney disease; NLR, neutrophil-to-lymphocyte ratio 3.3 | Increased neutrophil count was associated with CKD both in all subjects and 1:1 matched subjects, whereas NLR was associated with CKD in all subjects 3.4 | Spline regression: Continuous neutrophil count and NLR were both associated with the incidence of CKD in patients with diabetes After adjusting for gender and BMI, two spline regression models showed a significant relationship between continuous neutrophil count and NLR, and the incidence of CKD. The risk of developing CKD in patients with diabetes increased when neutrophil count exceeded 3.6 Â 10 9 /L and NLR exceeded 2.8 ( Figure 3).

| Increased neutrophil count predicted the development of CKD in the retrospective cohort study
As a retrospective cohort study 2060 patients with diabetes were recruited from two other centers. First, we analyzed dynamics of neutrophils and NLR, with reference parameters including mean, standard deviation, and coefficient of variation (CV). We found that the mean value of neutrophils and NLR was relatively stable during the 4 years, and the CV were all less than 25% and 30% respectively (Table S1 and Figure S1). All subjects were stratified at baseline into one of three groups according to tertile of neutrophil count and NLR. Across the lowest, middle, and highest groups of neutrophil tertile, there was a stepwise increase in the proportion of male patients (42.5%, 46.0%, 50.0%, p = 0.024) and patients with HBP (76.7%, 82.2%, 82.0%, p = 0.037), as well as the level of TG, Cr, fasting glucose, and HbA1c (all p < 0.001) and a stepwise decrease in the eGFR and HDL level (both p < 0.05). Across the lowest, middle and highest groups of NLR, there was also a stepwise increase in the proportion of male patients (40.0%, 44.8%, 53.4%, p < 0.001), as well as age, level of fasting blood glucose, and Cr (all p < 0.001), and a stepwise decrease in eGFR (p = 0.046) and TC, LDL, and HDL level (p < 0.05) ( Table 4).
Among the 2060 subjects with diabetes, there was an increasing incidence of eGFR decrease >30% in a graded T A B L E 5 Adjusted hazard ratios of incidence of estimated glomerular filtration rate decrease >30% in relation to tertile of neutrophil count among subjects with diabetes  Table 5). The cumulative incidence of CKD gradually increased with increasing neutrophil count ( Figure 4). Although Cox regression model of NLR was also implemented, the HR for eGFR decrease >30% showed no difference across the NLR tertiles (Table 5).

| DISCUSSION
This cross-sectional study and retrospective cohort study is the first with a large sample size to confirm an independent association of neutrophil count, not WBC, lymphocyte count or NLR, with the progression of CKD in patients with diabetes. A growing number of studies have described the crucial role of chronic inflammation in the development of CKD in patients with diabetes. [4][5][6][12][13][14] As the most important target of microvascular damage in diabetes, CKD is associated with both systemic and local renal inflammation with the participation of crucial inflammatory cells and molecules. As a simple indicator of inflammation, peripheral blood leukocytes comprising neutrophils, monocytes, and lymphocytes have previously been found to be related to the development of CKD in patients with diabetes. [7][8][9][10]15 Our study suggested that of the peripheral blood leucocytes, neutrophil count was the most reliable independent risk factor for CKD, both in the cross-sectional and cohort studies. Neutrophil count and lymphocyte count are the principal components of the WBC count, and increasing WBC in CKD would have been largely due to the increased number of neutrophils. Because NLR is the ratio of neutrophil count to lymphocyte count, the increased NLR would likely have been due to increased neutrophils and decreased lymphocytes. As the largest fraction of white blood cells, neutrophils are involved in chronic meta-inflammatory states such as obesity, insulin resistance, type 2 diabetes, gestational diabetes mellitus, and coronary artery disease. 10,[16][17][18] Previous studies have revealed lymphocytes to be protective in coronary heart disease and heart failure. 19,20 Although our study and results from Chung et al 7 both showed a decreased lymphocyte count, neither lymphocyte nor NLR were independent risk factors for CDK in patients with diabetes. It is proposed that the decreased lymphocyte count compensates for the increased neutrophils to maintain leukocyte homeostasis. Consequently, the study proved neutrophil count is most closely associated with CKD in patients with diabetes, not WBC, NLR, or lymphocyte count. This provides evidence of the important pathological role of innate immune cells in the development of CKD in patients with diabetes.
Increased neutrophils may participate in the occurrence and development of CKD in many ways. Studies from Mahfouz et al and Bolignano et al showed that neutrophil gelatinase-associated lipocalin (NGAL) produced by neutrophils is a potential predictor of CKD and its progression, 21,22 although NGAL is not only synthesized by neutrophils; other cells are also involved in its synthesis such as epithelial cells. 23 Other studies suggest that hyperglycemia promotes an increase in number of circulating neutrophils, and the migration of neutrophils through chemokines to the glomerular basement membrane injury site promotes the inflammatory cascade through further chemotaxis of mononuclear macrophages. 24 Neutrophils may also contribute to CKD progression through neutrophil elastase (NE) that is secreted by activated neutrophils. 25,26 The proinflammatory effect of NE has been confirmed in a variety of disease models. 27 Previous studies revealed that neutrophils contribute to the etiology of insulin resistance via secreted NE, which leads to the degradation of insulin receptor substrate 1. 27,28 NE also leads directly to renal cell damage, contributing to continuous progression of CKD in patients with type 2 diabetes. 29 All these results demonstrate that neutrophils may contribute to CKD.
Our study also found that UA, advanced age, duration of diabetes, and history of hypertension were risk factors for CKD in patients with diabetes, consistent with previous studies. 1,30 It is clear that the progression of diabetic nephropathy is a complex process with many factors involved in the pathogenesis. Sembach et al reported no or only mild gender differences in the functional and structural changes during CKD progression in an animal study. 31 On the contrary, in our study, female gender was an obvious risk factor for CKD in patients with diabetes. This may be due to hormonal changes in women after menopause and should be further verified in aged mice. We also found that advanced age was an independent risk factor for renal dysfunction, consistent with previous study demonstrating that eGFR declines in parallel with age. 32,33 Aging may therefore also be involved in the occurrence and development of CKD via its own mechanisms.
There were some limitations of this study. Although we have proved that among all white blood cells, neutrophils were the most predictive for CKD in patients with diabetes, the mechanism of the potentially pathophysiological role of neutrophils in the development of CKD has not been studied clinically. Further research in humans and in animal studies should be performed to clarify the function of neutrophil-secreted proteins, especially NE and NGAL.
In conclusion, our study is the first to demonstrate the close association of neutrophil count with CKD in patients with diabetes. Neutrophil count was an independent risk factor for CKD especially when it exceeded 3.6 Â 10 9 /L and increased neutrophil count could predict a faster decline of eGFR in patients with diabetes.

SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.