Relationship Between Biomarkers of Tubular Injury and Intrarenal Hemodynamic Dysfunction in Youth with Type 1 Diabetes

Abstract

Background:

Early identification of youth with type 1 diabetes (T1D) at risk for diabetic kidney disease may improve clinical outcomes. We examined the cross-sectional relationship between kidney biomarkers neutrophil gelatinase-associated lipocalin (NGAL), copeptin, interleukin-18 (IL-18), kidney injury molecule-1 (KIM-1), chitinase-3-like protein-1 (YKL-40), and monocyte chemoattractant protein-1 (MCP-1), and intrarenal hemodynamic function in adolescents with T1D.

Methods:

Urine albumin-to-creatinine ratio (UACR), renal vascular resistance (RVR), glomerular filtration rate (GFR), intraglomerular pressure (P_GLO), efferent arteriole resistance (R_E), afferent arteriolar resistance (R_A), and renal plasma flow (RPF), and the above indicated biomarkers were assessed in youth aged 12–21 years with and without T1D of <10 years duration.

Results:

Fifty adolescents with T1D (16.1±3.0 years, HbA1c 8.6±1.2%) and 20 adolescents of comparable BMI without T1D (16.1±2.9 years, HbA1c 5.2±0.2%) were enrolled. Adolescents with T1D demonstrated significantly higher GFR, RPF, R_E, and P_GLO than controls (39%, 33%, 74%, and 29%, respectively, all p<0.0001). Adolescents with T1D also exhibited significantly lower RVR and R_A than controls (25% and 155%, respectively, both p<0.0001). YKL-40 and KIM-1concentrations, respectively, were positively associated with GFR (r: 0.43, p=0.002; r: 0.41, p=0.003), RPF (r: 0.29, p=0.08; r: 0.34, p=0.04), UACR (r: 0.33, p=0.02; r: 0.50, p=0.0002), and P_GLO (r: 0.45, p=0.006; r: 0.52, p=0.001) in adolescents with T1D.

Conclusions:

Higher concentrations of biomarkers YKL-40 and KIM-1 may help define the risk for intraglomerular hemodynamic dysfunction in youth with T1D.

Graphical Abstract

Introduction

Diabetic kidney disease (DKD) is a well-established consequence of type 1 diabetes (T1D), and the number of people with chronic kidney disease (CKD) due to T1D has been steadily increasing for many years [1, 2]. The onset of DKD is difficult to detect because early DKD is clinically silent, even as kidney function declines and structural injury develops on kidney biopsy [3]. Early intraglomerular hemodynamic dysfunction characterized by elevated glomerular filtration rate (GFR), renal plasma flow (RPF), and intraglomerular pressure (P_GLO) is frequently observed in youth with T1D [4], but its effects on their risk of progressive DKD are uncertain. Current screening tools for DKD, which include urine albumin excretion and estimates of GFR, may not adequately capture this early stage of DKD [5]. Thus, discovery of early biomarkers of DKD in youth with T1D is necessary to identify young people at high risk for kidney disease progression prior to the onset of irreversible CKD.

Multiple kidney tubular injury biomarkers have been proposed as screening tools for early DKD detection, including neutrophil gelatinase-associated lipocalin (NGAL), monocyte chemoattractant protein-1 (MCP-1), interleukin-18 (IL-18), kidney injury molecule-1 (KIM-1), and chitinase-3-like protein-1 (YKL-40) [6–11]. Chronic inflammation, which stems from repeated diabetes-related immune responses, also results in generalized neurohormonal activation in DKD, including the stimulation of arginine vasopressin (AVP) release [12]. Studies have demonstrated that AVP increases the activity of the renin-angiotensin-aldosterone system and subsequently leads to intraglomerular hemodynamic dysfunction and increased sodium reabsorption, which further accentuates kidney energy expenditure and contributes to the development of DKD [1, 12]. Copeptin, a biomarker for AVP [12], has also been associated with elevated GFR in T1D and occurs in higher concentrations in youth with T1D than in controls. Copeptin concentration, therefore, may also serve as a promising biomarker to help predict the development and progression of DKD in T1D.

The relationship between the prominent intraglomerular hemodynamic dysfunction observed in young persons with T1D and the recognized biomarkers of tubular injury and neurohormonal activation described above remains poorly understood. In this analysis we examine the cross-sectional association of these biomarkers with intraglomerular hemodynamic function in 50 young persons with T1D. We hypothesize that biomarkers of tubular injury and neurohormonal activation will demonstrate a strong association with intraglomerular hemodynamic dysfunction in youth with T1D.

Methods

Study Design and Participants:

Fifty adolescents aged 12–21 years with T1D of less than 10 years duration and an HbA1c of less than 11% from the Copeptin in Adolescent Participants with Type 1 Diabetes and Early Renal Hemodynamic Function Study (CASPER, NCT03618420) and 20 healthy adolescent controls aged 12–21 years from the Renal Hemodynamics, Energetics, and Insulin Resistance in Youth Onset Type 2 Diabetes Study (Renal-HEIR, NCT03584217) were included in this analysis. Participants with T1D were recruited from the Barbara Davis Center for Diabetes pediatric clinic on the University of Colorado Anschutz Medical Campus in Aurora, Colorado [13]. American Diabetes Association criteria along with positive inulin, islet cell, glutamic acid decarboxylase, and/or zinc transporter 8 autoantibodies, and the persistent need for exogenous insulin since diabetes diagnosis define a diagnosis of T1D. All participants were examined in the Clinical and Translational Research Center (CTRC). Participants limited physical activity and consumed a fixed-macronutrient, sodium, and protein replete, weight-maintenance diet for 3 days prior to the study visit. Morning tests were performed following a 12 hour overnight fast. . The CASPER and Renal-HEIR cohorts have intentionally harmonized study protocols and were both approved by the Colorado Multiple Institutional Review Board (COMIRB). Guardians and/or the participants, as appropriate, provided informed and written consent to participate in the study.

GFR and RPF by Iohexol and P-aminohippurate Clearance Techniques, Urine Albumin-to-Creatinine Ratio, and Blood Pressure Assessments:

At a single study visit, two intravenous (IV) line were placed and then participants were asked to empty their bladders. Spot plasma and urine samples were collected prior to infusions of iohexol and p-aminohippurate (PAH) for assessment of GFR and RPF, respectively. Iohexol was administered through bolus IV injection (5 mL of 300 mg/mL [Omnipaque 300, GE Healthcare]). After an equilibration period of 120 minutes, blood collections for iohexol plasma clearance were collected at +120, +150, +180, +210, +240 minutes [11]. Because the Brøchner-Mortensen equation underestimates high values of GFR, the Jødal-Brøchner-Mortensen (JBM) equation was used to calculate the GFR [14]. A PAH bolus (2 g/10 mL, prepared at the University of Minnesota, dose of [weight in kg/75] × 4.2 mL, IND #140129) was given slowly over 5 minutes followed by a continuous infusion of 8 mL of PAH and 42 mL of normal saline at a rate of 24 mL/hour for 2 hours. After an equilibration period of 90 minutes, blood was drawn at +90 and +120 minutes, and the Schnurr method was used to calculate the PAH plasma clearance (in mL/min) [15]. PAH plasma clearance was computed as (12.8 mg/min)/S, where the PAH infusion rate was set at 12.8 mg/min, and S was the average steady-state plasma concentration (mL/min) at 90 and 120 minutes. The mean of the PAH plasma clearance at 90 minutes and 120 minutes was used to calculate RPF. RPF was calculated as PAH clearance divided by the estimated extraction ratio of PAH, which varies by GFR [16]. We report absolute GFR (mL/min) and RPF (mL/min) in the main analyses because the practice of indexing GFR and RPF for body surface area (BSA) underestimates hyperfiltration and hyperperfusion, and BSA calculations introduce noise into the clearance measurements [17].

In adolescents with T1D, GFR and RPF were measured during mild hyperglycemia (goal blood glucose 170–190 mg/dL [9.4–10.6 mmol/L]), which was achieved by a hyperglycemic clamp with paired 20% dextrose and insulin IV infusions. A hyperglycemic clamp was chosen to mimic the typical glycemic milieu of an adolescent with T1D (equivalent to an HbA1c ~7.6–8.2%) and maintain steady-state glycemic and insulin concentrations during the collection of the kidney measures [12, 18]. In the control participants, GFR and RPF were measured fasting without glucose modification to represent the normal physiology of an individual without diabetes. Urine albumin-to-creatinine ratio (UACR) was assessed by averaging measurements obtained from urine samples collected before and after kidney clearance studies.

Participants laid in a supine position for at least 5 minutes prior to completing blood pressure measurements which were obtained using an automatic blood pressure machine. Three measurements per participant were obtained and averaged. Mean arterial pressure (MAP) was computed as [systolic blood pressure + 2(diastolic pressure)]/3.

Intraglomerular Hemodynamic Parameter Calculations:

Indirect intraglomerular hemodynamic parameters, including P_GLO and afferent arteriolar resistance (R_A)/efferent arteriolar resistance (R_E) ratio, were estimated using equations derived by Gomez et al. [19, 20]. The following calculations were completed:

$$\text{Renal blood flow}\left(\text{RBF}\right)=\text{RPF}/\left(\text{1-Hematocrit}\right)$$

$$\text{Renal vascular resistance }\left(\text{RVR}\right)=\text{MAP}/\text{RBF}$$

Assumptions made by the Gomez equations include: (1) afferent, post-glomerular, and efferent intrarenal vascular resistances are assessed separately; (2) hydrostatic pressures within the Bowman’s space, kidney venules and tubules, and interstitium (P_BOW) are in equilibrium of 10 mm Hg; (3) the glomerulus, in contrast, is in disequilibrium during filtration; and (4) in a normal kidney, the gross filtration coefficient (K_FG) is 0.1733 mL/s/mm Hg. Additional intraglomerular hemodynamic parameters were also calculated using Gomez’s equations with the assumption of K_FG = 0.1012 mL/s/mm Hg for individuals with diabetes [21]. GFR (mL/s), RPF (mL/s), MAP (mm Hg), and total protein (g/dL) were used to calculate P_GLO (mm Hg), R_E (dyne*s/cm⁵), R_A (dyne*s/ cm⁵), filtration pressure across glomerular capillaries [ΔP_F (mm Hg)], and glomerular oncotic pressure [π_G (mm Hg)].

Glomerular capillary filtration pressure (ΔP_F) is calculated by:

$$\Delta {\text{P}}_{\text{F}}=\text{GFR}/{\text{K}}_{\text{FG}},{\text{where K}}_{\text{FG}}\text{is the gross filtration coefficient}$$

Glomerular oncotic pressure (π_G) from the capillary plasma mean protein concentration (C_M) is calculated by:

$${\text{C}}_{\text{M}}=\text{TP/FF}\times \text{ln}\left(\text{1}/\left(\text{1}-\text{FF}\right)\right)$$

$${\pi }_{\text{G}}=5\times \left({\text{C}}_{\text{M}}-\text{2}\right)$$

where C_M is determined by total protein (TP) and filtration fraction (FF).

Glomerular hydrostatic pressure (P_GLO) is calculated by the following equation, where P_BOW is assumed to be 10 mm Hg:

$${\text{P}}_{\text{GLO}}={\Delta \text{P}}_{\text{F}}+{\text{P}}_{\text{BOW}}+{\pi }_{\text{G}}$$

Ohm’s law principles were employed to estimate R_A and R_E, with a conversion factor of 1328 being used to convert to dyne/s/cm⁵:

$${\text{R}}_{\text{A}}\text{=}\left[\left(\text{MAP}-{\text{P}}_{\text{GLO}}\right)/\text{RBF}\right]\times 1328$$

$${\text{R}}_{\text{E}}=\left[\text{GFR}/\left({\text{K}}_{\text{FG}}\times \left(\text{RBF}-\text{GFR}\right)\right)\right]\times 1328$$

Tubular Injury and Neurohormonal Biomarkers:

Using an ultrasensitive assay on a KRYPTOR Compact Plus analyzer, copeptin concentrations were measured in serum via commercial sandwich immunoluminometric assays (Thermo Fisher Scientific, Waltham, MA). The copeptin assay has a 0.9 pmol/L lower limit of detection and a sensitivity of <2 pmol/L [1]. A Meso Scale Discovery (MSD) QuickPlex SQ120 platform was used to measure the serum IL-18, YKL-40, NGAL, MCP-1, and KIM-1 concentrations after a controlled sample thaw (Meso Scale Diagnostics, Gaithersburg, MD). The Meso Scale assay utilizes electrochemiluminescence detection and is a sandwich immunoassay organized in a patterned array format. MSD also provided the kidney injury panel 1 which was used to measure serum NGAL concentrations. The MSD U-plex assay was used to measure the serum IL-18, MCP-1, and YKL-40 concentrations, while the R-plex assay was used to measure KIM-1. From studies on adult healthy controls, the estimated normal range for NGAL was 28.7–167 ng/mL, for serum KIM-1 was <1 ng/mL, for IL-18 was 47–81 pg/mL, for MCP-1 was 13–55 pg/mL, and for YKL-40 was 69–197 ng/mL [8–10, 22, 23].

Laboratory Assessments:

The University of Colorado CTRC Core Labs or the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) laboratory in Phoenix, Arizona performed all laboratory assays for the CASPER cohort using standard methods.. Insulin was measured via Clinical Laboratory Improvement Amendments (CLIA)-certified chemiluminescent immunoassay (Beckman Coulter, Indianapolis, IN). Iohexol and PAH concentrations were measured in Phoenix by high-performance liquid chromatography (Waters, Milford, MA). HbA1c assay was calibrated to values traceable to the Diabetes Control and Complications Trial (DCCT).

Statistical Analyses:

Means and standard deviations were calculated for continuous variables with normal distributions. For all continuous variables with highly skewed distributions, medians and inter-quartile ranges (IQRs) were calculated. Outputs for categorical variables included numbers and percentages. T-tests and Chi-squared tests, as appropriate, were used to compare the baseline characteristics of adolescents with T1D and controls. Fisher’s exact test was used for categorical measures with limited observations. For non-normal continuous variables, a Wilcoxon rank-sum test was used. Pearson’s correlation coefficient and generalized linear regression models were used to examine the relationships among intraglomerular hemodynamic dysfunction, tubular injury biomarkers, and copeptin. Multivariable models were adjusted for age, sex, and HbA1c. Because data were decided a priori to be hypothesis generating, corrections for multiple comparisons were not performed. This study used an alpha value of 0.5 to define statistical significance. SAS version 9.4 was used to perform all statistical analyses (SAS Institute Inc).

Results

Participant Characteristics:

Fifty adolescents with T1D (16.1±3.0 years, HbA1c 8.6±1.2%) and 20 adolescents of comparable BMI without T1D (16.1±2.9 years, HbA1c 5.2±0.2%) were enrolled. Participants were stratified based on T1D status, and both groups had similar ages, pubertal stages, ethnicities, body mass indices (BMIs), fat masses, blood pressures, and lipid concentrations (Supplemental Table 1). There was a non-statistically significant higher ratio of females to males in the control group than the T1D group. The control group had significantly greater insulin sensitivity than the group with T1D (p<0.0001).

Intrarenal Hemodynamic Function Compared Between T1D and Control Groups:

GFR, RPF, R_E, and P_GLO were 39%, 33%, 74%, and 29% greater in the adolescents with T1D versus the controls (p<0.0001, Table 1). RVR and R_A were 29% and 155% lower in the adolescents with T1D versus the controls (p<0.0001).

Table 1.

Comparison of Kidney Function in Participants with and without Type 1 Diabetes

Kidney Function Variable	T1D (n = 50)	Controls (n = 20)	P value
GFR (mL/min)	189 ± 40	136 ± 22	<0.0001
GFR (mL/min/1.73m2)	183 ± 26	139 ± 8	<0.0001
RPF (mL/min)	820 ± 125	615 ± 65	<0.0001
RPF (ml/min/1.73m2)	824 ± 120	634 ± 85	<0.0001
RA (dyne/s/cm5)	977 ± 554	2494 ± 518	<0.0001
RE (dyne/s/cm5)	2041 ± 362	1173 ± 238	<0.0001
RVR (mm Hg/L/min)	0.07 ± 0.01	0.09 ± 0.01	<0.0001
PGLO (mm Hg)	72.76 ± 8.42	56.31 ± 4.38	<0.0001

^*Data are presented as mean ± standard deviation.

Abbreviations: GFR, glomerular filtration rate; P_GLO, glomerular pressure; R_A, afferent arteriolar resistance; R_E, efferent arteriolar resistance; RPF, renal plasma flow; RVR, renal vascular resistance; T1D, type 1 diabetes.

Relationship Between Biomarkers of Tubular Injury and Intrarenal Hemodynamic Function:

YKL-40 and KIM-1 concentrations, respectively, were positively associated with GFR (r: 0.43, p=0.002; r: 0.41, p=0.003), RPF (r: 0.29, p=0.08; r: 0.34, p=0.04), UACR (r: 0.33, p=0.02; r: 0.50, p=0.0002), and P_GLO (r: 0.45, p=0.006; r: 0.52, p=0.001) in adolescents with T1D (Table 2 and Table 3). Neither YKL-40 nor KIM-1 demonstrated associations with R_A, R_E, or RVR. NGAL, copeptin, and MCP-1 did not associate with any parameter of intrarenal hemodynamic function.

Table 2.

Relationships among Parameters of Intraglomerular Hemodynamic Function and Biomarkers of Tubular Injury and Copeptin in Participants with Type 1 Diabetes

Biomarker of Tubular Injury	GFR (mL/min)	RPF (mL/min)	UACR* (mg/g)	PGLO (mm Hg)	RA (dyne/s/cm5)	RE (dyne/s/cm5)	RVR (mm Hg/L/min)
IL-18* (pg/mL)	r: 0.13 p=0.36	r: 0.10 p=0.57	r: 0.01 p=0.96	r: 0.16 p=0.36	r: 0.17 p=0.33	r: 0.12 p=0.48	r: 0.13 p=0.43
YKL-40* (ng/mL)	r: 0.43 p=0.002	r: 0.29 p=0.08	r: 0.33 p=0.02	r: 0.45 p=0.006	r: −0.17 p=0.31	r: 0.36 p=0.03	r: −0.02 p=0.91
Copeptin (pmol/L)	r: 0.15 p=0.32	r: −0.10 p=0.56	r: −0.02 p=0.91	r: −0.05 p=0.79	r: −0.05 p=0.79	r: 0.06 p=0.71	r: −0.02 p=0.91
NGAL (ng/mL)	r: 0.05 p=0.72	r: 0.11 p=0.53	r: −0.09 p=0.55	r: 0.18 p=0.28	r: −0.19 p=0.25	r: 0.08 p=0.65	r: −0.08 p=0.62
MCP-1* (pg/mL)	r: −0.13 p=0.38	r: −0.00 p=0.98	r: −0.12 p=0.40	r: 0.01 p=0.95	r: −0.19 p=0.27	r: 0.01 p=0.94	r: −0.10 p=0.57
KIM-1 (ng/mL)	r: 0.41 p=0.003	r: 0.34 p=0.04	r: 0.50, p=0.0002	r: 0.52 p=0.001	r: −0.27 p=0.10	r: 0.24 p=0.16	r: −0.08 p=0.63

^*Indicates log transformation for normalization. All data are Pearson correlations.

Abbreviations: GFR, glomerular filtration rate; IL-18, interleukin-18; KIM-1, kidney injury molecule-1; MCP-1, monocyte chemoattractant protein-1; NGAL, neutrophil gelatinase-associated lipocalin; P_GLO, glomerular pressure; R_A, afferent arteriolar resistance; R_E, efferent arteriolar resistance; RPF, renal plasma flow; RVR, renal vascular resistance; T1D, type 1 diabetes; UACR, urinary albumin/creatinine ratio; YKL-40, chitinase-3-like protein-1.

Table 3.

Concentrations of Individual Biomarkers of Tubular Injury and Copeptin in Participants with Type 1 Diabetes

Biomarker of Tubular Injury	T1D
IL-18* (pg/mL)	695.56 (479.65 – 863.68)
YKL-40* (ng/mL)	24.91 (17.81 – 32.34)
NGAL (ng/mL)	96.02 ± 28.37
MCP-1 (pg/mL)	97.28 ± 19.83
KIM-1 (ng/mL)	0.027 ± 0.012
Copeptin (pmol/L)	8.26 ± 5.04

^*Data are presented as mean ± standard deviation or median (interquartile range), as appropriate.

Abbreviations: IL-18, interleukin-18; KIM-1, kidney injury molecule-1; MCP-1, monocyte chemoattractant protein-1; NGAL, neutrophil gelatinase-associated lipocalin; YKL-40, chitinase-3-like protein-1.

After adjusting for age, sex, and HbA1C, the relationships between YKL-40 and KIM-1, respectively, and log UACR (β±SE: 0.02±0.01, p=0.006; β±SE: 0.03±0.01, p=0.002), GFR (β±SE: 1.04±0.02, p=0.0001; β±SE: 1.39±0.42, p=0.002) and P_GLO (β±SE: 0.18±0.05; p=0.001, β±SE: 0.33±0.09, p=0.0005) remained statistically significant. In addition, the associations between YKL-40 and RPF (β±SE: 1.98±0.92, p=0.04) and R_E (β±SE: 5.60±2.72, p=0.049) also remained significant after multivariable adjustments.

Discussion

TID presents a significant challenge for the healthcare system in the United States, with an estimated prevalence of 1.6 million people affected in 2018, including 187,000 youth [2, 24]. DKD is a well-studied complication of T1D, and understanding and stratifying the risk for developing DKD in youth with T1D is of the utmost importance, as DKD remains one of the leading causes of morbidity and mortality in T1D [2, 24]. In this study, youth with short-standing T1D exhibited greater hyperfiltration, hyperperfusion, and intraglomerular hypertension than healthy controls, which suggests the development of intraglomerular hemodynamic dysfunction early in the course of T1D. Prolonged glomerular hypertension is thought to result in loss of glomerular charge and size selectivity. This loss of glomerular basement membrane integrity leads to increased transglomerular filtration of proteins and solutes, which can accentuate proximal tubular energy expenditure and predispose to hypoxia and injury. We presume that this dysfunction leads to tubular injury, and our results support this notion as demonstrated through the relationships between intraglomerular hemodynamic dysfunction and 2 different tubule injury biomarkers in the kidney, YKL-40 and KIM-1, which are elevated beyond a typical GFR effect. Consequently, YKL-40 and KIM-1 may serve as more targeted potential biomarkers for identifying and subsequently monitoring early kidney dysfunction in youth with T1D than classically used UACR.

Perturbations in intrarenal hemodynamic function currently represent the earliest indicators of developing DKD and have been demonstrated in both youth and adults with T1D prior to the development of significant albuminuria. Consistent with our findings, a Canadian study demonstrated abnormal baseline intraglomerular hemodynamic function in youth with T1D, including a higher GFR, RFP, and P_GLO and lower R_A than in either young adults or older adults with T1D [4]. Furthermore, perturbations in intrarenal hemodynamic function remain poor prognostic indicators as previous studies have demonstrated that intrarenal hemodynamic dysfunction predicts the development of future kidney failure in type 2 diabetes (T2D) [20]. In one recent study of Pima Indian adults with T2D, we found that an elevated R_A/R_E ratio and an elevated P_GLO that quickly declined were both associated with the development of kidney failure [25]. However, while intraglomerular hemodynamic function assessments remain the gold-standard for evaluating kidney function, our currently available methods are arduous, time-consuming, and expensive. Thus, the identification of easily accessible biomarkers that accurately predict the development of early intraglomerular hemodynamic dysfunction in individuals with diabetes is critical.

To date, few studies have assessed the use of kidney neurohormonal and tubular injury biomarkers to identify early DKD in people with T1D, and even fewer studies have been completed in youth with T1D. YKL-40 and KIM-1 are two examples of biomarkers that have recently undergone investigation in humans with diabetes. In a study of Japanese young adults aged 20–30 years with T1D, YKL-40 levels were shown to be associated with elevated urine albumin excretion rate [26]. Similarly, a case-cohort study involving 894 adult participants with diabetes and an eGFR <60 mL/min/1.73 m² from the Chronic Renal Insufficiency Cohort study, demonstrated associations among higher concentrations of YKL-40, KIM-1, and MCP-1 and a greater risk of DKD progression, as evidenced by the development of kidney failure or a 40% decline in eGFR [27]. By contrast, in a study of 100 youth and young adults aged 20 ± 2.8 years with T1D of 10.7 ± 5.2 years diabetes duration, researchers demonstrated that elevated KIM-1 levels were associated with elevated UACR and eGFR [28]. This study also employed kidney biopsies to demonstrate infiltration of KIM-1 positive T-cells in individuals with confirmed DKD [28]. These findings highlight the unique initial presentation of hyperperfusion and hyperfiltration in young persons with diabetes prior to an eventual decline in GFR associated with DKD progression, which often differs from older adults who experience more precipitous declines in GFR in the setting of DKD. We expand on findings that support YLK-40 and KIM-1 as important indicators of kidney dysfunction by demonstrating that both YKL-40 and KIM-1 were associated with an increase in GFR, RPF, and P_GLO, as well as UACR in a relatively healthy cohort of adolescents with T1D.

In addition to YKL-40 and KIM-1, the neurohormonal activator copeptin as well as tubular injury markers NGAL, IL-18, and MCP-1 have been proposed as biomarkers for DKD development. In a previous study of 169 youth with T1D, individuals in the highest tertile for copeptin demonstrated 4.29-fold greater odds of having elevated UACR after adjustments for age, sex, and eGFR [29]. Additionally, higher copeptin levels correlate with lower GFR in adults with T1D of >50 years duration [30]. In our study, we demonstrate a positive but statistically insignificant association between copeptin and parameters of intraglomerular hemodynamics, possibly secondary to changes in intraglomerular hemodynamics occurring earlier in the disease course than changes in UACR in youth with T1D. Additionally, elevated NGAL levels have also been shown to strongly associate with elevated copeptin levels in T1D [30], though the association was positive but not statistically significant in our study. Similarly, while IL-18 elevated in the setting of declining GFR in individuals with longstanding T1D and DKD [31], IL-18 also demonstrated non-significant positive associations with changes in GFR in our study. Lastly, urinary excretion of MCP-1 is low in adults with long-standing T1D and DKD [31]; however, we did not find any statistically significant associations with serum MCP-1 and intraglomerular hemodynamics to help support this finding. While our data describing the relationships between serum copeptin, NGAL, IL-18, and MCP-1 and intraglomerular hemodynamic function were not significant in this relatively small study of youth with T1D, we provide key preliminary data to serve as the foundation for future studies into the associations among tubular injury, neurohormonal biomarkers, and intrarenal hemodynamics in youth and young adults with diabetes.

This study has several strengths and limitations. We utilized a cross-sectional study design which restricts our data to a single timepoint and the relatively small sample size limits power. Participants with T1D demonstrated fairly well-controlled diabetes, as evidenced by their most recent HbA1c, which prevents generalizability of our results to individuals with poorly controlled T1D. However, because we did achieve relatively good average glycemic control in our participants with T1D, we minimized our result variability and thus increased the overall strength of the study. Additionally, while the biomarkers studied here have been previously researched in adults with T1D, we present the first study that we know of in youth and young adults with T1D. We also employed gold-standard measures of intraglomerular hemodynamics which improves accuracy through the collection of direct rather than estimated assessments. Lastly, we recognize the limitations of applying the Gomez equation to our study population since the ultrafiltration coefficients in the equation were derived from animal kidney micropuncture studies. At the time of publication, no gold standard data exists on glomerular size selectivity in adolescents with T1D [20].

In conclusion, we demonstrate intraglomerular hemodynamic dysfunction in youth with T1D of <10 years duration and identify strong associations between the tubular injury biomarkers YKL-40 and KIM-1 and both albuminuria and parameters of abnormal intraglomerular hemodynamic function. Future directions of this work include evaluations of the predictive capacity of YKL-40 and KIM-1 for future decline in kidney function and for assessments of YKL-40 and KIM-1 in the setting of nephroprotective agents including glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors in youth with T1D. Currently ongoing kidney biopsy studies will permit us to examine relationships between these circulating tubular injury biomarkers and intrarenal expression patterns of structural evidence of diabetic kidney injury.

Availability of data and material:

The datasets generated and analyzed during this study are available from the corresponding author upon reasonable request.