Urinary early kidney injury molecules in children with beta-thalassemia major

Abstract

Background: The aim of this study was to investigate novel urinary biomarkers including N-acetyl-β-d-glucosaminidase (NAG), neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and liver-type fatty acid binding protein (L-FABP) in children with β-thalassemia major (β-TM). Materials and methods: Totally, 52 patients (29 boys, 23 girls) with β-TM and 29 healthy controls (3–17 years) were included. Various demographic characteristics and blood transfusions/year, disease duration, and chelation therapy were recorded. Serum urea, creatinine, electrolytes, and ferritin and urinary creatinine, protein, calcium, phosphorus, sodium, potassium, and uric acid in first morning urine samples were measured and estimated glomerular filtration rate (eGFR) was calculated. Routine serum and urinary biochemical variables, urinary NAG to Creatinine (U_NAG/Cr), U_NGAL/Cr, U_KIM-1/Cr, and U_L-FABP/Cr ratios were determined. Results: Patients had similar mean serum urea, creatinine and eGFR levels compared with controls (p > 0.05 for all). The mean urinary protein to creatinine (U_Protein/Cr) ratio was significantly higher in patients compared to the healthy subjects (0.13 ± 0.09 mg/mg and 0.07 ± 0.04 mg/mg, respectively; p < 0.001). Significantly increased U_NAG/Cr (0.48 ± 0.58 vs. 0.23 ± 0.16, p = 0.026) and U_NGAL/Cr (22.1 ± 18.5 vs. 11.5 ± 6.17, p = 0.01) ratios were found in β-TM patients compared with healthy controls. However, no differences were found in serum and urinary electrolytes or U_KIM-1/Cr and U_L-FABP/Cr ratios between patients and controls (p > 0.05). Significant correlations were found between urinary biomarkers and urinary electrolytes (p < 0.05). Conclusions: Our results suggest that urinary NAG and NGAL may be considered to be reliable markers to monitor renal injury in β-TM patients.

• Early molecule
• FABP
• kidney injury
• KIM-1
• NAG
• NGAL
• Thalassemia

Introduction

β-Thalassemia is an inherited microcytic anemia that is characterized by mutations of the β-globin gene, leading to impairment in the biosynthesis of beta-globin chains.1 In β-thalassemia major (β-TM), regular transfusions of packed red blood cells are a mainstay of treatment but lead to iron overload.2 As a result of chronic transfusions, iron deposition occurs in many organs, such as the liver, heart, and endocrine glands.3

Due to the increase in patients’ lifespan by early diagnosis, blood transfusions, and iron chelation therapy, previously unobserved associated disorders have been encountered in β-TM.1 Although major system dysfunctions, including cardiopulmonary and reticuloendothelial systems, have been reported in patients with TM, little information is available on renal involvement.4

In recent years, some studies have investigated renal proximal tubular damage using biomarkers including β2 microglobulin, N-acetyl-d-glucosaminidase (NAG), proteinuria, and aminoaciduria in thalassemia patients.5^,6 Moreover, a number of studies showed that patients with TM had increased urinary excretion of calcium, phosphate, magnesium, and uric acid.7 Even less is known about the effects of thalassemia on the changes of estimated glomerular filtration rate (eGFR).8

Shortened red cell life span, rapid iron turnover, iron overload, and anemia are likely to be the main factors responsible for renal abnormalities in thalassemia. In addition, it has been reported that chelation therapy can potentially lead to nephrotoxicity via an unknown mechanism.9 Therefore, monitoring of renal functions in transfusion-dependent thalassemic patients receiving chelation therapy has been recommended.1

The detection of renal damage has been made by traditional biomarkers such as serum creatinine and blood urea nitrogen (BUN), and this has remained unaltered for a few decades. However, these biomarkers have several shortcomings as early and sensitive diagnostic markers of kidney injury.10 In this regard, novel biomarkers are needed to allow detection of early renal damage. Over the past few years, with the use of genomic and proteomic instruments, some novel early kidney injury biomarkers have been recognized and proposed.11 The most promising biomarkers are urine N-acetyl-b-d-glucosaminidase (NAG), neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and liver-type fatty acid binding protein (L-FABP).12

NAG is a lysosomal enzyme found in the renal tubules, and high urinary NAG activity has been shown during active renal disease. Thus, NAG can serve as an exclusive urinary marker for damaged tubular cells.13 NGAL is also expressed by the proximal tubule epithelial cells, and urinary NGAL levels have been proposed as a beneficial early predictor of renal injury.14 KIM-1 is a trans-membrane glycoprotein that is expressed at high levels in de-differentiated proximal tubule epithelial cells in kidneys after toxic or ischemic injury.14 L-FABPs are small cytoplasmic proteins that bind selectively to free fatty acid (FFA). Recent studies performed in children have shown that L-FABP is a sensitive biomarker in ischemic renal injury.15

Few data are available regarding renal functional abnormalities in children with thalassemia.2 There are very few studies that investigate renal dysfunction in patients with TM.1^,16 Moreover, we could not find any studies in the literature on urinary molecules indicative of early kidney injury, including NAG, NGAL, KIM-1, and L-FABP, in children with TM. Early identification of kidney injury may enable particular interventions to reduce renal damage. Therefore, we aimed to investigate renal functions using both traditional and early biomarkers of tubular and glomerular dysfunction. To the best of our knowledge, this study is the first cross-sectional study investigating renal function using novel biomarkers such as NAG, NGAL, KIM-1, and L-FABP in thalassemia.

Patients and methods

Subjects and demographics

A total of 52 patients (29 boys, 23 girls) with β-thalassemia major aged between 3 and 17 years that were followed at the Hematology Clinics of Dicle University Hospital were included. The patients were recruited in a cross-sectional manner between November 2013 and January 2014. A total of 29 age- and gender-matched healthy controls (16 boys, 13 girls) were recruited as a control group. The control group consisted of children without any potential diseases affecting renal function, which were undergoing routine checkups or pre-surgical examinations for elective minor surgery such as inguinal hernia or circumcision.

We assessed the following data at the time of the Pediatric Hematology Outpatient Clinic visits: patient’s age, gender, height, weight, number of blood transfusions/year, disease duration, history of splenectomy, and chelation therapy.

The diagnosis of β-TM was made based on the results of a complete blood count and hemoglobin electrophoresis. Patients were under regular follow-up and were selected for assessment of renal function on the basis of conventional and novel urinary biomarkers. In addition, all children were in a stable stage of their β-TM disease course with regular packed red blood cell transfusion every 3–4 weeks. A detailed medical history and clinical examination were performed on all patients. The patients were investigated for the existence of diseases other than β-TM on presentation to the Pediatric Hematology outpatient clinics. Anthropometric measurements including weight and height were performed and recorded. Patients with concomitant diseases that could affect kidney functions or potentially lead to renal damage, such as diabetes, rheumatic heart disease, thyroid disease, hepatic diseases, sepsis, or consumption of nephrotoxic drugs were excluded. Patients with a history suggestive of recurrent urinary tract infections, a family history of hereditary renal diseases, any active infection, renal stones, hydronephrosis, urinary reflux, or intake of corticosteroid, trimethoprim or cephalosporin in the past week were also excluded.

All patients received blood transfusions and were treated with chelation therapy using deferasirox with a dose of 20–40 mg/kg/day when their ferritin levels were above 1000 ng/mL. Splenectomy in children over the age of 5 years was done when blood transfusion requirements increased to 200–250 mL/kg/year to achieve target hemoglobin or in the presence of signs of hypersplenism.

Biochemical analyses

Patients were advised not to take any medications for 24 h prior to blood sampling. Blood samples were collected from each patient, preferably after an overnight fast, to measure complete blood counts (CBC), serum biochemical analyses such as serum urea, creatinine (SCr), electrolytes, and ferritin. Fresh first morning urine samples were centrifuged then immediately divided into five different aliquots in Eppendorf tubes and stored at −80 °C until analysis for creatinine, total protein, calcium (Ca), phosphorus (P), sodium (Na), potassium (K), uric acid, and early urine markers including NAG, NGAL, KIM-1, and L-FABP.

Serum ferritin levels and serum and urinary Ca, P, Na, K, protein, and uric acid levels were measured by an enzymatic colorimetric method performed using an Abbott ARCHITECT C16000 (Abbott Park, IL) instrument in the Biochemistry Laboratory.

Estimated glomerular filtration rate (eGFR) was calculated according to the modified Schwartz formula for children: eGFR (mL/min/1.73 m²) = height (cm) × 0.413/serum creatinine (mg/dL). The eGFR was considered to be normal if the value was ≥90 mL/min/1.73 m², while an eGFR of <90 mL/min/1.73 m² was considered to be decreased GFR.17 For each patient, serum Cr was matched with reference values for age and sex. All biochemical tests were performed in the same laboratory.

Measurement of urinary biomarkers

Urine levels of NAG, NGAL, KIM-1, and L-FABP were determined using a commercially available quantitative sandwich immunoassay technique (SunRed Biotechnology Company, Shanghai, China), as per manufacturers’ instructions. All urinary biomarker levels were normalized by dividing by urine creatinine.

Our study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Dicle University Medical School. Written informed consent was obtained from each participant and/or from his or her legal caregivers prior to study entry.

Statistical analysis

All statistical analyses were performed using SPSS (Statistical Package for Social Sciences) version 15.0 for Windows (SPSS, Inc., Chicago, IL). Numerical variables were expressed as mean ± standard deviation, median, and maximum–minimum values. Categorical variables were expressed as counts and percentages. The Kolmogorov–Smirnov and Shapiro–Wilk tests were performed for all continuous variables to check the distribution of normality. Independent samples t-test or Mann–Whitney U-test were used for comparison of numerical data of patients and the controls. Chi-square test was used for comparison of nominal or ordinal variables. Pearson or Spearman correlation analyses were used to examine relationships between numerical variables. A p value of less than 0.05 was accepted as statistically significant.

Results

A total of 52 (29 boys, 23 girls) children with β-TM and 29 (16 boys, 13 girls) age- and gender-matched healthy control subjects were enrolled in this study between November 2013 and January 2014. The mean age of the thalassemic children was 9.1 ± 4.4 years and that of the control subjects was 8.8 ± 4.0 years (range, 3–17 years; p > 0.05). No significant differences were found between the patients and controls in terms of age, gender distribution, and anthropometric measurements (p > 0.05; Table 1). The demographic and biochemical characteristics of the patients and the healthy subjects are listed in Table 1. Serum sodium, potassium, calcium, and phosphate of patients and controls were within normal limits and no significant difference was found between patients and the controls (p > 0.05 for all; Table 1). Serum urea and creatinine levels of patients were within normal limits and no significant differences were found in serum urea and creatinine levels between children with β-TM and control subjects (p > 0.05; Table 1). No significant difference was found in eGFR values between children with β-TM and controls (p > 0.05; Table 1).

Table 1. Demographic and biochemical characteristics of the patients and the control group.

	Thalassemia Group (n = 52)	Controls (n = 29)	p
Age, years	9.1 ± 4.4 (8.0; 3.0–17.0)	8.8 ± 4.0 (8.0; 3.0–17.0)	NS
Height, cm	124.2 ± 22.9 (128.0; 75.0–159.0)	125.5 ± 20.3 (122.0; 92.0–159.0)	NS
Weight, kg	30.3 ± 12.2 (28.0; 10.0–60.0)	29.6 ± 13.5 (25.0; 13.0–56.0)	NS
Serum glucose (mg/dL)	93.5 ± 13.4 (92.5; 64.0–134.0)	95.2 ± 10.5 (95.0; 76.0–110.0)	NS
Serum urea (mg/dL)	26.9 ± 7.2 (26.5; 8.0–40.0)	27.3 ± 7.9 (21.0; 8.0–36.0)	NS
Serum creatinine (mg/dL)	0.47 ± 0.10 (0.46; 0.30–0.80)	0.50 ± 0.08 (0.50; 0.38–0.70)	NS
eGFR (ml/min/1.73 m^2)	121 ± 29 (119; 71–189)	116 ± 18 (120; 92–157)	NS
Serum sodium (mEq/L)	137.3 ± 2.5 (137.0; 131.0–143.0)	137.7 ± 2.1 (138.0; 134.0–142.0)	NS
Serum potassium (mEq/L)	4.2 ± 0.28 (4.2; 3.6–5.1)	4.1 ± 0.42 (4.0; 3.5–5.1)	NS
Serum calcium (mg/dL)	9.3 ± 0.5 (9.4; 8.2–10.2)	9.4 ± 0.62 (9.4; 8.5–10.5)	NS
Serum phosphate (mg/dL)	3.5 ± 0.6 (3.5; 2.5–4.9)	3.4 ± 0.5 (3.5; 2.5–4.3)	NS

Notes: Data presented are [mean, (median; minimum-maximum)]; eGFR, estimated glomerular filtration rate; NS: Not significant.

Significantly lower hemoglobin, hematocrit, and red blood cell count and higher ferritin levels were found in children with β-TM compared with healthy controls (p < 0.001 for all; Table 2). High serum ferritin levels were found in all of our patients, and 50 (96.2%) had anemia based on normal reference ranges for hemoglobin.18 The mean level of serum ferritin was 2552 ± 1413 (minimum 859, maximum 6826) ng/mL, and 49 (94.2%) patients had ferritin levels greater than 1000 ng/mL.

Table 2. Hematological characteristics of β-TM children and healthy controls [mean, (median; minimum–maximum)].

	Thalassemia group (n = 52)	Control group (n = 29)	p
Hemoglobin (g/100 mL)	8.2 ± 1.5 (8.5; 5.1–11.5)	11.8 ± 1.1 (11.5; 10.7–14.1)	<0.001
Hematocrit (%)	23.7 ± 4.7 (23.5; 12.9–33.6)	36.7 ± 3.3 (36.1; 30.8–44.1)	<0.001
RBC (×10^6 cells/µL)	3.19 ± 0.76 (3.12; 1.65–5.20)	4.79 ± 0.40 (4.78; 4.12–5.36)	<0.001
Platelet count (×10^3 cells/µL)	390 ± 208 (349; 115–1235)	308 ± 95 (346; 149–429)	NS
Ferritin (ng/mL)	2552 ± 1413 (1988; 859–6826)	78 ± 33 (82; 19–143)	<0.001

Notes: β-TM, β-thalassemia major; RBC: Red blood count; NS: Not significant.

Table 3 summarizes the urinary electrolytes, uric acid, and protein to creatinine ratios of the patients and controls. No significant differences were found between patients and controls in terms of urinary sodium-, potassium-, calcium-, phosphorus-, magnesium-, and uric acid-to-creatinine ratios (p > 0.05 for all). However, a significantly higher protein-to-creatinine ratio was found in patients with β-TM compared with the control subjects (0.13 ± 0.09 and 0.07 ± 0.04, respectively; p < 0.001; Table 3).

Table 3. Urinary electrolytes, uric acid, and protein–creatinine ratios of the patients and the controls [mean, (median; minimum–maximum)].

	Thalassemia (n = 52)	Controls (n = 29)	p
UNa/Cr	2.28 ± 1.85 (2.13; 0.27–11.88)	1.97 ± 1.08 (1.78; 0.17–4.51)	NS
UK/Cr	1.11 ± 0.72 (0.97; 0.07–4.00)	0.94 ± 0.58 (0.84; 0.30–3.04)	NS
UCa/Cr	0.12 ± 0.04 (0.12; 0.02–0.42)	0.11 ± 0.09 (0.06; 0.02–0.21)	NS
UP/Cr	0.58 ± 0.36 (0.46; 0.12–1.42)	0.69 ± 0.41 (0.63; 0.12–1.77)	NS
UMg/Cr	0.13 ± 0.04 (0.12; 0.05–0.25)	0.13 ± 0.06 (0.12; 0.04–0.25)	NS
UProtein/Cr	0.13 ± 0.09 (0.10; 0.04–0.47)	0.07 ± 0.04 (0.06; 0.03–0.20)	<0.001
UUric acid/Cr	0.55 ± 0.29 (0.50; 0.10–1.42)	0.44 ± 0.29 (0.32; 0.13–1.12)	NS

Notes: U, Urinary; Cr, Creatinine; Na, Sodium; K, Potassium; Ca, Calcium; P, Phosphorus; Mg, Magnesium; NS, Not significant.

Patients with β-TM had a significantly elevated urinary NAG-to-creatinine ratio (U_NAG/Cr) compared with controls (0.48 ± 0.58 vs. 0.23 ± 0.16, p = 0.026; Table 4). In addition, a significantly higher mean urinary NGAL-to-creatinine ratio (U_NGAL/Cr) was found in patients with β-TM compared with the control group (22.1 ± 18.5 vs. 11.5 ± 6.17, p = 0.01). However, no significant differences were found in urinary KIM-1-to-creatinine (U_KIM-1/Cr) and urinary L-FABP-to-creatinine (U_L-FABP/Cr) ratios between the patient and the control groups (p = 0.795 and p = 0.622, respectively; Table 4).

Table 4. Urinary novel early biomarkers–creatinine ratios of the β-thalassemia major and the control groups [mean, (median; minimum–maximum)].

	Thalassemia group (n = 52)	Controls (n = 29)	p
UNAG/Cr	0.48 ± 0.58 (0.28; 0.09–2.85)	0.23 ± 0.16 (0.19; 0.04–0.61)	0.026
UNGAL/Cr	22.1 ± 18.5 (14.4; 5.8–91.0)	11.5 ± 6.17 (10.7; 0.92–22.8)	0.010
UKİM1/Cr	0.062 ± 0.049 (0.045; 0.01–0.23)	0.044 ± 0.018 (0.036; 0.02–0.08)	NS
UL-FABP/Cr	0.133 ± 0.105 (0.102; 0.030–0.460)	0.126 ± 0.080 (0.119; 0.030–0.350)	NS

Notes: U_NAG/Cr, urinary N-acetyl-beta-d-glucosaminidase to creatinine; U_NGAL/Cr, urinary neutrophil gelatinase associated lipocalin to creatinine; U_KIM-1/Cr, urinary kidney injury molecule-1 to creatinine; U_L-FABP/Cr, urinary liver-type fatty acid binding protein to creatinine; NS, Not significant.

Significant positive correlations were found between U_NAG/Cr, U_NGAL/Cr, U_KIM-1/Cr, and U_L-FABP/Cr (p < 0.001 for each; Table 5). In addition, the urinary calcium-to-creatinine ratio had significant positive correlations with urinary biomarkers-to-creatinine ratios (U_NAG/Cr, U_NGAL/Cr, U_KIM-1/Cr, and U_L-FABP/Cr; p < 0.001, for each). Urinary sodium-to-creatinine and uric acid-to-creatinine ratios had significant weak positive correlations with urinary biomarkers (p < 0.05; Table 5). Urinary potassium-to-creatinine ratios (U_K/Cr) had weak positive correlations with U_NGAL/Cr and U_KIM-1/Cr, however no significant correlations were found between U_K/Cr and U_NAG/Cr and U_L-FABP/Cr (p > 0.05; Table 5).

Table 5. Spearman’s correlation coefficients in β-thalassemia major group of urinary novel early biomarkers with other variables.

Variables		UNAG/Cr	UNGAL/Cr	UKIM-1/Cr	UL-FABP/Cr
UNAG/Cr	r		0.772	0.809	0.707
	p		<0.001	<0.001	<0.001
UNGAL/Cr	r			0.882	0.716
	p			<0.001	<0.001
UKIM-1/Cr	r				0.751
	p				<0.001
UK/Cr	r	NS	0.316	0.318	NS
	p		0.039	0.031
UCa/Cr	r	0.606	0.558	0.612	0.528
	p	<0.001	<0.001	<0.001	<0.001
UNa/Cr	r	0.409	0.338	0.434	0.364
	p	0.008	0.027	0.003	0.018
UUric acid/Cr	r	0.412	0.423	0.418	0.310
	p	0.007	0.005	0.004	0.046

Notes: U_NAG/Cr, urinary N-acetyl-beta-d-glucosaminidase to creatinine; U_NGAL/Cr, urinary neutrophil gelatinase associated lipocalin to creatinine; U_KIM-1/Cr; urinary kidney injury molecule-1 to creatinine; U_L-FABP/Cr, urinary liver-type fatty acid binding protein to creatinine; NS, Not significant.

There were no significant relationships between eGFR and urinary biomarkers, urinary electrolyte-to-creatinine, urinary protein-to-creatinine, and urinary uric acid-to-creatinine ratios (data not shown, p > 0.05).

No correlations were found between urinary markers-to-creatinine ratios (U_NAG/Cr, U_NGAL/Cr, U_KIM-1/Cr, and U_L-FABP/Cr) and patient’s age at diagnosis, duration of illness, first transfusion age, and transfusion frequency in the patient group (data not shown, p > 0.05). There were also no correlations of urinary markers-to-creatinine ratios with hematological variables including hemoglobin concentration, hematocrit, red blood cell count, platelet counts, and ferritin (data not shown, p > 0.05).

Discussion

Thalassemia is a chronic blood disease that affects many organs in various ways. The long-term survival of children with thalassemia has improved dramatically with respect to both life expectancy and quality of life.9 This is partly attributable to improved treatment regimens, including regular blood transfusions and iron chelation therapy. However, though the quality of life in patients with β-TM has improved significantly, this success has permitted the emergence of previously unrecognized complications and morbidities of the disease. While functional abnormalities have been well-studied in many systems of patients with TM, including cardiac, endocrine, pulmonary and hepatic disorders, the renal effects, and manifestations of the disease have been poorly investigated.19

Some studies on renal involvement in children with β-TM demonstrated GFR abnormalities and varying degrees of renal tubular dysfunction.16^,19^,20 Chronic anemia, increased iron deposition and possible toxicity of specific iron chelators have been proposed as underlying etiologies of renal involvement in patients with β-TM.19

Biomarkers are measurable biological entities that allow differentiation between injury and normal function. Some novel urine biomarkers have been tested for early detection of acute kidney injury. These biomarkers are produced and released by tubular cells as measurable proteins in a variety of renal injuries. Renal tubular injury can be reflected by upregulation of NAG, NGAL, KIM-1, L-FABP, and IL-8.21 Increased levels of these markers can be indicative of subclinical renal injury, and these novel markers may allow early assessment of kidney injury and prognosis. Assessments of kidney functions in β-TM patients are mainly done in clinical settings, since renal abnormalities can be associated with many short- and long-term effects. Early renal damage recognition may be of critical importance for optimum care to ameliorate patient outcomes. Therefore, defining an early and reliable biomarker of kidney involvement in thalassemia is very important.

In this study, we aimed to explore these new generation urinary biomarkers in β-TM with spot urine samples. Our study group was similar to the control group in the male-to-female ratio, mean age, and anthropometric characteristics. Thalassemia patients had normal values of serum sodium, potassium, calcium, and phosphate, which can all be affected by renal dysfunction, and no significant differences were found between patients and controls. Although no significant differences were found in serum urea, creatinine and eGFR values, the thalassemia group had a significantly higher urine protein-to-creatinine ratio compared to control subjects (Tables 1 and 3). Aldudak et al. found that serum levels of potassium, phosphorus, uric acid, and U_Protein/Cr levels were higher in β-thalassanemia major patients than in normal subjects.22 Several authors also reported proteinuria and albuminuria in thalassanemia patients. Our finding of higher ratios of U_Protein/Cr in patients is in accordance with these prior reports.4^,5^,22^,23 This proteinuria might be explained primarily by the impairment of proximal tubular reabsorption, which may be due to severe iron overload in the tissues where it also stimulates the production of reactive oxygen radicals and results in cellular injury.24 As urinary protein/Cr ratio is 0.13 in our study, this degree of proteinuria is consistent with proximal tubular injury. It is well-known that significant glomerular proteinuria indicates protein/Cr > 0.2. Increased urinary levels of NAG and NGAL, together with low proteinuria suggested proximal tubular injury in β-TM. Previous studies also presented strong clinical evidence consistent with proximal tubular injury in β-TM.7^,22

As expected, patients with anemia of β-TM had significantly lower hemoglobin, hematocrit and red blood cell counts and higher serum ferritin levels compared with the control subjects (Table 2).

A recent study showed that Cr levels were higher in children with β-TM when compared to controls and showed impaired eGFR in these patients.25 Mohkam et al. also reported high serum Cr in thalassemia patients.23 In contrast, several authors found normal serum Cr in β-TM patients.22 Some researchers demonstrated decreased eGFR levels in β-TM patients.8^,20 They hypothesized that large variations in hemoglobin levels were also associated with an increased risk of decline in eGFR. We suggest that despite existence of severe anemia combined with iron deposition in β-TM patients, renal injury did not reach clinically detectable abnormalities in serum urea and creatinine or eGFR in our patients.

We found significant positive correlations between U_NAG/Cr, U_NGAL/Cr, U_KIM-1/Cr, and U_L-FABP/Cr levels in β-TM patients (Table 5). These correlations may indicate the reliability of urinary biomarker measurements and parallel changes of all biomarkers in β-TM. In addition, significant positive correlations between biomarkers and U_Na/Cr, U_K/Cr, U_Ca/Cr, and U_Uricacid/Cr are consistent with the fact that both urinary biomarkers and urinary solutes represent renal tubular functions.

In this study, significantly higher U_NAG/Cr and U_NGAL/Cr were found in children with β-TM compared with the matched healthy control subjects. However, no significant differences were found in U_KIM-1/Cr and U_LFABP/Cr between the study and control subjects (Table 4).

NAG is predominantly a biomarker of proximal tubular damage, and increase in its level reflects tubular injury.13 Several authors have reported increased urinary excretion of markers of tubular injury, such as NAG, malondialdehyde, and β₂-microglobulin, in patients with TM.22^,23 A recent case–control study demonstrated a significant increase in urinary NAG in β-TM patients in comparison with controls, with most cases having high NAG levels.26 In this study, significantly elevated U_NAG/Cr and U_NGAL/Cr levels in patients indicated renal proximal tubular damage. A probable mechanism for this outcome may be that thalassemia itself leads to proximal tubular dysfunction either through chronic hypoxia from persistent anemia, through iron deposition or iron chelation. We think that the lack of correlation between urinary markers and Hb, Hct and ferritin levels may be related to very early phase of tubular dysfunction so that elevation was taken place only in NAG and NGAL, but not in KIM-1 and L-FABP or urinary electrolytes.

NGAL is a 25-kDa lipocalin iron-carrying protein secreted by activated neutrophils and expressed in epithelial cells such as those in the proximal tubule, distal tubule, and loop of Henle segments. It is upregulated in renal tubular injury. NGAL excretion in urine occurs when there is proximal tubular injury that disrupts NGAL reabsorption or increases NGAL synthesis. NGAL rises dramatically with acute kidney injury (AKI).27 However, NGAL is also elevated in non-AKI patients such as lupus nephritis, IgA nephropathy, and urinary tract infection.28 Nishida et al. reported significant correlations between the levels of urinary NGAL and the degree of proteinuria in pediatric patients with chronic renal disease from diverse etiologies.29 In the current study, U_NGAL/Cr levels were also shown to be increased in our patients. Urinary NGAL offers potential as a biomarker in screening for renal dysfunction and hence to determine patients who are likely to experience impairment of renal function in the future.

In this cross-sectional study, we found no difference in U_KIM-1 levels of patients with β-TM and the control group. Previous reports on increased U_KIM-1 levels are probably related to KIM-1 expression in kidney disease, suggesting that U_KIM-1 is a very promising biomarker for the presence of tubulointerstitial damage. Many studies on renal injury have shown the feasibility of this marker in predicting damage due to acute kidney injury.30 The chronic nature of kidney damage in β-TM, likely without renal interstitial damage, may explain the normal U_KIM-1 values of our patients.

Urinary L-FABP is a urinary tubular biomarker associated with functional and structural renal injury. Urinary levels of L-FABP are not affected by its serum levels because urinary L-FABP originates mainly from the tubular cells. The utility of urinary L-FABP in various chronic kidney disease (CKDs) has been reported.31 A number of studies have suggested a possible role for urinary FABP in clinical diagnosis of renal disease.32 Ferguson et al. have reported a substantial increment of urinary L-FABP in several etiologies of acute kidney injury (AKI) such as acute tubular necrosis, nephrotoxic exposure, and sepsis.33 In contrast to previous studies, we did not detect high U_L-FABP/Cr levels in our patients. Normal urinary KIM-1 and L-FABP levels in our children with β-TM may also be secondary to the beneficial effects of deferasirox. Perhaps L-FABP, like creatinine, may not be a significant instrument for analyzing renal dysfunction in β-TM. Further investigation in various chronic clinical pediatric settings should help to approve the potential of U_L-FABP for clinical practice.

We found weak positive significant correlations between the biomarkers themselves and between biomarkers and the levels of U_Na/Cr, U_K/Cr, U_Ca/Cr, and U_Uricacid/Cr (Table 5). Although, all urinary biomarkers (NAG, NGAL, KIM-1, and L-FABP) levels and most urine electrolytes were found higher in β-TM group compared with the healthy controls, these differences reached to significant levels only in NAG and NGAL (Table 4). Therefore, we think the difference being significant or not, elevated levels of biomarkers and electrolytes in the patient group are responsible from positive correlations of urinary biomarkers and electrolytes. Early biomarkers are very sensitive proteins, therefore increased urinary NAG and NGAL, as signalers of proximal renal injury, were found in our β-TM patients. We think that significant correlations between biomarkers and urinary electrolytes may be an indication of subclinical mild tubular dysfunction in our patients. Therefore, before reaching to significantly increased levels, KIM-1, L-FABP, and urinary electrolytes were found to be correlated with each other. The efficacy of chelation therapy on kidney function is remarkable. Oral deferasirox has become a routine therapy for the treatment of iron overload in β-TM patients. Several adverse effects on renal function have been shown with increasing frequency.34 The mechanism of renal injury associated with chelation is not exactly known. Kidney injury with chelators is unexplained but recognized. Several mechanisms were proposed for deferasirox-induced tubular dysfunction. In a recent study, the toxic effect of deferasirox on tubular epithelial cells was suggested as a cause of kidney damage in β-TM patients.25 Brosnahan et al. suggested that a drug hypersensitivity reaction might lead to kidney damage in patients with in β-TM.35 In contrast, safety data with deferasirox in patients with β-TM have now been reported for up to 5 years of treatment and confirm the absence of progressive increases in serum creatinine over longer-term treatment.36

It is believed that anemia, associated potential chronic hypoxia, iron overload, and chelation therapy may be important key factors in the development of kidney dysfunction. Our results indicate that glomerular and tubular dysfunctions exist in children with β-TM. These abnormalities are mainly sub-clinical, but, with protracted recurrent tubule injury, progression of renal damage may occur.

A potential limitation of our study was that it was a cross-sectional analysis, so we did not follow individual subjects over time. In addition, we did not obtain a “gold standard” measurement of eGFR, including the clearance of iohexol or inulin. However, this is a preliminary study, which provided information regarding renal dysfunction in children with β-TM by evaluating early urinary biomarkers.

Conclusion

In conclusion, our findings show a diverse biomarker type in β-TM, identifying urinary NAG and NGAL, but not urinary KIM-1 and L_FABP, as two potential biomarkers of renal dysfunction. It is improbable that a sole marker will satisfy the necessity of predicting renal disease. It is more likely that a panel of non-invasive urine biomarkers to define renal damage may be a better strategy than using a sole one. Further longitudinal, prospective analyses of novel urinary biomarkers for their potential role as monitoring instruments for renal dysfunction in β-TM remain to be performed.

Acknowledgments

The authors wish to thank all the β-thalassemia major patients and the controls that took part in this study.

Declaration of interest

Authors thank to DUBAP (Dicle University Fund of Scientific Research Projects) for scientific and financial support.