Prevalence and risk factors for albuminuria and glomerular hyperfiltration in a large cohort of children with sickle cell anemia

Abstract

Sickle cell anemia (SCA) considerably modifies kidney structure and function, and the group of kidney abnormalities are called sickle cell nephropathy (SCN). Hyperfiltration is one of the most commonly recognized cause of kidney injury in children with SCA and occurs in up to 76% of the pediatric SCA population.1 Albuminuria, indicating early glomerular damage, affects 15.8%-28.8% of children with SCA, and contributes to the development and progression of CKD.2 Few studies with pediatric patients have shown that some laboratory parameters and genetic variants might influence GFR or albuminuria in individuals with SCA. However, contradictory results have been reported. Therefore, there is a clear need for large studies to accurately determine the prevalence and risk factors for the development of pediatric SCN. In this study, we investigated the prevalence and risk factors for albuminuria and hyperfiltration in a large cohort of SCA children from the state of Minas Gerais, Brazil. The detailed methods of the manuscript were provided in online supporting information (see Supporting Information File S1). The study population comprised 555 HbSS children. The protocol was approved by institutional review boards and informed written consents were obtained. Medical records were reviewed to obtain childrensʼ demographics, laboratory, clinical and therapeutic data. Random spot urine and blood specimens were collected at the time of enrollment in the study. The HBA deletions were evaluated by gap-PCR and rs743811 polymorphism (nearby HMOX1) and rs4646994 Insertion/Deletion (ACE) by qPCR. Normoalbuminuria, and albuminuria were defined as Urine Albumin-to-Creatinine Ratio (UACR) < 30 mg/g, ≥ 30 mg/g, respectively. The estimated glomerular filtration rate (eGFR) was calculated using the updated pediatric Schwartz formula, and we considered glomerular hyperfiltration when eGFR ≥140 mL/min/1.73m2. Continuous variables were expressed as mean and standard deviation (SD) or median and interquartile range as appropriated. Categorical variables were expressed as percentages of total. Cumulative risk of albuminuria was estimated by using a Kaplan-Meier method (function [1−survival]) and the log rank test was used to compare different subgroups based on risk factors. Coxʼs regression was used to determine the independent effect of each risk factor on albuminuria and hyperfiltration. The demographic, clinical, laboratory, and genetic characteristics of the study population are described in supporting information (Table S1). The study cohort consisted of 555 children (283 [51%] male) ranging in age from 1.6 to 20.9 years. One hundred nine children (19.6%, 95%CI: 16.3%-22.9%) had albuminuria, and the cumulative risk of albuminuria by 20 years of age was 88.7% (95% CI, 82.7%-94.7%; Figure 1A). Prevalence of albuminuria increased with aging (P < .001; Figure 1B). There was no difference in prevalence of albuminuria according to treatment and gender (data not shown). The final Cox multivariate model for albuminuria retained five variables (Table S2). Platelet count (HR = 1.002, 95% confidence interval [CI]: 1.001-1.004; P = .007), WBC counts (HR = 1.062, 95% CI: 1.003-1.126; P = .04) and eGFR (HR = 1.007, 95% CI: 1.003-1.011; P < .001) were statistically significant risk factors for albuminuria. Among genetic factors, the risk for albuminuria significantly reduced in children with coexisting of alpha-thalassemia (HR = 0.59, 95% CI: 0.36-0.99; P = .047) and in those with at least one copy of the C allele of HMOX1 rs743811 (HR = 0.51, 95% CI: 0.33-0.78; P = .002). Heart rate, respiratory rate, systolic blood pressure, and diastolic blood pressure were significantly associated with albuminuria in the bivariate analysis, but were not included in the Cox multivariate analysis. The cumulative risk of albuminuria was significantly lower in children with HBA deletion (P = .01; Figure 1C), and lower for children carrying TC or CC genotype (rs743811) (P = .013; Figure 1D). The prevalence of hyperfiltration was 25.4% (95% CI: 21.7%-29%; n = 137/540). Hyperfiltration was present in 39.3% (n = 42/107) and 21.9% (n = 95/433) of children with and without albuminuria (P < .001), respectively. The final Cox multivariate model for hyperfiltration (Table S2) included Hb F (HR = 0.96, 95% CI: 0.93-0.99; P = .022) and alpha-thalassemia (HR = 0.45, 95% CI: 0.26-0.79; P = .006). We explored the effect of risk factors on eGFR as a continuous variable, and the final linear regression multivariate model also included Hb F and alpha-thalassemia (Table S3). This is the largest study to date that evaluated the prevalence and risk factors for albuminuria and glomerular hyperfiltration in children with SCA, and the first conducted in a Brazilian population. We confirm previous reports of a reduction in GFR and protection from albuminuria in individuals with coexisting alpha-thalassemia.3 However, this is the first study to provide an association between alpha-thalassemia and renal function changes in children with SCA, suggesting that HBA deletions may play a role in early stages of kidney disease physiopathology in individuals with SCA. Alpha-thalassemia is associated with improved rheological features of red blood cells and reduced hemolysis, which, in turn, might decrease infarction and hemolysis-related glomerular damage and, by so doing, might preserve glomerular function. The rs743811 SNP is located nearby HMOX1 gene and has been significantly associated with chronic kidney disease (CKD) and end-stagerenal disease (ESRD) in individuals with SCD.4 In contrast with the results of Saraf et al.4 for whom rs743811 C allele was associated with an increased risk of CKD and ESRD, our findings showed instead that rs743811 C allele was associated with a decreased risk of albuminuria. Therefore, the association of SNP rs743811 with kidney disease in individuals with SCA was not consistent between populations. The reason for this effect in different directions is unclear. The functional consequence of the SNP rs743811, located in a noncoding region, has not been established yet. Saraf et al.4 did not observe an association of the rs743811 SNP with transcript levels of HMOX1 in peripheral blood mononuclear cells. We performed in silico functional annotation of the rs743811 SNP in order to assess its potential regulatory role on modulating tissue-specific gene expression (Figure S1). The rs743811 SNP is a splicing Quantitative Trait Loci for the MCM5 gene in the whole blood and spleen. The allele C is associated with lower Intron-Excision ratio, which might lead to reduction on MCM5 expression. The MCM proteins are crucial to DNA replication, and MCM5 overexpression has been associated with various types of cancer, including renal cell carcinoma proliferation, progression and prognosis.5 Although elucidation and interpretation of the biological basis for the association between rs743811 SNP and kidney outcomes in SCA remains challenging, our findings provide insight into genes and pathways that may have a role in the pathophysiology of SCN. One novel finding derived from this study was that platelet count was significantly associated with the occurrence of albuminuria, and HbF levels was associated with eGFR. To our knowledge, this is the first study to show these associations, which may have a role in the pathophysiology of SCN in children with SCA. SCA is characterized by a hypercoagulable state with chronic activation of coagulation cascade that causes tissue inflammation and organ damage, including in the kidney. Platelets promote adhesion of red blood cells and WBC to endothelium, contributing to endothelial dysfunction and vaso-occlusion. Platelets from SCD patients can also secrete higher amounts of inflammatory molecules than controls, including interleukin 6. Note, IL-6 is a cytokine with a pivotal role in the pathophysiology of albuminuria in children with SCA.6 Novel risk factors may improve the ability to predict the likelihood of CKD and ESRD in the future. Individuals at highest risk would substantially benefit from more intensive monitoring and preventive therapies. The results from our multivariate models showed that alpha-thalassemia was significantly associated with eGFR and hyperfiltration, whereas an inverse association was found between HbF levels and eGFR and hyperfiltration. Hemoglobin F is the major modulator of the clinical phenotype of SCD. HU therapy can theoretically increase levels of HbF. Our data suggests that HU therapy may benefit SCA children with glomerular hyperfiltration, reducing the progressive renal dysfunction. Our data showed associations between higher heart and respiratory rates and the presence of albuminuria in children with SCA. These data may suggest that early changes to renal function may directly affect the function of other organs, including heart and lung. Thus, treatment of children with albuminuria might reduce the risk of severe life-threatening clinical complications and, also, mortality. Clinical trials with pediatric patients are needed to confirm and extend these findings. In summary, the prevalence of albuminuria and hyperfiltration in children with SCA in this large cohort study was 19.6% and 25.4%, respectively. Age, laboratory features, and coexistence of alpha-thalassemia significantly affected occurrence of albuminuria and hyperfiltration. These features may contribute to early identification of patients at high-risk and to improve screening strategies, leading to preventive and therapeutic interventions. The authors acknowledge all subjects and parents for their cooperation in the study. This work was supported by a grant from the “Fundação de Amparo à Pesquisa de Minas Gerais” (FAPEMIG; grant # APQ-00131-17, APQ-04261-17 and BIP-00005-18). The authors declare no conflict of interest. A.R.B. designed the study, generally supervised the data and wrote the first draft of the manuscript; J.A.A., F.G.M., D.M.M.S., and W.P. performed experimental assays, helped in data collection and sample preparation. P.V.R., designed the study, collected the biological samples and clinical data; C.M.S. and A.C.B. collected the biological samples and clinical data; R.R.S. performed in silico functional annotation and took part in the statistical analysis. M.B.V. designed the study and revised the manuscript; A.C.S.S. designed and coordinated the study and made the final revision of the manuscript. All authors approved the final version of the manuscript and the submission to American Journal of Hematology. Appendix S1. Supporting Information. Figure S1. Violin plots from rs743811 SNP reported as splicing Quantitative Trait Loci for MCM5 in whole blood (A) and spleen (B) human samples from GTEx portal. Each copy of the C allele decreases the Normalized Intron-Excision Ratio of MCM5 in whole blood (P-value = 4.1e−18; Normalized Effect Size = −0.48) and spleen (P-value = 8.3e−7; Normalized Effect Size = −0.48). The blue parenthesis in X-axis correspond to number of sample size. Table S1. Main characteristics of the cohort of children with sickle cell anemia. Table S2. Risk factors analysis for albuminuria and glomerular hyperfiltration in pediatric patients with SCA. Table S3. Risk factors analysis for estimated glomerular filtration rate in pediatric patients with SCA. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts