Is hyperfiltration associated with higher urine albumin-to-creatinine ratio at follow up among Indigenous Australians? The eGFR follow-up study

Abstract

Background

Glomerular hyperfiltration is not able to be detected in clinical practice. We assessed whether hyperfiltration is associated with albuminuria progression among Indigenous Australians at high risk of diabetes and kidney disease to determine its role in kidney disease progression.

Methods

Longitudinal observational study of Indigenous Australians aged ≥18 years recruited from >20 sites, across diabetes and/or kidney function strata. At baseline, iohexol clearance was used to measure glomerular filtration rate (mGFR) and hyperfiltration was defined as (i) a mGFR of ≥125 mL/min/1.73 m², and (ii) an age-adjusted definition, with the top 10% of the mGFR for each 10 year age group at baseline. Baseline and follow-up urine albumin-to-creatinine ratio (uACR) was collected, and linear regression was used to assess the associations of hyperfiltration and uACR at follow up.

Results

407 individuals (33% men, mean age 47 years) were followed-up for a median of 3 years. At baseline, 234 had normoalbuminuria and 173 had albuminuria. Among participants with normoalbuminuria, those with mGFR ≥125 mL/min/1.73 m² had 32% higher uACR at follow-up (p = 0.08), and those with age-adjusted hyperfiltration had 60% higher uACR (p = 0.037) compared to those who had normofiltration. These associations were independent of uACR at baseline, but attenuated by HbA_1c. Associations were stronger among those without than those with albuminuria at baseline.

Conclusions

Although not available for assessment in current clinical practice, hyperfiltration may represent a marker of subsequent albuminuria progression among individuals who have not yet developed albuminuria.

Introduction

Indigenous populations, including Indigenous Australians, experience very high rates of both type 2 diabetes and kidney disease.1., 2., 3. Diabetes is the leading cause of end stage kidney disease in Australia. Up to 77% of Indigenous Australians diagnosed with end stage kidney disease have diabetes as a co-morbidity compared to 33% of non-Indigenous Australians with end stage kidney disease.⁴ Early diabetes has been characterised by elevated glomerular filtration. It is thought that the processes that occur during hyperfiltration may have a detrimental effect on the kidney.⁵ However, the significance of hyperfiltration in terms of causing or predicting the development of albuminuria and chronic kidney disease remains uncertain.

Several structural and functional changes, including hyperfiltration (early on) and increased albuminuria (later on) may be associated with the development of chronic kidney disease (CKD) in patients with diabetes.6., 7., 8., 9. The presence of albuminuria is an important and well recognized risk factor for the development of CKD, particularly in Indigenous Australians^10,11 and in other Indigenous populations across the world including in Pima Indians with diabetes.¹² The association of albuminuria with accelerated glomerular filtration rate (GFR) loss has been confirmed in several studies,^13,14 and the CKD classification incorporates albuminuria stage along with the estimated GFR.¹⁵ Although the majority of people with diabetes follow a classical albuminuria-based pathway,⁸ normoalbuminuric progression of diabetic kidney disease is also recognized.^8,16

The significance of hyperfiltration in terms of causing or predicting the development of albuminuria and CKD remains uncertain as previous studies report conflicting findings, partly due to methodological problems associated with accurately measuring true GFR in the hyperfiltering range, and also due to a lack of studies utilising measured GFR with sufficient long term follow-up.¹⁷ Evidence that hyperfiltration has a pathological impact on renal function has mostly been elucidated from animal studies.¹⁸ In humans, some,¹⁹ but not all,²⁰ studies have demonstrated an association between the presence of hyperfiltration at baseline and higher follow-up urine albumin-to-creatinine ratio (uACR). Most studies involve individuals with type 1 diabetes,19., 20., 21., 22. and fewer studies have assessed the contribution of hyperfiltration to progression of albuminuria among individuals with type 2 diabetes.²³ The limitations of the studies on participants with type 2 diabetes include small sample sizes,^24,25 use of estimated GFR (eGFR) instead of measured GFR (mGFR)²⁶ and the inclusion of participants with pre-existing albuminuria,²³ making it difficult to assess the temporal association between hyperfiltration and albuminuria progression.

There are no studies assessing the relationship between hyperfiltration and albuminuria progression in Indigenous Australians, who have a very high prevalence of diabetes and kidney disease. Previously, we have shown in Indigenous Australians that albuminuria is a strong predictor of eGFR decline,²⁷ and demonstrated that the prevalence of hyperfiltration varied from 7% to 27% depending on the definition of hyperfiltration employed.²⁸ Furthermore, while hyperfiltration was more prevalent among those with, than those without type 2 diabetes, a significant proportion of those with intermediate hyperglycaemia also had hyperfiltration,²⁸ suggesting that hyperfiltration may be associated with metabolic abnormalities before diabetes is diagnosed.

In this analysis of longitudinal data from the eGFR study cohort, we now aim to assess the temporal associations between baseline hyperfiltration determined from directly measured GFR, and uACR at follow-up, after considering baseline uACR, HbA_1c and other covariates. As it is likely that those with pre-existing albuminuria may already be on a pathway for albuminuria progression, we chose to undertake analyses separately in those with normoalbuminuria and albuminuria at baseline. We hypothesized that in this population with a high burden of metabolic syndrome, diabetes and kidney disease, the presence of hyperfiltration at baseline will be associated with an increased risk of greater uACR at follow up.