Insulin Sensitivity and Risk for Hyperfiltration in Youth with Type 2 Diabetes Mellitus

The leading cause of end-stage renal disease (ESRD) is diabetic kidney disease. Among children and adolescents with type 2 diabetes mellitus (T2DM), early diabetic kidney disease, including hyperfiltration and increased excretion of albumin, is common; progression occurs at “an alarming rate,” according to Peter Bjornstad, MD, and colleagues. In a small cohort of 46 adolescents with T2DM, the prevalence of hyperfiltration was 24% and of increased albumin excretion, 34%.

Dr. Bjornstad et al. previously reported longitudinal data from the TODAY (Treatment Options for Type 2 Diabetes in Adolescents and Youth) study. The results demonstrated that the prevalence of increased excretion of albumin increased from 6.3% at baseline to 16.6% by the end of the study. The data were reported after only 3.9 years of follow-up; further, there were no laboratory data available to assess hyperfiltration.

Children and adolescents diagnosed with T2DM appear to have a particularly high risk for progression to diabetic kidney disease; the risk is significantly greater than that in youth with type 1 diabetes or in adults diagnosed with T2DM with similar duration of disease. Hyperfiltration is an early indicator of diabetic kidney disease and often precedes increased excretion of albumin and decline in kidney function, making identification of clinical phenotypes associated with hyperfiltration and predictive of diabetic kidney disease a key factor in improving outcomes in adolescents with T2DM.

In adults with and without T2DM, reduction in insulin sensitivity has been associated with the development of future kidney disease. The mechanisms underlying the association between reduced insulin sensitivity and diabetic kidney disease are unclear; however, experimental evidence indicates that the energy profile of T2DM cannot accommodate the renal hypermetabolism of diabetic kidney disease. In a small cross-sectional cohort of adolescents with T2DM, associations between insulin resistance and hyperfiltration have been reported; however, data in larger cohorts are lacking. In addition, there are few data available on the role of hyperfiltration in adolescents with T2DM.

In a recent issue of the American Journal of Kidney Diseases [2018;71(1):65-74], Dr. Bjornstad and colleagues added to their previous work by describing the prevalence and incidence of hyperfiltration in the TODAY study cohort at baseline and over 5 years; data on albumin excretion out to 5 years is also reported. The researchers also sought to examine the longitudinal relationship between estimated insulin sensitivity and renal outcomes during the 5 years of the study to test the hypothesis that there would be an association between lower estimated insulin sensitivity in adolescents with T2DM and diabetic kidney disease, reflected by increased risk for hyperfiltration and increased albumin excretion.

The current analysis included 532 participants; 63.9% (n=340) were female, mean age was 13.9 years, mean duration of diabetes was 7.9 months, 33.6% (n=179) were black non-Hispanic, 43.0% (n=229) were Hispanic, 20.1% (n=107) were white non-Hispanic, and 3.2% (n=17) were other race/ethnicity. Mean body mass index (BMI) was 35 kg/m2, 4.7% (n=25) used an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker (ACE/ARB), 10.8% (n=100) had a diagnosis of hypertension, mean serum creatinine was 0.61 mg/dL, and mean serum cystatin was 0.75 mg/L.

During the study period, there was an annual increase in estimated glomerular filtration rate (eGFR) by a mean of 2.45 mL/min/1.73 m2 (P<.001), an average increase in albumin-creatinine ratio (ACR) of 5.5% per year (P=.002), and an average decrease in estimated insulin sensitivity of 5.4% per year (P<.001). In fully adjusted models, these linear trends remained significant for eGFR.

At baseline, 1.5% of participants had severely increased excretion of albumin; there was no significant increase over time. In a subgroup of participants who remained in glycemic control, the increase in eGFR remained (P=.01) and estimated insulin sensitivity decreased over time (P=.01), at a lower but still significant rate. In fully adjusted models, the trends remained significant.

At baseline, 12.8% of participants had an elevated eGFR (≥140 mL/min/1.73 m2); at year 5, 26.8% had elevated eGFRs. Based on the first of three consecutive annual elevated eGFRs, hyperfiltration was observed in 7.0% of participants at baseline and in 13.3% at 5 years. At baseline, the prevalence of increased excretion of albumin was 6.0%, increasing to 18.0% at 5 years; cumulative incidence over 5 years was 13.4%.

Both hyperfiltration and increased excretion of albumin were seen in 0.8% of participants at baseline and in 4.0% at year 5. In analysis of data of a subgroup of participants who remained in glycemic control, there was no increase over time in the prevalences of hyperfiltration and elevated excretion of albumin.

In a univariate model, there was an association between a 2.1-fold increased risk for hyperfiltration with lower estimated insulin sensitivity (an 8% increase in the risk for hyperfiltration per 10% lower estimated insulin sensitivity). In adjusted models, the association remained significant. There was no association of hemoglobin A1c (HbA1c) concentration, BMI, age, loss of glycemic control, development of hypertension, or treatment group with increased risk for developing hyperfiltration. There was an association between male sex and decreased risk for developing hyperfiltration (P=.03).

In sensitivity analyses in the multivariate model, the researchers adjusted for ACE-inhibitor/ARB use rather than the development of hypertension; the association between estimated insulin sensitivity and incident hyperfiltration remained statistically significant (hazard ratio [HR] 2.05; 95% confidence interval [CI], 1.11-3.79; P=.02). The model was also rerun without HbA1c concentration: loss of glycemic control conferred greater risk for hyperfiltration (HR, 3.49; 95% CI, 1.43-8.47; P=.006).

In unadjusted and adjusted models, there was an association between a 10% worsening of estimated insulin sensitivity relative to baseline with a 7% increase in hyperfiltration. There was no association between change in HbA1c concentration or change in BMI from baseline with increased risk of hyperfiltration. Following adjustment for ACE inhibitor/ARB use rather than a diagnosis of hypertension, the relationship between change in estimated insulin sensitivity and incident hyperfiltration was of borderline statistical significance (HR, 1.54; 95% CI, 1.00-2.35; P=.05).

Finally, there was no association between increased risk for developing increased albumin excretion and change in estimated insulin sensitivity or change in BMI from baseline in both unadjusted and adjusted models. There was an association between change in HbA1c concentration from baseline and a diagnosis of hypertension with increased risk for increased albumin excretion.

The researchers cited some limitations to the study, including the use of calculated GFR and estimated insulin sensitivity and the length of follow-up.

“In summary, the high rates of increased albumin excretion and hyperfiltration in youth with T2DM forecast early renal morbidity and mortality. Improved understanding of the determinants of diabetic kidney disease risk and progression over longer periods of follow-up is essential to improve outcomes in youth with T2DM. We found that insulin resistance rather than adiposity, blood pressure, or glycemic control was associated with greater risk for hyperfiltration during the 5-year study period in adolescents with T2DM. Worsening insulin sensitivity may therefore represent an early risk factor for future progression of diabetic kidney disease in adolescents with T2DM,” the researchers said.

Takeaway Points

  1. Researchers provided an update to previously reported data from the TODAY (Treatment Options for Type 2 Diabetes in Adolescents and Youth) study on the longitudinal association of insulin sensitivity with hyperfiltration and increased excretion of albumin in adolescents with type 2 diabetes mellitus.
  2. At baseline, hyperfiltration was seen in 7.0% of participants; at 5 years, the incidence was 13.3%, with a cumulative incidence of 5.0% over 5 years.
  3. There was an association between lower estimated insulin sensitivity and the risk for hyperfiltration over time; increased albumin excretion was associated with hyperglycemia in this patient population.