Insulin Sensitivity and Diabetic Kidney Disease in Children and Adolescents With Type 2 Diabetes: An Observational Analysis of Data From the TODAY Clinical Trial

Abstract

Background: Diabetic kidney disease is a major cause of premature mortality in type 2 diabetes mellitus (T2DM). Worsening insulin sensitivity independent of glycemic control may contribute to the development of diabetic kidney disease. We investigated the longitudinal association of insulin sensitivity with hyperfiltration and increased albumin excretion in adolescents with T2DM.

Study design: Observational prospective cohort study.

Setting & participants: 532 TODAY (Treatment Options for Type 2 Diabetes in Adolescents and Youth) participants aged 12 to 17 years with T2DM duration less than 2 years at baseline. The TODAY Study was a multicenter randomized clinical trial that examined the efficacy of 3 treatment regimens (metformin monotherapy, metformin plus rosiglitazone, or metformin plus an intensive lifestyle intervention program) to achieve durable glycemic control.

Predictors: Natural log-transformed estimated insulin sensitivity (reciprocal of fasting insulin), hemoglobin A_1c concentration, age, race-ethnicity, treatment group, body mass index, loss of glycemic control, and hypertension.

Outcomes: Hyperfiltration was defined as 99th percentile or higher of estimated glomerular filtration rate (≥140mL/min/1.73m²) when referenced to healthy adolescents (NHANES 1999-2002) and albumin-creatinine ratio ≥ 30μg/mg at 3 consecutive annual visits.

Results: Hyperfiltration was observed in 7.0% of participants at baseline and in 13.3% by 5 years, with a cumulative incidence of 5.0% over 5 years. The prevalence of increased albumin excretion was 6% at baseline and 18% by 5 years, with a cumulative incidence of 13.4%. There was an 8% increase in risk for hyperfiltration per 10% lower estimated insulin sensitivity in unadjusted and adjusted models (P=0.01). Increased albumin excretion was associated with hemoglobin A_1c concentration, but not estimated insulin sensitivity.

Limitations: Longer follow-up is needed to capture the transition from hyperfiltration to rapid glomerular filtration rate decline in youth-onset T2DM.

Conclusions: Lower estimated insulin sensitivity was associated with risk for hyperfiltration over time, whereas increased albumin excretion was associated with hyperglycemia in youth-onset T2DM.

Keywords: Type 2 diabetes mellitus (T2DM); adolescents; albumin-creatinine ratio (ACR); children; creatinine; cystatin C; diabetic kidney disease (DKD); disease progression; estimated glomerular filtration rate (eGFR); hyperfiltration; increased albumin excretion; insulin sensitivity; kidney function; youth-onset T2DM.

Figure 1. Annual longitudinal changes for (A) mean eGFR, (B) median ACR and (C) median eIS (along with 95% CI error bars) across the 5 years in the TODAY study^*

^*The arithmetic mean and 95% CI error bars is shown for eGFR (A) values over time. The geometric mean and 95% CI error bars is shown for ACR (B) and eIS (C) due to skewed distributions. The geometric mean is a good approximation of the median as the log-transformed data is approximately symmetric. Data shown in figures from all participants, irrespective of whether participants lost glycemic control. ^†P-value from an unadjusted model testing for linear trend over time, and from adjusted model including sex, race-ethnicity, treatment group, baseline BMI, baseline HbA1c, age at baseline, treatment failure status (loss of glycemic control yes/no), and hypertension diagnosis.

Figure 2. Cumulative incidence curves for time to hyperfiltration by Zappitelli combined creatinine and cystatin C equation during TODAY^*

^*Participants with hyperfiltration at baseline are excluded. Time to hyperfiltration defined at the first of three consecutive annual visits with eGFR ≥ 99^th percentile for NHANES controls (i.e, eGFR ≥ 140ml/min/1.73m²); Data shown in figure from all participants, irrespective of whether participants lost glycemic control.