# Acidosis at Diagnosis of Type 1 Diabetes Mellitus: Relation With Kidney Function

## TL;DR

This study shows that kidney damage at the start of type 1 diabetes worsens acidosis by reducing bicarbonate reabsorption.

## Contribution

The study identifies a strong link between renal tubular damage and acidosis at diagnosis of type 1 diabetes.

## Key findings

- Renal tubular damage (RTD) is strongly associated with severe acidosis at T1DM onset.
- Urinary NGAL and β2-microglobulin levels correlate inversely with serum bicarbonate and blood pH.
- The ketone-to-bicarbonate ratio effectively predicts RTD at T1DM diagnosis.

## Abstract

Acidosis at type 1 diabetes mellitus (T1DM) onset results from unregulated massive overproduction of ketoacids. However, renal tubular damage (RTD), a complication of T1DM onset, may impair bicarbonate reabsorption, exacerbating acidosis. We aimed to assess RTD role in acidosis at T1DM onset.

RTD was defined by urinary β2-microglobulin > 0.33 mg/L and/or neutrophil gelatinase-associated lipocalin (NGAL) > 95th percentile for age. Acute kidney injury (AKI) was classified using Kidney Disease/Improving Global Outcomes (KDIGO) criteria. Participants were grouped by serum ketone levels (cut-off: 3 mmol/L, above which indicates diabetic ketoacidosis [DKA]) and bicarbonate levels (cut-off: 22 mmol/L, below which indicates acidosis):– Group 1: Ketones ≥ 3 mmol/L, bicarbonate < 22 mmol/L.– Group 2: Ketones < 3 mmol/L, bicarbonate < 22 mmol/L.– Group 3: Ketones ≥ 3 mmol/L, bicarbonate ≥ 22 mmol/L.– Group 4: Ketones < 3 mmol/L, bicarbonate ≥ 22 mmol/L.

– Group 1: Ketones ≥ 3 mmol/L, bicarbonate < 22 mmol/L.

– Group 2: Ketones < 3 mmol/L, bicarbonate < 22 mmol/L.

– Group 3: Ketones ≥ 3 mmol/L, bicarbonate ≥ 22 mmol/L.

– Group 4: Ketones < 3 mmol/L, bicarbonate ≥ 22 mmol/L.

Of 185 individuals, 111 (60%) were in Group 1, 18 (9.7%) in Group 2, 8 (4.3%) in Group 3, and 48 (26%) in Group 4. Group 1 had the most severe clinical and biochemical derangements, followed by Groups 2, 3, and 4. Logistic regression, adjusted for AKI, relative difference of weight loss and glycated hemoglobin (HbA1c), identified RTD (odds ratio [OR] = 22.3; 95% confidence interval [CI]: 6.9–71.5; p  < 0.001), and relative difference of weight loss, (OR = 1.2; 95% CI: 1.1–1.4; p  < 0.006), as significant factors associated with Group 1 and only RTD (OR = 29.9; 95% CI: 3.0–292.9; p=0.004) as significant factor associated with Group 2. Serum bicarbonate and blood pH showed an inverse correlation with urinary NGAL (r2 = 0.61 and 0.56, respectively, both p  < 0.001) and β2-microglobulin (r2 = 0.67 and 0.59, respectively, both p  < 0.001), regardless of ketone levels. The ketone-to-bicarbonate ratio predicted RTD (area under the receiver-operating characteristic (ROC) curve (AUROC) = 0.94; 95% CI: 0.91–0.97; p  < 0.001), while serum bicarbonate levels predicted normal renal tubular function (AUROC = 0.95; 95% CI: 0.92–0.98; p  < 0.001).

A link exists between biological markers of RTD and metabolic acidosis at T1DM onset.