Association of urinary citrate excretion, pH, and net gastrointestinal alkali absorption with diet, diuretic use, and blood glucose concentration

Abstract

Urinary citrate (Ucit) protects against urinary stone formation. Acid base status and diet influence Ucit. However, the effect of demographics, diet, and glucose metabolism on Ucit excretion, urinary pH (U-pH) and net gastrointestinal alkali absorption (NAA) are not known. Twenty-four hour urine samples, blood glucose, creatinine, and cystatin C were obtained from non-Hispanic white sibships in Rochester, MN (n = 446; 64.5 ± 9 years; 58% female). Diet was assessed by a food frequency questionnaire. The impact of blood glucose, demographics and dietary elements on Ucit excretion, U-pH, and NAA were evaluated in bivariate and multivariable models and interaction models that included age, sex, and weight. NAA significantly associated with Ucit and U-pH In multivariate models Ucit increased with age, weight, eGFR_Cys, and blood glucose, but decreased with loop diuretic and thiazide use. U-pH decreased with serum creatinine, blood glucose, and dietary protein but increased with dietary potassium. NAA was higher in males and increased with age, weight, eGFR_Cys and dietary potassium. Significant interactions were observed for Ucit excretion with age and blood glucose, weight and eGFR_Cys, and sex and thiazide use. Blood glucose had a significant and independent effect on U-pH and also Ucit. This study provides the first evidence that blood glucose could influence urinary stone risk independent of urinary pH, potentially providing new insight into the association of obesity and urinary stone disease.

Keywords: pH; Citrate; diabetes mellitus; diet; glucose; nephrolithiasis.

Figure 1

Urinary citrate versus net alkali absorption. Adjusted urinary citrate was significantly, positively associated with adjusted net alkali absorption (β = 4.7062; P < 0.0001). Urinary citrate was adjusted for age, sex, serum creatinine, dietary potassium intake, weight, blood glucose, eGFR _cys, diuretic loop use, and diuretic Thiazide use. Net alkali absorption was adjusted for age, sex, serum creatinine, dietary potassium, eGFR _cys, and diuretic loop use.

Figure 2

Urine pH versus net alkali absorption. Adjusted urine pH was significantly, positively associated with adjusted net alkali absorption values (β = 0.009; P < 0.0001). Urine pH was adjusted for age, sex, serum creatinine dietary potassium intake, blood glucose, diuretic loop use, and dietary protein intake. Net alkali absorption was adjusted for age, sex, serum creatinine, dietary potassium, eGFR _cys, and diuretic loop use.

Figure 3

Urinary citrate versus urine pH. Adjusted urinary citrate was not significantly associated with urine pH (β = 40.18; P = 0.10). Urinary citrate was adjusted for age, sex, serum creatinine, dietary potassium intake, weight, blood glucose, eGFR _cys, diuretic loop use, and diuretic Thiazide use. Urine pH was adjusted for age, sex, serum creatinine dietary potassium intake, blood glucose, diuretic loop use, and dietary protein intake.

Figure 4

Interaction models. (A) Age and blood glucose on urinary citrate excretion. Higher age decreased the effect of blood glucose on urinary citrate excretion (β _{age x blood glucose} = −0.20, P = 0.01). SD, age (years), blood glucose (mg/dL). (B) eGFR_{C ys} and weight on urinary citrate excretion. Higher weight increased the effect of eGFR_{C ys} on urinary citrate excretion (β _{weight x eGFR cys} = 0.07, P = 0.01). eGFRcys: estimated Glomular Filtration Rate (mL/min/1.73 m²), weight in kg, SD. (C) Sex and Thiazide diuretic use on urinary citrate excretion. Thiazide use decreased urinary citrate excretion more in males than females (β _{sex x thiazide use} = −114.22, P = 0.04). SD. (D) Weight and loop diuretic use on net alkali absorption. Individuals taking loop diuretics showed increased net alkali absorption as weight increased, while those not taking loop diuretics showed a decreased net alkali absorption as weight increased. (β _{weight x loop diuretic use} = −0.45, P = 0.04).