Effect of Being Overweight on Urinary Metabolic Risk Factors for Kidney Stone Formation

Linda Shavit; Pietro Manuel Ferraro; Nikhil Johri; William Robertson; Steven B. Walsh; Shabbir Moochhala; Robert Unwin


Nephrol Dial Transplant. 2015;30(4):607-613. 

In This Article


A total of 2132 patients was studied, of whom 833 (39%) were NW, 863 (40.5%) were OW and 436 (20.5%) were obese.

Figure 1 presents the trends of BMI over time (1995–2012). A statistically significant and stepwise increase in average BMI (+0.14 kg/m2 for the 2000–01 period, +0.97 kg/m2 for the 2002–05 period, +1.16 kg/m2 for the 2006–12 period compared with the 1995–99 period, P for trend <0.001) was observed over this time period.

Figure 1.

Trends of BMI of the study population over time—years 1995 to 2012. Statistically significant and stepwise increase in average BMI (+0.14 kg/m2 for the 2000–01 period, +0.97 kg/m2 for the 2002–05 period, +1.16 kg/m2 for the 2006–12 period compared with the 1995–99 period, P-value for trend <0.001) was observed during this period of time.

The demographic and clinical characteristics of the population broken down by BMI categories are shown in Table 1. OW and obese KSF were older (mean age in years 43 ± 15 in NW, 48 ± 13 in OW, and 50 ± 12 in obese; P for trend <0.001), demonstrated a higher female preponderance and increased prevalence of diabetes, hypertension and gout (P for trend <0.001 for all the above variables). The prevalence of recurrent stone disease was 38% in NW, 61% in OW and 54% in obese KSF (P for trend <0.001). The median duration of kidney stone disease was not significantly longer in OW and obese KSF (P for trend = 0.04).

Comparison of 24-h urine composition in patients stratified by categories of BMI is presented in Table 2. Both, crude and adjusted analyses are presented. Adjustment was performed for age, sex and all urinary parameters (mainly volume and creatinine excretion). Since diabetes and hypertension are well-recognized consequences of obesity and metabolic syndrome, adjustment for these confounders was not performed to avoid over adjustment. There were no statistically significant differences in U.Vol, and U.Mg among the groups. Significantly higher levels of U.Ca, U.Ox, U.Cit were detected by crude analyses only. However, significantly higher levels of U.UA (3.3 ± 1.1 versus 3.8 ± 1.2 versus 4.0 ± 1.2 mmol/L; P < 0.001 for trend), U.Na (151 ± 57 versus 165 ± 60 versus 184 ± 63 mmol/L; P < 0.001 for trend), U.K (69 ± 22 versus 72 ± 22 versus 72 ± 22, P = 0.01) and lower U.pH (6.3 ± 0.5 versus 6.1 ± 0.5 versus 6.0 ± 0.6; P < 0.001 for trend) were found in OW and obese KSF in both crude and multivariate adjusted models. As expected, prevalence of the low urinary pH (<6) was more common in patients with hypocitraturia (42 versus 36%, P = 0.01) and this finding was similar for all BMI groups.

Crude and age- and sex-adjusted analyses of the prevalence of urinary metabolic abnormalities among the study groups are set out in Table 3. There were no statistically significant differences in the prevalence of low urine volume (7.0% versus 4.8% versus 4.9%, P = 0.87), hypocitraturia (25% versus 21% versus 21%, P = 0.06) and hyperoxaluria (6.8% versus 8.2% versus 11%, P = 0.07) among NW, OW and obese KSF, respectively. However, hyperuricosuria (9.3% versus 21% versus 27%, P < 0.001) and hypercalciuria (32% versus 39% versus 34%, P = 0.01) were significantly more prevalent between OW and obese patients. In addition, increased prevalence of higher Psf for CaOx, UA and CaOx/UA stone types was detected in OW and obese compared with NW KSF (Table 3).

Data on stone composition were available for 640 of the 2132 patients (30%) (Table 4). More than 60% of samples were stones composed of more than one constituent. The prevalence of calcium oxalate stones was slightly more than 50% in all BMI categories. However, calcium phosphate stones were significantly more prevalent in NW KSF, whereas the prevalence of uric acid stones was positively correlated with BMI (4.3 ± 18.9% versus 14 ± 34% versus 24 ± 41%, P for trend <0.001). There was no difference in the prevalence of calcium oxalate stones across the BMI groups (52 ± 39 versus 58 ± 39 versus 54 ± 41, P = 0.33).

No interaction by sex was detected in all of the above-mentioned analyses (24-h urine parameters and stone composition comparisons, P > 0.05 for all interactions).

Contribution of dietary parameters on urinary and stone composition was evaluated. The following dietary covariates were analysed: intake of fluid, calcium, magnesium, oxalate, animal protein and sodium. No statistically significant association was found between specific dietary components and urine metabolic parameters and stone composition.