Abstract Background and Aims Patients with kidney stones (KS) are at increased risk for chronic kidney disease. Research published so far has focused on factors associated with low glomerular filtration rate in KS patients, however, factors associated with proteinuria have not been fully described. The aim of this retrospective study was to assess the presence of proteinuria in our cohort of patients with known or presumed calcium KS and to analyze which common clinical and metabolic variables might be associated with proteinuria. Method We performed a retrospective cross-sectional study in patients (pts) treated at our stone clinic at the University Medical Centre Ljubljana in the ten-year period from January 1st 2010 to December 31st 2019. We excluded patients who did not have 24-hour proteinuria determined and patients with known urate, struvite or cystine stones. We noted the presence of arterial hypertension (AH), diabetes mellitus, a family history of kidney stones, age, sex, eGFR, body mass index, history of renal colic events and urological procedures for kidney stones, and 24-hour urine measurements (calcium, urate, oxalate, citrate, sodium, and proteinuria). Categorical variables were analyzed with Chi-Square test, normally distributed numerical values with Student's t-test and non-normally distributed with Mann-Whitney U test. P-values < 0.05 were considered statistically significant. Associations with p-values < 0.1 were included into a binary logistic regression analysis to determine the probability of predicting the presence of 24-hour proteinuria, defined as > 0.25 g/day in our laboratory. Results In the allocated time period 335 pts were examined. After exclusion, 251 pts were included into the analysis (mean age 49.5 ± 14.2 years, 116 (46.2%) males, 135 (53.8%) females). Complete 24-hour urine metabolic analysis was available for 238 pts. Proteinuria was present in 40 pts (15.9%) and was significantly more common in pts with AH compared to normotensive pts (18/77 (23.4%) vs. 22/174 (12.6%), p 0.027) and more common in males compared to females (23/116 (19.8%) vs. 17/135 (12.6%), p 0.083). Numerical variables with a p-value < 0.1 regarding the association with the presence of proteinuria are presented in Table 1. In regression analysis the model significantly predicted the presence of proteinuria (χ2 20.738, p 0.014, R2 0.343) and was able to predict the result overall in 85.2% of cases. Of all the variables included only citrate significantly predicted the presence of proteinuria (OR 1.732; 95% CI 1.009, 2.972; p 0.046). Proteinuria was significantly more common in pts with hypocitraturia (<1.67 mmol/day) compared to pts with normal urinary citrate (29/116 (25%) vs. 9/122 (7.4%), p < 0.001) and in pts with immeasurably low urinary citrate (<0.2 mmol/day) compared to higher citrate (15/27 (55.6%) vs. 23/211 (10.9%), p < 0.001). Conclusion Proteinuria was relatively common in our cohort. The main factor associated with proteinuria was urinary citrate. Proteinuria was significantly more common in pts with hypocitraturia compared to pts with normal urinary citrate and was present in the majority of pts with immeasurably low urinary citrate. Further studies are warranted to clarify the clinical relevance and implications of this findings.
Read full abstract