Defining Variation in Urinary Oxalate in Hyperoxaluric Stone Formers

Abstract

Background and Purpose:

The development of effective preventive therapy for renal calculi in patients with secondary hyperoxaluria (2°HO) relies on establishing the pattern of normal variation in urinary oxalate (uOx) and attempting to reduce it. Therefore, we evaluated uOx at baseline and at subsequent time points in stone formers with 2°HO.

Methods:

We reviewed the charts of 201 recurrent stone formers with 2°HO (uOx ≥40 mg/day). The 24-hour urine collections at baseline and after initiation of clinician-directed therapies were analyzed. Mixed models were constructed to analyze uOx over time for individual patients and as a group. Subgroup analyses were performed for enteric and idiopathic 2°HO. Coefficients of variation were computed using the root mean square error from linear models.

Results:

The etiology of 2°HO was enteric in 17.9% and idiopathic in 82.1% of patients. Among the 943 urine collections analyzed, 196 oxalate values were derived from the enteric group and 747 from the idiopathic group. The median number of uOx values measured per person was four. The median 24-hour uOx (mg/day) was significantly higher for the enteric group than for the idiopathic group at the time of diagnosis: 64.4 (interquartile range [IQR]=48–90) vs 46.0 (IQR=38–56), P<0.001) and during follow-up (58.2 [IQR=46–86] vs 44.2 [IQR=35–53], P<0.001). Over a median follow-up of 22.5 months, 44.4% of the enteric and 61.8% of the idiopathic patients had at least one normal uOx value (P=0.06). The coefficients of variation for the enteric and idiopathic groups were 40.8% and 27.3%, respectively, with variation randomly displayed in either direction for both groups.

Conclusions:

Among patients with 2°HO, uOx demonstrates significant random variation over time even with the incorporation of standard treatments, with enteric HO demonstrating higher values and greater variance than idiopathic HO.

Introduction

Oxalate is a salt-forming ion with an affinity for calcium, producing a highly insoluble calcium oxalate (CaOx) complex in biologic fluids. Increasing urinary oxalate (uOx) leads to supersaturation of urine with respect to CaOx, thereby promoting the development of nephrocalcinosis and/or urinary stones.¹ Hyperoxaluria (HO), defined as a uOx excretion exceeding 40 to 45 mg/24 hours, has an estimated prevalence of 12% to 47% in recurrent CaOx stone formers.^1

–4

The etiologies of HO can be divided into two broad categories: primary and secondary. Secondary HO (2°HO) occurs in settings of increased intestinal absorption of oxalate as a result of excessive dietary oxalate intake (dietary)⁵ or states of fat and/or bile acid malabsorption (enteric).⁶ Other causes of 2°HO that are under investigation include alterations in transporters responsible for intestinal oxalate secretion^7,8 and depletion or absence of microorganisms in the intestine that have been shown to both degrade oxalate⁹ and promote intestinal oxalate secretion.¹⁰

Treatment strategies to reduce urinary stone formation in the setting of 2°HO, such as dietary oxalate and/or fat restriction,^11,12 calcium supplementation,¹³ pyridoxine,¹⁴ and probiotic therapy,¹⁵ have been associated with limited compliance and success; consequently, stone recurrence is common.¹⁶ Stone formation in 2°HO is driven by CaOx supersaturation, and reducing uOx should effectively decrease the driving force for stone formation. An important step toward achieving this therapeutic goal, and for which data are lacking, is characterization of the variation in uOx levels among HO stone formers. As such, we sought to define the variation in uOx at baseline and during standard treatment regimens in stone formers with 2°HO. Our goal was to establish whether standard treatment resulted in a consistent decline in uOx or if uOx continued to demonstrate a pattern of normal variation.

Patients and Methods

Subjects

After obtaining Institutional Review Board approval, the charts of all patients evaluated for stone disease at the Charles and Jane Pak Center for Mineral Metabolism and/or in the urology department at the University of Texas Southwestern Medical Center between June 1989 and January 2013 were reviewed. Inclusion criteria included age ≥18 years, clinical history of at least one lifetime urinary stone, and diagnosis of 2°HO (two or more uOx values ≥40 mg/day). A total of 201 patients met the inclusion criteria, and they were categorized into two groups: “enteric” (those with intestinal pathology associated with fat malabsorption including small intestine resection, gastric bypass or banding, Crohn disease, ulcerative colitis, or celiac disease) and “idiopathic” (excessive dietary oxalate intake or unknown cause).

24-hour urine collections

To determine uOx variation, at least two 24-hour urine collections were needed, with a minimum follow-up of 24 hours and a maximum follow-up of 10 years. All 24-hour uOx values, regardless of concomitant therapeutic intervention directed at HO, were identified, and only those values from complete 24-hour collections (defined as a 24-hour urinary creatinine excretion within 20% of the median urinary creatinine excretion for each person, using all available 24-hour urine collections for each patient) were included in the analyses.

Statistical analysis

Mixed linear models were constructed to analyze uOx values over time for individual patients, modeling each patient as a random effect. Planned subgroup analyses were performed according to the etiology of HO (enteric and idiopathic) and compared with t tests for independent observations and mixed models for correlated observations. Coefficients of variation were computed using the root mean square error from linear models. Categorical variables were compared with the Fisher exact test. Time to normalization of uOx was analyzed with Kaplan-Meier estimates and the log-rank test. Associations with a P≤0.05 were considered statistically significant. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC).

Results

The demographic data for the 201 included study patients are shown in Table 1. The diagnosis of 2°HO was enteric in 17.9% and idiopathic in 82.1% of patients. The majority of hyperoxaluric stone-forming patients were male (78% male vs 22% female), and this sex distribution was consistent in both the enteric and idiopathic groups. The majority of patients in both groups were Caucasian. The enteric group was significantly older than the idiopathic group, and there was no significant difference in body mass index between groups.

Table 1.

Demographic and Clinical Characteristics for Hyperoxaluric Stone Formers Stratified by Diagnosis

		Diagnosis
	Total	Enteric	Idiopathic	P value
Number of patients (%)	201 (100)	36 (17.9)	165 (82.1)
Urine collections per patient
Median (IQR)	4 (2–18)	4 (2–18)	3 (2–17)
Age (years)
Mean (SD)	52 (12.8)	56 (10.8)	51 (13.3)	0.032^*
Median (IQR)	54 (43–61)	55 (36–80)	52 (19–81)
Sex
Male (%)	157 (78)	27 (75)	130 (79)	0.6577
Female (%)	44 (22)	9 (25)	35 (21)
Ethnicity
Caucasian (%)	192 (96)	35 (97)	157 (95)	0.837
Black (%)	4 (2)	1 (3)	3 (2)
Hispanic (%)	2 (1)	0 (0)	2 (1)
Asian (%)	3 (1)	0 (0)	3 (2)
BMI (kg/m²)
Mean (SD)	31.0 (8.0)	30.5 (10.7)	31.3 (7.4)	0.345
Median (IQR)	29.7 (26–35)	27.2 (15.7–64.3)	30.2 (15.1–57.9)

Denotes significance, P<0.05.

SD=standard deviation; IQR=interquartile range; BMI=body mass index.

A total of 943 uOx values were identified among the 201 patients, including 196 from the enteric group and 747 from the idiopathic group (Table 2). The median number of uOx values measured per patient was four (interquartile range [IQR]=2–18). The median 24-hour uOx (mg/day) at the time of diagnosis was significantly higher for the enteric (64.4 [IQR=48–90]) compared with the idiopathic group (46.0 [IQR=38–56], P<0.001) (Table 2). When all subsequent collections after diagnosis were considered, the median uOx (mg/day) remained significantly higher for the enteric compared with the idiopathic group (58.2 [IQR=46–86] vs 44.2 [IQR=35–53], P<0.001) (Table 2).

Table 2.

Twenty–Four Hour Urinary Oxalate at Diagnosis, During Follow–Up, and Total Stratified by Diagnosis

		Diagnosis
	Total	Enteric	Idiopathic	P value
Number of 24–hour uOx values
At diagnosis	201	36	165
After diagnosis	742	160	582
All values	943	196	747
24–hour uOx (mg/d) at diagnosis
Mean (SD)	53.2 (23.3)	72.8 (35.9)	49.0 (16.9)	<0.001^*
Median (IQR)	47.6 (38–61)	64.4 (48–90)	46.0 (38–56)
24–hour uOx (mg/d) after diagnosis
Mean (SD)	51.1 (23.3)	68.5 (33.3)	46.2 (16.8)	<0.001^*
Median (IQR)	47.0 (37–58)	58.2 (46–86)	44.2 (35–53)
24–hour uOx (mg/d), all values
Mean (SD)	51.5 (23.3)	69.3 (33.7)	46.8 (16.8)	<0.001^*
Median (IQR)	47.0 (37–58)	59.5 (46–86)	45.0 (36–54)

Denotes significance, P<0.05.

uOx=urine oxalate; SD=standard deviation; IQR=interquartile range.

Over a median follow-up of 22.5 months (IQR=7–45), a significantly greater proportion of idiopathic patients compared with enteric patients demonstrated at least one normal uOx value (61.8% vs 44.4%, P=0.06). Furthermore, among patients who attained normalization of uOx at least once, the time interval to normalization of uOx (<40 mg/day) was significantly longer for the enteric group (median of 52.3 months [IQR=3.9–86.7]) compared with the idiopathic group (median of 7.1 months [IQR=0–42.1], P<0.001) (Fig. 1). The coefficients of variation for the enteric and idiopathic groups were 40.8% and 27.3%, respectively, and the variation was randomly displayed in either direction.

FIG. 1.

Time to normalization of urine oxalate (<40 mg/day) by diagnostic group.

Discussion

The formation of urinary stones associated with HO is driven largely by CaOx supersaturation. Therefore, knowledge of uOx variation not only provides valuable insight into the disease process but also defines a therapeutic goal for stone prevention. A new therapy designed to reduce stone formation in 2°HO must prove to consistently reduce uOx beyond physiologic variation. Here, we provide the first description of the variation over time in uOx in stone formers with 2°HO with or without treatment. We found that patients with enteric HO demonstrated higher median uOx values at diagnosis and over the course of follow-up, had more variability in uOx over time, were less likely to normalize uOx values with standard treatment, and, when uOx did normalize, took longer to do so compared with those who had a diagnosis of idiopathic (including dietary) HO.

Investigators have demonstrated differences between enteric and idiopathic HO with regard to histopathologic evidence of CaOx stone formation. Evan and associates^17,18 procured renal papillary biopsies from patients undergoing percutaneous nephrolithotomy and found that in stone formers with mild HO (40–50 mg/day), hydroxyapatite crystals (Randall's plaques) originate in the basement membranes of the thin loops of Henle and spread through the interstitium to the renal papilla, where they serve as nidi for CaOx crystal deposition and ultimately stone formation. Conversely, in patients with enteric HO, they observed intraluminal apatite deposits in the inner medullary collecting ducts. In this scenario, stones were not adherent to plaque but instead were free-floating in the renal collecting system. These differences in crystal formation and deposition between enteric and idiopathic (including dietary) HO underscore their distinct pathophysiologic processes and suggest that patients may respond differently to varying levels of uOx and ultimately to the therapeutic interventions directed at altering uOx.

Despite differences in the pathogenesis of stone formation in enteric and idiopathic HO, CaOx urinary supersaturation likely plays a role in both etiologies, and uOx excretion is an important contributor to CaOx supersaturation. Historically, dietary oxalate was thought to be a relatively minor contributor to uOx (10%–20%).¹⁹ Holmes and colleagues,⁵ however, showed that dietary oxalate contributes up to 50% of uOx, the remainder attributed to endogenous oxalate production. In a metabolic study of 12 healthy subjects given prepared meals with total daily measured oxalate contents of 10, 50, and 250 mg, they found significant differences in uOx between diets, ranging from 24.4 mg/day in the 10 mg oxalate diet to 41.5 mg/day in the 250 mg diet.

Despite the significant contribution of dietary oxalate to uOx, however, stone formers with dietary HO typically have only mild to moderate uOx elevation (40–55 mg/day), and adherence to a low oxalate diet along with normal calcium intake and/or calcium supplementation has been shown to significantly decrease uOx excretion and CaOx supersaturation.²⁰ Accordingly, those patients in our analysis with idiopathic (largely dietary) HO had lower uOx at diagnosis and throughout follow-up and a quicker and more frequent normalization of uOx compared with those in the enteric group.

On the other hand, enteric HO occurs secondary to fat malabsorption caused by small bowel resection, Crohn disease, ulcerative colitis, celiac disease, or bariatric surgical procedures. Fat malabsorption increases intestinal oxalate absorption by two mechanisms: Increased colonic oxalate permeability because of the increased luminal bile salts and fatty acids, and sequestration of intraluminal calcium by fatty acids, thereby decreasing complexation of calcium and oxalate and increasing absorption of free soluble oxalate.^6,21 The uOx values for enteric HO have been described as moderate to severe with mean uOx values of approximately 80 mg/day described in the literature.²¹ Hence, in our analysis, the enteric group had a higher uOx at diagnosis and throughout follow-up, and was less likely to demonstrate normalization of uOx compared with the idiopathic group.

The coefficient of variation for the enteric group was higher than for the idiopathic group (40.8% vs 27.3%, respectively), with both groups displaying random variation in either direction. Consequently, individual uOx values, particularly in those with enteric HO, should be interpreted with caution. Further, therapies directed at correcting HO must consistently overcome these variations in uOx to affect a positive change. Future trial designs can use this information when powering and calculating sample sizes for observations deemed clinically relevant.

Although this series represents the first published description of the variation in uOx among HO stone formers, our results should be tempered by some inherent limitations. First is the retrospective design, which is vulnerable to recall and case-mix biases. Second, while one of the inclusion criteria was a lifetime history of stones, we did not correlate 24-hour urine values with stone burden, which in future studies might provide a helpful context in which to elucidate trends in uOx kinetics. In addition, we did not record stone composition or limit our analysis to CaOx stone formers, because determination of uOx variation could be applicable in any population.

Finally, a significant limitation is that we did not identify if and what type of treatment patients underwent throughout the study period. By omitting this from our analysis of patients seen, and presumably treated, at a tertiary care metabolic stone clinic, one could argue that the variation we observed may be blunted by treatment effect, preventing us from appreciating the true magnitude of variation that exists without treatment. Likewise, if the majority of these patients experienced considerable random uOx variation despite HO-directed therapy, particularly in the enteric group, one might conclude that existing treatments are ineffective at consistently lowering or stabilizing uOx, particularly in enteric HO, at least during our follow-up interval. It is sobering that uOx levels failed to show a consistent decline with initiation of treatment aimed at lowering uOx; however, we did not specifically analyze treatment effect in subgroups according to dietary/medication intervention.

Despite these limitations, our conclusions represent a sound analysis of the largest series of hyperoxaluric stone-forming patients for whom variations in uOx were assessed. Future directions to refine this work include assessing concomitant treatment in this series to establish a more granular understanding of true baseline uOx variation. In addition, correlation of uOx values with clinical outcomes (stone burden and type) will help to interpret these results in a clinically meaningful context so that they can ultimately be applied to prevention strategies for urinary stone formation.

Conclusions

Our results provide the first description of the variation in uOx that exists in hyperoxaluric stone formers with a diagnosis of 2°HO. We found that among patients with 2°HO, uOx demonstrates significant variation over time. Furthermore, patients with enteric HO had higher median uOx values at diagnosis and during follow-up, had more variability in uOx over time, were less likely to normalize uOx values, and, when uOx did normalize, took longer to do so compared with those with idiopathic (including dietary) HO.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Asplin

. Hyperoxaluric calcium nephrolithiasis. Endocrinol Metab Clin North Am, 2002; 31:927–949.

Marengo

, Romani

. Oxalate in renal stone disease: The terminal metabolite that just won't go away. Nat Clin Pract Nephrol, 2008; 4:368–377.

Curhan

, Willett

, Speizer

, Stampfer

. Twenty-four-hour urine chemistries and the risk of kidney stones among women and men. Kidney Int, 2001; 59:2290–2298.

Laminski

, Meyers

, Kruger

, et al. Hyperoxaluria in patients with recurrent calcium oxalate calculi: Dietary and other risk factors. Br J Urol, 1991; 68:454–458.

Holmes

, Goodman

, Assimos

. Contribution of dietary oxalate to urinary oxalate excretion. Kidney Int, 2001; 59:270–276.

Worcester

. Stones from bowel disease. Endocrinol Metab Clin North Am, 2002; 31:979–999.

Clark

, Vandorpe

, Chernova

, et al. Species differences in Cl- affinity and in electrogenicity of SLC26A6-mediated oxalate/Cl- exchange correlate with the distinct human and mouse susceptibilities to nephrolithiasis. J Physiol, 2008; 586:1291–1306.

Freel

, Hatch

, Green

, Soleimani

. Ileal oxalate absorption and urinary oxalate excretion are enhanced in Slc26a6 null mice. Am J Physiol Gastgrointest Liver Physiol, 2006; 290:G719–G728.

Siva

, Barrack

, Reddy

, et al. A critical analysis of the role of gut Oxalobacter formigenes in oxalate stone disease. BJU Int, 2009; 103:1821.

10.

Hatch

, Cornelius

, Allison

, et al. Oxalobacter sp. reduces urinary oxalate excretion by promoting enteric oxalate secretion. Kidney Int, 2006; 69:691–698.

11.

Parivar

, Low

, Stoller

. The influence of diet on urinary stone disease. J Urol, 1996; 155:432–440.

12.

Stauffer

. Hyperoxaluria and intestinal disease. The role of steatorrhea and dietary calcium in regulating intestinal oxalate absorption. Am J Dig Dis, 1977; 22:921–928.

13.

Hylander

, Jarnum

, Nielsen

. Calcium treatment of enteric hyperoxaluria after jejunoileal bypass for morbid obesity. Scand J Gastroenterol, 1980; 15:349–352.

14.

Ortiz-Alvarado

, Miyaoka

, Kriedberg

, et al. Pyridoxine and dietary counseling for the management of idiopathic hyperoxaluria in stone-forming patients. Urology, 2011; 77:1054–1058.

15.

Kaufman

, Kelly

, Curhan

, et al. Oxalobacter formigenes may reduce the risk of calcium oxalate kidney stones. J Am Soc Nephrol, 2008; 19:1197–1203.

16.

Borghi

, Nouvenne

, Meschi

. Probiotics and dietary manipulations in calcium oxalate nephrolithiasis: Two sides of the same coin?. Kidney Int, 2010; 78:1063–1065.

17.

Evan

, Lingeman

, Coe

, Worcester

. Randall's plaque: Pathogenesis and role in calcium oxalate nephrolithiasis. Kidney Int, 2006; 69:1313–1318.

18.

Evan

, Coe

, Lingeman

, et al. Mechanism of formation of human calcium oxalate renal stones on Randall's plaque. Anat Rec (Hoboken), 2007; 290:1315–1323.

19.

Williams

, Wandzilak

. Oxalate synthesis, transport and the hyperoxaluric syndromes. J Urol, 1989; 141:742–749.

20.

Penniston

, Nakada

. Effect of dietary changes on urinary oxalate excretion and calcium oxalate supersaturation in patients with hyperoxaluric stone formation. Urology, 2009; 73:484–489.

21.

Asplin

, Coe

. Hyperoxaluria in kidney stone formers treated with modern bariatric surgery. J Urol, 2007; 177:565–569.