Predicting Patients with Inadequate 24- or 48-Hour Urine Collections at Time of Metabolic Stone Evaluation

Abstract

Purpose:

We aimed to understand the characteristics of patients who are less likely to submit adequate urine collections at metabolic stone evaluation.

Methods:

Inadequate urine collection was defined using two definitions: (1) Reference ranges for 24-hour creatinine/kilogram (Cr/24) and (2) discrepancy in total 24-hour urine Cr between 24-hour urine collections. There were 1502 patients with ≥1 kidney stone between 1998 and 2014 who performed a 24- or 48-hour urine collection at Northwestern Memorial Hospital and who were identified retrospectively. Multivariate analysis was performed to analyze predictor variables for adequate urine collection.

Results:

A total of 2852 urine collections were analyzed. Mean age for males was 54.4 years (range 17–86), and for females was 50.2 years (range 8–90). One patient in the study was younger than 17 years old. (1) Analysis based on the Cr 24/kg definition: There were 50.7% of patients who supplied an inadequate sample. Females were nearly 50% less likely to supply an adequate sample compared with men, P<0.001. Diabetes (odds ratio [OR] 1.42 [1.04–1.94], P=0.026) and vitamin D supplementation (OR 0.64 [0.43–0.95], P=0.028) predicted receiving an adequate/inadequate sample, respectively. (2) Analysis based on differences between total urinary Cr: The model was stratified based on percentage differences between samples up to 50%. At 10%, 20%, 30%, 40%, and 50% differences, inadequate collections were achieved in 82.8%, 66.9%, 51.7%, 38.5%, and 26.4% of patients, respectively. Statistical significance was observed based on differences of ≥40%, and this was defined as the threshold for an inadequate sample. Female sex (OR 0.73 [0.54–0.98], P=0.037) predicted supplying inadequate samples. Adequate collections were more likely to be received on a Sunday (OR 1.6 [1.03–2.58], P=0.038) and by sedentary workers (OR 2.3 [1.12–4.72], P=0.023).

Conclusion:

Urine collections from patients during metabolic evaluation for nephrolithiasis may be considered inadequate based on two commonly used clinical definitions. This may have therapeutic or economic ramifications and the propensity for females to supply inadequate samples should be investigated further.

Introduction

Kidney stones are common, increasing in incidence, and have a high recurrence rate.¹ Urinary evaluation in the assessment of stone-forming patients can identify factors that can be manipulated to prevent stone formation. American Association of Urology (AUA) guidelines recommend “either one or two 24-hour urine collections in high-risk, motivated, first-time stone formers or recurrent stone formers.” This recommendation, however, is not universally agreed on.^1,2 Some argue that a single collection is adequate for diagnostic purposes, cost-effective, and more convenient for patients.³ Others argue that multiple urine collections improve diagnosis and help accurately direct therapy.^4
–6

Although patients are usually given verbal and/or written instructions on proper collection of the urine sample, clinicians can assess adequacy of collection when interpreting the results through two useful methods. The first method relies on creatinine per kilogram (Cr/24), which is the total Cr excreted in 24 hours divided by the weight of the patient. This parameter is easy to interpret because there are recommended reference ranges, although the reference ranges are largely arbitrary and not universally agreed on, as is the case with other urinary values.⁷ A high value can indicate overcollection; a low value can indicate undercollection, and false positives can occur in patients with high (e.g., muscular athletes) or low muscle mass (e.g., spina bifida).

A second method relies on comparing the total Cr excreted in a 24-hour period to previous values and can be a good indicator of a correctly collected sample. For example, when examining consecutive total urinary Cr values for variation, we would normally take an arbitrary value of >20% of the baseline Cr to indicate that the sample may be inadequate. Clinically, this is a more reliable method compared with Cr/24, although “normal” values have never been demonstrated and, as such, there is no reference range to assist the clinician.

In practice, 24- and 48-hour urine collections are often evaluated for adequacy based on the above definitions, yet no clinically defined reference range for 24-hour adequacy currently exists. This practice can have important implications. Based on the adequacy of the collection, therapeutic pharmacologic interventions may be initiated or, alternatively, the entire result of a collection may be ignored because it is deemed to be inadequately collected. Furthermore, clinicians often compare the total 24-hour urine Cr values between samples as a benchmark for sample adequacy. Samples may be disregarded based on low arbitrary cutoffs (e.g., 10%–20%). In fact, our study suggests that larger differences in total Cr may actually be acceptable. Identifying patients more likely to submit inadequate samples could help clinicians target them for education to increase compliance and improve the quality and quantity of samples received.

In this study, we aimed to describe the demographics of the patients who supply inadequate samples and attempt to define a clinically useful threshold for describing adequate samples.

Methods

The Institutional Review Board for Northwestern University approved this retrospective study. Patients presenting to a tertiary metabolic stone evaluation clinic between 1998 and 2014, with a history of ≥1 stone were assessed by serum and 24- or 48-hour urine collection, evaluated by Litholink^® Corporation. Patients submitted one or two separate 24-hour urine collections performed while consuming a regular diet and without any particular advice with regard to fluids. The serum and urinary results of these patients were collected for analysis. The results of serum (Cr, Cr clearance), and 48-hour urine collection were analyzed. Body mass index (BMI), sex, age, race, and median household income were the primary anthropologic variables.

Sample adequacy was assessed in two different ways. First, sample adequacy was assessed by examining patient results for abnormal Litholink values for the 24-hour Cr/24—i.e., those above or below the “normal” values described by Litholink. An adequate sample for men was defined as Cr/24 within the range of 18 to 24 mg/kg/day and women between 15 to 20 mg/kg/day. The second method for assessing sample adequacy considered the discrepancy in total 24-hour urine Cr between two 24-hour urine collections. Adequacy was assessed at 10%, 20%, 30%, 40%, and 50% differences between the 24-hour Cr values, such that any two samples with a difference greater than any one of the cutoffs described above would be considered inadequate.

In addition, we did not want patients with chronic kidney disease (CKD) to skew data because they are more likely to have discrepant urinary volume compared with the normal population, as well as increased variability in urinary metabolites, especially if receiving medication such as diuretics. As such, independent analysis was performed excluding patients with stage three, four, or five CKD to identify potential confounding variables. CKD stage was calculated using the Cockcroft-Gault equation and defined as follows: Stage 1: glomerular filtration rate (GFR) ≥90 mL/min; stage 2: 60≤GFR–89 mL/min; stage 3: 30≤GFR–59 mL/min; stage 4: 15≤GFR–29 mL/min; and stage 5: GFR≤15 mL/min.

Values were assessed using a variety of statistical methods including Pearson correlation coefficient and Pearson chi-square. Frequencies and proportions were generated for categoric variables, and the chi-square test was used to assess the statistical significance of differences in populations and odds ratios (ORs) of sample adequacy. Means, medians, and ranges were computed for continuously coded variables. A binary logistic regression model was computed for the prediction of sample adequacy based on the covariates of sex, diabetes, vitamin D supplementation, urine collection day, occupation, and smoking status. Statistical significance was set at P<0.05. Analyses were conducted using SPSS^® version 22.

Results

Demographic data

A total of 2852 urine collections in 1502 patients were analyzed, 60% (898) of which were supplied by males and 40% (604) by females. The mean age for males was 54.4 years (range 17–86) and for females was 50.2 years (range 8–90). The majority of patients were middle-aged (mean age 52.7±14.3 standard deviation [SD]), overweight (mean BMI 28.7 kg/m²±7.4 SD), white (54.4%), and sedentary workers (96%) (Table 1). Demographic variables and stone-forming risk factors in the study population are listed in Table 1.

Table 1.

Select Demographics of the Study Population

Study Population
% Male (n)	60% (898)	% Female (n)	40% (604)
Mean age (years, min–max)	54.4 (17–86)		50.2 (8–90)
Mean BMI (kg/m², min–max)	28.7 (13.6–69.0)		28.3 (15.3–76.2)
	Minimum	Maximum	Mean	SD
BMI (kg/m²)	13.6	76.2	28.7	7.4
Age (years)	8.0	90.0	52.7	14.3
Median household income by ZIP code (US dollars)	$19,623	$229,792	$71,837	$26,058
	White	Black	Asian	Hispanic	NHL
Race	54.4% (819)	5.9% (77)	2.6% (39)	4.7% (71)	20.1% (302)
Occupation: % Sedentary (n)	96.0% (1,214)	Occupation: % Manual (n)	4.0% (50)
	All	% of males	% of females	OR (95% CI)	P value
Smoking status	36.0% (445)	37.8%	32.3%	0.79 (0.62–1.00)	0.047
Vitamin D supplementation	13.1% (197)	9.1%	20.5%	2.60 (1.90–3.50)	0.000
Diabetes	13.0% (190)	13.9%	12.1%	0.85 (0.62–1.20)	0.325
Calcium supplementation	7.2% (108)	4.1%	12.6%	3.40 (2.20–5.10)	0.000
Gout or allopurinol use	6.5% (94)	9.0%	2.9%	0.31 (0.20–0.52)	0.000
Bowel disorders^*	4.4% (67)	3.8%	5.5%	^*	^*
Hyperparathyroidism	2.8% (41)	1.8%	4.5%	2.60 (1.40–5.00)	0.002
Bariatric surgery	1.0% (15)	0.6%	1.7%	3.00 (1.00–8.80)	0.037
Topiramate use	0.8% (12)	0.4%	1.6%	4.50 (1.20–16.60)	0.014
Sarcoid	0.4% (6)	0.6%	0.2%	0.29 (0.03–2.52)	0.235
Renal tubular acidosis	0.3% (5)	0.1%	0.7%	5.90 (0.66–53.20)	0.071
Cystinuria	0.3% (4)	0.2%	0.3%	1.50 (0.21–10.50)	0.695
Primary hyperoxaluria	0.1% (2)	0.2%	0.0%	0.60 (0.57–0.62)	0.244

Bowel disorders=Crohn disease, ulcerative colitis, chronic diarrhea, short gut syndrome, or history of bowel resection.

BMI=body mass index; SD=standard deviation, NHL=Not Hispanic or Latino.

Analysis based on the Cr/24 definition

In total, more than half (50.7%) of all patients supplied an inadequate sample. Stratified by sex, females more commonly supplied inadequate samples (60.3% females vs 44.8% males, P<0.001). Excluding patients with advanced CKD did not considerably alter this outcome (58.3% females vs 44.6% males, P<0.001) (Table 2B). The demographics of these patients stratified based on age, BMI, and CKD≥stage 3 are described in Table 2A.

Table 2A

Sex Differences in Adequacy of Urine Collections Submitted for Analysis

	Sample inadequacy based on urine collection normal range 24-hour creatinine per kilogram
	Total	Male (18–24 mg/kg/d)	Female (15–20 mg/kg/d)	OR (95% CI)	P value
All patients	50.7% (761/1502)	44.8% (398)	60.3% (363)	0.53 (0.43–0.66)	0.000
Mean age (years)	52.4±14.7	55.0±13.2	50.1±15.9		0.017
Mean BMI (kg/m²)	28.7±7.6	28.9±6.0	28.4±8.9		0.000
Excluding CKD stages ≥3 Patients	48.0% (465/968)	44.6% (307)	58.3% (158)	0.58 (0.43–0.76)	0.000
Mean age (years)	50.6±14.9	53.3±13.3	45.2±16.5		0.000
Mean BMI (kg/m²)	28.8±7.3	28.9±6.1	28.5±9.2		0.000

OR=odds ratio; CI=confidence interval; CKD=chronic kidney disease.

Table 2B

Univariate and Multivariate Analysis of Predictor Variables of Adequate Samples Based on Urine Collection Normal Ranges for Cr/24 for All Patients Submitting Either 24– or 48–Hour Urine Collections (n=1502; Excluding CKD Patients, n=968)

Predictors of adequate urine collections based on normal range for 24-hour creatinine per kilogram
	Univariate		Multivariate
Predictor variable	OR (95% CI)	P value	OR (95% CI)	P value
Female sex	0.53 (0.43–0.66)	0.000	0.57 (0.45–0.72)	0.000
Excluding CKD stages ≥3	0.58 (0.43–0.76)	0.000	0.53 (0.42–0.67)	0.010
Diabetes	1.42 (1.04–1.94)	0.026	1.33 (0.95–1.85)	0.098
Excluding CKD stages ≥3	1.45 (0.97–2.15)	0.069	1.41 (1.02–1.94)	0.037
Vitamin D	0.64 (0.43–0.95)	0.028	0.75 (0.54–1.04)	0.088
Excluding CKD stages ≥3	0.64 (0.43–0.95)	0.028	0.80 (0.58–1.10)	0.049
Urine collection day (Sunday)	–	0.508	1.30 (0.84–1.90)	0.703
Excluding CKD stages ≥3	–	0.508	0.89 (0.54–1.50)	0.269
Occupation: sedentary	0.65 (0.36–1.17)	0.147	0.60 (0.32–1.09)	0.093
Excluding CKD stages ≥3	0.59 (0.28–1.25)	0.167	0.55 (0.29–1.05)	0.068
Smoker	0.96 (0.76–1.21)	0.735	0.94 (0.72–1.21)	0.620
Excluding CKD stages ≥3	0.94 (0.70–1.25)	0.659	0.93 (0.72–1.21)	0.598

Univariate and multivariate analysis was performed to assess if predictors of an adequate sample could be identified. Females were nearly 50% less likely to supply an adequate sample compared with men on univariate and multivariate models, P<0.001 (Table 2B). A history of diabetes (OR 1.42 [1.04–1.94], P=0.026, Table 2B) predicted supplying an adequate sample, and this result remained significant on multivariable analysis when excluding patients with end-stage kidney disease (OR 1.41 [1.02–1.94] P=0.037, Table 2B). Vitamin D supplementation (OR 0.64 [0.43–0.95], P=0.028, Table 2B) was a significant predictor of inadequate urine collection.

Separate univariate and multivariate analysis was performed based on sex. In males, only diabetes was a significant predictor of an adequate urine collection on univariate (OR 1.65 [1.10–2.50], P=0.016) and multivariate analysis (OR 1.60 [1.0–2.5], P=0.047); Supplementary Tables 1A and B; supplementary data are available online at www.liebertpub.com/end. No other factor remained significant for males. In females, when excluding patients with end-stage renal disease, vitamin D supplementation remained the only significant predictor for inadequate urine collection (OR 0.35 [0.16–0.76], P=0.008); Supplementary Table 1B; supplementary data are available online at www.liebertpub.com/end.

Analysis based on differences between total urinary Cr

The predictive model was stratified based on percentage differences between samples ranging from 10% to 50% (Table 3A). If a 10% difference were used to predict an inadequate sample, the majority of samples supplied would be considered inadequate (82.2%). With 20%, 30%, 40%, and 50% differences, inadequate collections would be achieved in 66.9%, 51.7%, 38.5%, and 26.4% of patients, respectively. We observed statistical significance in the model based on differences of greater than 40% only (Supplementary Table 1; supplementary data are available online at www.liebertpub.com/end) and defined this as the threshold for an inadequate sample because univariate and multivariate analysis consistently demonstrated statistically significant differences in sample adequacy based on sex, vitamin D supplementation, occupation, and diabetes status at the 40% difference level for 24-hour urine Cr between urine collections.

Table 3A

Description of Patients Submitting an Inadequate Urine Collection Based on the Percentage Difference in 24-Hour Creatinine per Kilogram Between Two 24-Hour Urine Collections ^*

	Sample inadequacy based on percentage difference in 24-hour creatinine between two urine collections
Cutoff (% difference in Cr 24)	10% difference	20% difference	30% difference	40% difference	50% difference	Total
All patients (% male/% female)	82.8% (58.2/41.8)	66.9% (59.0/41.0)	51.7% (59.0/41.0)	38.5% (56.2/43.8)	26.4% (55.8/44.2)	1,181 (698/482)
Mean age (years) (Male/female)	52.7±14.4 (54.5/50.3)	52.5±14.6 (54.1/50.2)	52.3±14.8 (54.2/49.5)	52.4±14.2 (54.3/50.3)	51.7±13.9 (53.2/49.7)	52.7±14.3 (54.4/50.2)
Mean BMI (kg/m²) (Male/female)	28.6±7.3 (29.1/27.9)	28.8±7.3 (29.4/28.0)	28.7±7.3 (29.2/28.0)	28.6±7.3 (29.2/27.8)	28.5±7.2 (29.1/27.7)	28.6±7.5 (29.0/28.1)
Excluding CKD Sstages ≥3 patients (% Male/% female)	83% (67.4/32.6)	67.3% (68.1/31.9)	53.3% (67.5/32.5)	40.1% (63.5/36.5)	27.8% (62.7/37.3)	820 (555/265)
Mean age (years) (Male/female)	51.0±14.7 (53.3/46.3)	50.7±14.9 (53.0/45.8)	50.3±15.1 (53.0/45.0)	50.7±14.4 (53.0/46.6)	50.0±14.4 (52.1/46.6)	51.2±14.6 (53.4/46.5)
Mean BMI (kg/m²) (Male/Female)	28.7±7.3 (29.2/27.5)	29±7.3 (29.5/27.7)	28.7±7.1 (29.4/27.3)	28.7±7.1 (29.5/27.2)	28.8±7.3 (29.3/27.8)	28.6±7.2 (29.2/27.3)

“Total” represents all patients who submitted two 24-hour urine collections. For example, at “50%,” two urine collections submitted by a single patient that differ by more than 50% for 24-hour creatinine per kilogram are considered inadequate. See Methods for additional explanation of the methodology.

Cr 24=24-hour creatinine per kilogram.

Univariate and multivariate analysis was performed on this cohort with a greater than 40% difference between total urinary Cr (Table 3B). Female sex was more likely to supply an inadequate collection only when those with advanced CKD were excluded on both univariate (OR 0.73 [0.54–0.98], P=0.037, Table 3B) and multivariate analysis (OR 0.72 [0.53–0.98], P=0.041, Table 3B). In addition, patients who performed the collection on a Sunday were more than 1.5 times more likely to supply an adequate collection (OR 1.6 [1.03–2.58], P=0.038, Table 3B) in the multivariate model only. Sedentary workers were twice as likely to supply an adequate collection compared with manual workers (OR 2.3 [1.12–4.72], P=0.023, Table 3B).

Table 3B

Univariate and Multivariate Analysis of Predictor Variables of Adequate Samples Based on a 40% Difference (>40% Difference Between Samples Considered Inadequate) in 24-Hour Creatinine per Kilogram Between Litholink® Samples (n=1181; Excluding CKD Patients, n=820)

Predictors of adequate urine collection based on percentage difference in 24-hour creatinine between collections
	Univariate		Multivariate
Predictor variable	OR (95% CI)	P value	OR (95% CI)	P value
Female sex	0.82 (0.65–1.04)	0.099	0.81 (0.63–1.03)	0.086
Excluding CKD stages ≥3	0.73 (0.54–0.98)	0.037	0.72 (0.53–0.98)	0.041
Diabetes	1.00 (0.69–1.42)	0.958	0.88 (0.59–1.30)	0.550
Excluding CKD stages ≥3	1.08 (0.69–1.69)	0.738	0.96 (0.58–1.60)	0.888
Vitamin D	0.94 (0.67–1.32)	0.701	0.90 (0.61–1.33)	0.606
Excluding CKD stages ≥3	0.94 (0.61–1.50)	0.791	0.99 (0.61–1.60)	0.966
Urine collection day (Sunday)	–	0.487	1.60 (1.03–2.58)	0.038
Excluding CKD stages ≥3	–	0.487	2.10 (1.20–3.80)	0.008
Occupation: sedentary	2.20 (1.10–4.30)	0.018	2.30 (1.12–4.72)	0.023
Excluding CKD stages ≥3	1.90 (0.92–4.10)	0.076	2.10 (0.94–4.60)	0.071
Smoker	1.30 (0.97–1.70)	0.079	1.22 (0.91–1.65)	0.182
Excluding CKD stages ≥3	1.30 (0.97–1.80)	0.075	1.30 (0.95–1.90)	0.099

Discussion

We attempted to define the type of patient who supplies an inadequate urine collection based on two common clinically useful parameters for interpreting 24- and 48- hour urine collections in stone-forming patients. In addition, we have attempted to define cutoff values of a commonly used clinical technique in which values of 24-hour Cr are compared against each other to assess adequacy of collection. Our data reveal that a majority of patients submit inadequate urine collections regardless of which definition is used, and this was especially true in females.

Other predictive factors of an inadequate collection were a history of vitamin D supplementation and an occupation entailing manual labor. While it is not entirely clear why patients who supplement with vitamin D are more likely to provide an inadequate sample, we hypothesize that older patients at higher risk of osteoporosis or with a diagnosis of brittle bone disease and associated conditions such as osteoarthritis or rheumatoid arthritis are more likely to have a higher rate of vitamin D supplementation to avoid or at least slow the progression of osteoporosis. These conditions would almost certainly cause these patients to have a more difficult time properly collecting urine because of joint pain, stiffness, and discomfort.

Alternatively, predictive factors for submitting an adequately collected sample were a history of diabetes and if the sample was collected on a Sunday. It comes at no surprise that diabetics are subject to a battery of urine collections to measure glucose, ketones, and protein for diagnostic purposes when undergoing a workup for diabetes, and it is plausible that these patients are well practiced in the urine collection procedure. Furthermore, many patients with diabetes continue to submit regular urine collections to monitor kidney function, test for albuminuria, and to monitor progression of diabetes. Therefore, given that patients with diabetes have more experience collecting urine samples than other types of patients with kidney stones, it is expected that they would be more likely to submit a properly collected urine sample.

If adequacy of collection is assessed by comparison of 24-hour urinary Cr, our data suggest that the percentage difference is actually quite high at 40% (if based on the presence of significant predictive variables in the study population). Herein lies the value of our results; based on this study, physicians may interpret differences in total 24-hour Cr between samples with greater caution, because large differences may in fact be acceptable. In addition, certain patient groups may be targeted for extra education to improve sample collection.

While other studies have examined the adequacy of urine collection in stone-forming patients, they have been based solely on Cr/24.⁸ As shown in our data as well, Sawyer and associates⁸ reported a 50% inadequate urine collection rate in their cohort. Although commonly used in clinical practice, the Cr/24kg definition is based on individual laboratory definitions, and reference ranges vary.^8,9 The results of this test are influenced by many factors such as BMI, muscle mass, and age.⁸ Given these variables and the large number of inadequate collections, it has been suggested that the range be reconsidered. In fact, changing the reference range has important clinical consequences. For instance, there can be as much as a 63% difference between the median values for calcium concentration depending on the definition.⁸

Based on our own clinical experience, it would seem very unlikely that 50% of patients are supplying inadequate samples at the time of metabolic evaluation. In our opinion, it is highly unlikely that 50% of our patients fulfill the criteria that would bias the Cr/24kg results and suggests that the reference range does not correctly represent the modern stone-forming patient. Based on the high proportion of patients excluded using this reference range in both our data and that of Sawyer and associates,⁸ it is possible that many patients in other institutions are being excluded incorrectly if the physician accepts that the reference range for Cr/24 is correct. The clinical ramifications of this include the fact that many unnecessary repeated urine evaluations may be ordered, as well as the delay of appropriate dietary or pharmacologic interventions, or indeed potentially initiating pharmacologic therapy based on insufficient data.

Difference in total 24-hour urinary Cr as an alternative means for assessing adequacy of urine collection has not been studied previously and is a novel clinical tool in this setting. It is commonly used in clinical practice, yet no guideline or reference exists to aid the clinician. We wanted to examine this method of analysis, because in our own practice, we would arbitrarily take a difference of anything between 20% to 30% between samples as an indication for an inadequately supplied sample. Of course, when examining the 24-hour urinary Cr result in a patient who has supplied many collections over time, it is obvious if one of the samples is clearly different than previous samples and can be easily identified. If a clinician, however, is comparing two or three 24-hour collections, this is much more difficult.

We examined the difference in total Cr, and it was interesting to note how many patients have differences between samples of 10% (82% of patients) and even up to 50% (25% of patients). Based on our analysis, 40% discordance in 24-hour urinary Cr is a better means for assessing sample inadequacy compared with Cr/24; however, this conclusion is drawn tentatively because this is based on significance attained on predicting variables only.

Univariate and multivariate analysis was conducted to assess predictor variables of inadequate urine collections at 10%, 20%, 30%, 40%, and 50% difference between 24-hour urine Cr values. More specifically, each of the above-mentioned percentages served as a cutoff for adequacy such that if two urine collections had a discrepancy greater than the percentage cutoff, then the urine collection was considered inadequate. The five percentage cutoffs mentioned above were each tested, and a logistic regression model showed statistical significance at the 40% cutoff only. This is an important finding because there are no well-established reference ranges for sample adequacy based on 24-hour urine Cr even though this measure is often used in clinical practice. This also indicates that a single 24-hour urine collection may be an unreliable tool for metabolic stone evaluation, and that multiple 24-hour urine collections may aid in accurate evaluation.

In both methods, female sex is predictive of providing an inadequate collection. Females may be more likely to provide an inadequate sample because of anatomic and mechanical challenges during urine collection. To our knowledge, this has not been examined before and will be a future focus of our research. It is intuitive that a urine sample collected on a Sunday will more likely be adequate because patients are not working and can remain at home with the collection container. This, of course, presents its own issues, because the sample may not be representative of the patient's normal diet and urinary metabolites that occur on the other six days of the week. It is also intuitive that a laborer would not supply an adequate sample because of the difficulty of collecting urine for 24-hours with the limitations of being outside, and these patients could be targeted for special education.

Limitations of our data include the retrospective nature of the analysis, the possibility of a mechanical rather than physiologic confounding factor—e.g., males vs female anatomic factors or another unidentified confounding variable—and the lack of data regarding incontinence, especially in women who are more often affected, in the study sample. Patients with any extent of incontinence will presumably submit urine collections that underestimate total volume and certain urinary values such as 24-hour Cr. Without fully knowing the extent of the incontinence, however, it is impossible to predict what the effect of this condition would have on the predictors used in our model, but this is an area of future study. Finally, expanding the study to include additional institutions and geographies beyond the midwestern United States would help to further validate our findings.

Conclusion

Many of the urine collections performed on patients during metabolic workup for nephrolithiasis may be considered inadequate based on two commonly used clinical definitions. This may have therapeutic or economic ramifications and, furthermore, the propensity for females supplying inadequate samples should be investigated.

Footnotes

Acknowledgment

The authors would like to thank Dr. John Asplin from Litholink^® Corporation for advice received during this study.

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Pearle

, Goldfarb

, Assimos

, et al. Medical management of kidney stones: AUA guideline. J Urol, 2014; 192:316–324.

Castle

, Cooperberg

, Sadetsky

, et al. Adequacy of a single 24-hour urine collection for metabolic evaluation of recurrent nephrolithiasis. J Urol, 2010; 184:579–583.

Pak

, Peterson

, Poindexter

. Adequacy of a single stone risk analysis in the medical evaluation of urolithiasis. J Urol, 2001; 165:378–381.

Parks

, Goldfisher

, Asplin

, Coe

. A single 24-hour urine collection is inadequate for the medical evaluation of nephrolithiasis. J Urol, 2002; 167:1607–1612.

Nayan

, Elkoushy

, Andonian

. Variations between two 24-hour urine collections in patients presenting to a tertiary stone clinic. Can Urol Assoc, 2012; 6:30–33.

Healy

, Hubosky

, Bagley

. 24-hour urine collection in the metabolic evaluation of stone formers: Is one study adequate?. J Endourol, 2013; 27:374–378.

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.

Sawyer

, Dietrich

, Pickens

, et al. Adequate or not? A comparison of 24-hour urine studies for renal stone prevention by creatinine to weight ratio. J Endourol, 2013; 27:366–369.

Semins

, Matlaga

. Medical evaluation and management of urolithiasis. Ther Adv Urol, 2010; 2:3–9.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB