Metabolic Evaluation of Urolithiasis and Obesity in a

Midwestern Pediatric Population

John T. Roddy, Anas I. Ghousheh, Melissa A. Christensen and Charles T. Durkee*
From the Medical College of Wisconsin (JTR) and Department of Urology, Children’s Hospital of Wisconsin (MAC, CTD),
Milwaukee, Wisconsin, and Al Khaldi Medical Plaza, Amman, Jordan (AIG)

Purpose: The incidence of urolithiasis has been proved to be increasing in the

Abbreviations
adult population, and evidence to date suggests that the same holds true for the
and Acronyms
pediatric population. While adult urolithiasis is clearly linked to obesity, studies
of pediatric patients have been less conclusive. We hypothesized that a popula- BMI ¼ body mass index
tion of otherwise healthy children with stones would have an increased body BMI% ¼ body mass index
mass index compared to a control population, and that obese pediatric stone percentile
formers would have results on metabolic assessment that are distinct from BSA ¼ body surface area
nonobese stone formers.
Accepted for publication September 27, 2013.
Materials and Methods: We retrospectively reviewed the charts of all patients
Study received Children’s Hospital of
10 to 17 years old with upper tract urolithiasis without comorbidities treated Wisconsin institutional review board approval.
between 2006 and 2011. Mean body mass index of our population was compared * Correspondence: Children’s Hospital of
Wisconsin, 999 N. 92nd St., Suite 330, Milwau-
to state data, and 24-hour urine collection results were compared between obese
kee, Wisconsin 53226 (telephone: 414-266-1656;
and nonobese patients with stones. FAX: 414-266-1752; e-mail: CDurkee@chw.org).
Results: The obesity rate in 117 patients with urolithiasis did not differ signifi-
See Editorial on page 579.
cantly from the obesity rate derived from the 2007 National Survey of Children’s
Health (observed/expected ratio 1.11, 95% CI 0.54e1.95). Using t-test and
chi-square comparisons, overall 24-hour urine collection data did not show sta-
tistically significant differences.
Conclusions: Our results do not confirm obesity as a risk factor for pediatric
urolithiasis in otherwise healthy patients. We also found no substantial meta-
bolic differences between healthy nonobese stone formers and obese patients.
While the pediatric literature is mixed, our study supports the majority of pub-
lished series that have failed to establish a link between pediatric urolithiasis
and obesity.

Key Words: obesity, pediatrics, urolithiasis

THE incidence of kidney stones is increases may exist because of better

increasing in adults and appears to
be increasing in children, although
the evidence is more limited in the
latter. Analysis of the Pediatric
Health Information System database
demonstrated an increase in admis-
sions for urolithiasis, from 13.9 cases
per hospital in 1999 to 32.6 cases in
2008.1 Single institutional reporting
has revealed the same trend.2,3 These
reporting, increased diagnosis due to
more frequent use of computerized
tomography or changes in referral
patterns. However, the contemporary
trend of increased incidence of stone
disease continues, a finding that
would be unexpected if driven by bias
alone.4 Given the physical costs
associated with this diagnosis, along
with the societal costs, targeting

0022-5347/14/1913-0771/0 http://dx.doi.org/10.1016/j.juro.2013.09.064
THE JOURNAL OF UROLOGY®
© 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
Vol. 191, 771-776, March 2014
Printed in U.S.A.
www.jurology.com j 771
772 UROLITHIASIS AND OBESITY IN MIDWESTERN PEDIATRIC POPULATION
appropriate risk factors for stone formation is
highly relevant.
A number of factors contribute to the increasing
incidence of urolithiasis in the adult population,
including a well-defined relationship between high
BMI and stone formation.5e7 While it is well
established that the prevalence of high BMI is
increasing in the pediatric population,8,9 the link
between obesity and pediatric stone disease is less
clear. We sought to find evidence of a relationship
with obesity in an otherwise healthy sample of
pediatric stone formers.
Others have examined this association previ-
ously. Conclusions have been mixed regarding a
correlation with BMI. Kim et al used age matched
controls and failed to show a statistical link.10 Two
additional studies using population matched con-
trols also failed to demonstrate an increased inci-
dence of obesity among stone formers.4,11 Conversely
urolithiasis was associated with higher odds of
obesity in a matched case control study using the
Pediatric Health Information System database.12
Similarly metabolic assessment, a key diagnostic
tool in preventing pediatric stone disease, has
resulted in differing conclusions regarding obese
vs nonobese patients. A 2009 California study of
43 children suggested that as calculated BMI
percentile increased, supersaturation of calcium
phosphate increased and urinary oxalate excretion
decreased, while urinary calcium, citrate and pH
were unchanged.13 However, a 2012 series revealed
only a decreased urinary citrate level when
comparing overweight to normal weight children.11
A study from Turkey in 2008 found that urinary
oxalate excretion increased and urinary citrate
decreased in children with urolithiasis and high
BMI compared to nonobese patients.14
Thus, the pediatric literature is limited and con-
tradictory regarding the link between high BMI and
risk of urolithiasis, as well as findings regarding
metabolic assessment. It is unclear whether the
disparities are related to differences in study design
or represent a true variation in regional findings.
Our study adds further knowledge in this area. Our
primary hypothesis is that otherwise healthy
pediatric patients with stones will have an
increased incidence of high BMI compared to con-
trols. Our secondary hypothesis is that obese pedi-
atric stone formers will have results on metabolic
assessment that are distinct from stone formers
with a normal BMI.
MATERIALS AND METHODS
This study consisted of a retrospective case-control chart
review of children presenting with urinary calculi at a
major tertiary care pediatric hospital and health system.
This series compares disease characteristics of urolithia-
sis in groups categorized as obese (case) and nonobese
(control). Comparisons of obesity rates were also made
between the case group of stone formers with data from
the general pediatric population in Wisconsin via the 2007
National Survey of Children’s Health.15 The study was
approved by the Children’s Hospital of Wisconsin insti-
tutional review board. Statistical calculations were per-
formed using MicrosoftÒ ExcelÒ 2007.
Selection Criteria
The target population of this study was chosen as a
representative sample of pediatric stone formers. Initial
inclusion criteria consisted of a diagnosis of “calculus of
kidney,” “calculus of ureter” or “urinary calculus, un-
specified” (ICD-9 codes 592.0, 592.1 and 592.9, respec-
tively) from 2006 to 2011. These diagnoses were confirmed
by the presence of radiographic evidence or stone analysis
in patient charts. A search for “uric acid calculus” di-
agnoses (ICD-9 code 274.11) was performed but yielded
no additional subjects for inclusion. Patient charts with
ICD-9 codes 788.0 and 789.0 (representing renal colic and
abdominal pain, respectively) were also examined to
ensure that no urolithiasis was missed. However, these
codes failed to supply additional subjects. Subjects from
birth to 18 years at initial presentation or with recurring
presentation between 2006 and 2011 were included. These
inclusion criteria yielded 299 patients with a history of
urolithiasis.
Diagnoses of lower urinary tract stones (urethral and
bladder stones) and congenital abnormalities of the
kidneys (ICD-9 codes 594.2, 594.8, 594.9 and 753.3) were
used as exclusion criteria to avoid confounding bias.
Any patient with a confirmed history of cystine stones,
spina bifida, cerebral palsy, extended immobility within
3 months of stone presentation, metabolic abnormality,
endocrine abnormality, rheumatological abnormality,
gastrointestinal abnormality, surgical genitourinary
reconstruction, chromosomal abnormality, profound
developmental delay, malignancy or neurogenic bladder
was excluded for the same reason. After excluding for
confounders 177 of the 299 subjects younger than 19
years remained.
Study subjects were further focused to most closely
mirror adults with urolithiasis because of known corre-
lations between obesity and stone formation in adults.
The final study population included 125 patients 10 to 17
years old. Demographic data characterizing study sub-
jects according to gender, race and age at presentation are
outlined in table 1.
Data Collection
Weight and height were recorded within 4 months before
or after initial or earliest documented stone presentation
within the study period. One exception was a subject with
no recorded height within the allotted time frame and
a recorded stable height outside the time frame. Body mass
index was calculated for each subject if height and weight
were recorded within acceptable time parameters using
BMI ¼ weight (kg)/height2 (m2). Body mass index percen-
tile was further calculated to account for the variation
observed in children of different ages and genders,10,16
 UROLITHIASIS AND OBESITY IN MIDWESTERN PEDIATRIC POPULATION 773

Table 1. Patient demographics

No. Obese No. Nonobese Totals

Pts (%) Pts (%) (%)
Gender:
Male 9 (81.8) 23 (31.5) 32 (38.1)
Female 2 (18.2) 50 (68.5) 52 (61.9)
Race:
NonHispanic white 8 (72.7) 62 (84.9) 70 (83.3)
NonHispanic black 0 (0) 5 (6.8) 5 (6.0)
Hispanic 2 (18.2) 3 (4.1) 5 (6.0)
Asian 0 (0) 1 (1.4) 1 (1.2)
Other 1 (9.1) 2 (2.7) 3 (3.6)
Age at earliest presentation (yrs):
10-13.9 5 (45.5) 29 (39.7) 34 (40.5)
14-17.9 6 (54.5) 44 (60.3) 50 (59.5)

using the Centers for Disease Control and Prevention

BMI% calculator.17 BMI at or above the 95th percentile
was categorized as obese. Data were available for BMI%
calculation in 84 subjects.
Stone analysis was recorded by percent composition.
Results of 24-hour urine collection chemical analyses were Comparison of obesity rate for stone study subjects and state
also recorded from patient charts. Of the 77 subjects with data derived from National Survey of Children’s Health15
demonstrates nonsignificant difference.
24-hour urine studies available 58 also had BMI% in the
chart, thus allowing for comparison in this subset. Data
such as recurrence, stone size, degree of family history,
gender, age, race, asthma diagnosis, prematurity and obese, with BMI greater than 35 and obesity related
vitamin supplementation were also recorded to assess complications. Average BMI of the obese subjects
correlation with BMI category in stone formers. was 31.2. However, categorization by BMI% is more
informative, since obesity is defined by age depen-
24-Hour Urine Chemistry Analysis dent BMI% in children.
Two laboratories were used for 24-hour urine chemistry Urine volume adjusted for weight and unadjusted
analysis during the study, resulting in different measures
24-hour uric acid levels showed significant differ-
for supersaturation indices for calcium phosphate, cal-
cium oxalate and uric acid. The normal ranges from both
ences between the study groups. On chi-square
analyses were used to place the respective variables into analysis there was no significant difference be-
categories that could be compared across all subjects. The tween obese and nonobese status in the proportion of
2 reported reference ranges were from Quest Diag- subjects with low urine output and hyperuricosuria
nosticsÒ and LitholinkÒ Laboratory Reporting SystemÔ. adjusted for creatinine, nor hyperoxaluria, hyper-
Abnormal values used to categorize variables for analysis calciuria, hypocitraturia or abnormal levels of su-
from 24-hour urine collection were urine volume less than persaturation of calcium phosphate, calcium oxalate
1.0 ml/kg per hour, calcium greater than 4.0 mg/kg per or uric acid (table 2). Several 24-hour urine factors
24 hours, citrate less than 180 mg/gm creatinine, uric acid displayed no significant differences between obese
greater than 815 mg/1.73 m2 BSA and oxalate greater and nonobese subjects on t-test, including calcium
than 52 mg/1.73 m2 BSA.4 BSA was calculated based on
adjusted for creatinine and weight, citrate, oxalate
weight and height using the Mosteller equation,18 BSA
approximately equals (height  weight/3,600)1/2.
Table 2. Multivariate analysis of potential urinary risk factors
for pediatric urolithiasis
RESULTS % Obese % Nonobese p Value
To compare obesity rates between the study group Abnormal 24-Hr Urine Factor Pts Pts (No.)*
and the population sample, 41 subjects without Low urine output 100.0 71.7 0.083 (61)
available BMI% data were excluded, yielding Hypercalciuria 0.0 24.5 0.115 (57)
Hypocitraturia 12.5 2.0 0.136 (57)
84 subjects in the study group, of whom 11 were Hyperuricosuria 4.1 0.0 0.586 (56)
obese. These subjects were compared to the 930 Hyperoxaluria 12.5 6.1 0.513 (57)
state survey responders with recorded BMI% Increased supersaturation CaOx 25.0 56.3 0.101 (56)
Decreased supersaturation CaOx 0.0 16.7 0.212 (56)
from the National Survey of Children’s Health. No Increased supersaturation CaP 62.5 46.8 0.412 (55)
significant difference was found between rates of Increased supersaturation uric acid 25.0 31.3 0.722 (56)
obesity in the study population and state norms
Patients with BMI percentile 95% or greater were considered obese, while those
(observed/expected ratio 1.11, 95% CI 0.54e1.95, see with BMI percentile 5% to less than 95% were considered nonobese.
figure). Of the 11 obese adolescents 3 were morbidly * Chi-square test.
774 UROLITHIASIS AND OBESITY IN MIDWESTERN PEDIATRIC POPULATION

and sodium adjusted for creatinine, potassium
adjusted for weight, specific gravity and pH (table 3).
Since not all subjects with available BMI% had
24-hour urine collection recorded, the sample size
in the analyses of urinary risk factors was smaller
and depended on which subjects had the relevant
data recorded.
We were interested in a number of factors
relating to subject presentation with urolithiasis,
disease progression, family history and vitamin
supplementation. Proportion of first degree family
history did not differ significantly between patients
with high body weight and normal BMI% (chi-
square[1, sample size 81] ¼ 0.03, p ¼ 0.85), nor for
stone size greater than 5 mm (chi-square[1, 81]
¼ 2.13, p ¼ 0.15), formation of multiple stones
(chi-square[1, 79] ¼ 1.72, p ¼ 0.19), recurrence (chi-
square[1, 81] ¼ 0.06, p ¼ 0.80) or number requiring
procedures (chi-square[1, 81] ¼ 0.04, p ¼ 0.85).
Factors such as asthma status (chi-square[1, sample
size 80] ¼ 0.84, p ¼ 0.36), prematurity (chi-square
[1, 75] ¼ 1.05, p ¼ 0.31) and vitamin supplementa-
tion (chi-square[1, 80] ¼ 0.14, p ¼ 0.71) likewise
did not differ significantly between obese and non-
obese subjects.
DISCUSSION
Our results do not demonstrate that obesity is a risk
factor for pediatric urolithiasis in otherwise healthy
patients. We also found no substantial metabolic
differences between healthy normal weight stone
formers and those with a high BMI. There is
conclusive evidence in adults of increased obesity,
increasing urolithiasis and, most importantly, a link
between adult obesity and stone disease. Therefore,
it is rational to attempt to find associations between
childhood obesity and increased stone disease,
although our study did not bear out this relationship.
Our findings are similar to other single institution
studies revealing no correlation between obesity and
stones.4,11 A population based pediatric study from
Minnesota showed no correlation compared to na-
tional norms, and no increase in obesity rates during
a 25-year period, while demonstrating a 6% annual
increase in the incidence of stones in children 12 to
17 years old.3 Another series using matched controls
across 30 pediatric cases revealed no association
between obesity and urolithiasis.10 Conversely
based on the Pediatric Health Information System
database, patients with urolithiasis are 30% more
likely to be obese compared to matched controls after
adjusting for confounding variables.12 However,
a study from the Kids’ Inpatient Database failed to
show any link between stones and obesity across all
4 age groups studied.19
Adults with a high BMI excrete more urinary
oxalate, uric acid, sodium and phosphorus, and have
lower urinary pH. These factors are all known to
be promoters of calcium oxalate urolithiasis.7
Increased uric acid excretion combined with low
pH leads to an increase in uric acid stones and
calcium oxalate stones through heterologous nucle-
ation.20 This activity occurs partly due to develop-
ment of insulin resistant type II diabetes mellitus
associated with obesity and the metabolic syn-
drome. Insulin resistance leads to urinary acidifi-
cation and hypercalciuria. Type II diabetes is also
rapidly increasing in the pediatric population,
although this constellation of findings is not present
in any of the metabolic urolithiasis studies in pedi-
atrics. However, a study of teenagers with proved
metabolic syndrome but no history of neph-
rolithiasis demonstrated reduced urine pH with an
increasing number of metabolic syndrome traits,
decreasing from a mean of 6.12 with no traits to 5.23
with 4 traits.21 The relative saturation ratio of cal-
cium oxalate also increased with increasing meta-
bolic syndrome traits in the study. The study group
had a mean BMI of 36 (severely obese).
Due to the known associations of urolithiasis and
metabolic profile in obese adults, we hypothesized
that there would also be unique metabolic attributes
in obese pediatric patients with stones. However, in

Table 3. Comparison of obese and nonobese stone subjects with available 24-hour urine collection data

Urinary Factor Obese Pts (mean  SD) Nonobese Pts (mean  SD) p Value (No.)*
Urine vol/wt (ml/kg) 16.4  3.7 20.3  9.3 0.044 (61)
Potassium/wt (mEq/kg) 10.6  5.9 13.5  5.0 0.23 (58)
Uric acid (gm) 0.68  0.20 0.28  0.13 0.037 (56)
Uric acid/creatinine (gm/gm) 0.44  0.18 0.43  0.12 0.98 (53)
Ca/creatinine (gm/gm) 148  68 154  64 0.84 (54)
Ca/wt (gm/kg) 2.68  0.76 3.15  1.3 0.17 (57)
Citrate/creatinine (mg/gm) 414  264 507  244 0.37 (57)
Oxalate/creatinine (mg/gm) 21.4  8.5 25.9  12 0.25 (54)
Na/creatinine (mEq/gm) 148  60 129  54 0.45 (53)
Urine specific gravity 1.023  0.009 1.022  0.012 0.85 (75)
pH 6.8  0.83 6.4  0.68 0.18 (77)

Patients with BMI percentile 95% or greater were considered obese, while those with BMI percentile 5% to less than 95% were considered nonobese.
* T-test. Although significant differences were noted in means between groups for 24-hour urine volume/weight and 24-hour uric acid, neither category displayed increased
rate of abnormally high values on chi-square analysis.
 UROLITHIASIS AND OBESITY IN MIDWESTERN PEDIATRIC POPULATION 775

our study no significant differences were found
within the measured metabolic variables between
patient groups using t-test and chi-square test.
Urine output per kilogram was less in the obese
group by t-test but output per kilogram in obese
patients is probably not the best measurement
of overall hydration. No difference was seen by
chi-square analysis, and specific gravity of urine,
perhaps a more accurate measurement of hydration
in otherwise healthy patients, was essentially
identical between groups.
Uric acid excretion was increased in the obese
group but only when measured on a total gram
basis. No differences were seen when corrected for
weight by uric acid/creatinine ratio and by chi-
square testing. Urine pH was 6.8 in obese patients
vs 6.4 in normal weight patients. This nonacidic pH
makes it less likely that heterologous nucleation is
occurring. Stone analysis was performed by Fourier
transform spectroscopy, a method that would detect
the presence of uric acid either as a pure stone or
as a stone nucleus, and no uric acid elements were
identified. More well-known risks for stone disease,
including citrate, calcium and oxalate excretion,
revealed no differences between groups.
Metabolic studies from other research groups
have failed to show consistent findings between
obese and nonobese patients. Stratifying pediatric
patients with nephrolithiasis into 4 groups based
on BMI, Eisner et al reported supersaturation of
calcium phosphate as the only positive correlation
with increasing BMI.13 When reviewing the data,
the supersaturation value was actually less in the
fourth quartile of BMI as opposed to the second and
third quartiles, raising the question of why the
trend did not hold as BMI increased to the most
overweight group. Our data did not demonstrate
a significant difference between obese and non-
obese patients.
A Turkish study compared 24-hour urine chem-
istry analyses of obese and nonobese patients.14 The
authors found increased oxalate excretion in the
obese group, in contrast to the findings of Eisner
et al.13 Our levels revealed no statistical difference.
The same series also found hypocitraturia in the
obese patients, while our series and that of Eisner
et al did not show a difference. The reason for the
variation in results of these studies is unknown but
obesity in and of itself clearly does not result in
predictable metabolic changes in the urine of pedi-
atric stone formers.
The limitations of our study include its retro-
spective nature and missing data points for some
patients, which could introduce unintentional bias.
The relatively low incidence of pediatric stone dis-
ease was a limitation to our study, as it decreased
the power achievable within the context of a single
institution study. Matched controls may be a more
accurate comparison group than population based
controls. We are currently enrolling our patients in
a prospective registry to address some of these
limitations, and we plan to look more closely at the
regional incidence of pediatric stone disease.
CONCLUSIONS
Urolithiasis in pediatric patients is increasing, as is
obesity. However, the 2 conditions do not appear to
be directly linked, and other factors are likely
driving the trend. Future studies are needed to find
a more global explanation for the increase in pedi-
atric stone disease.

REFERENCES
1. Routh JC, Graham DA and Nelson CP:
Epidemiological trends in pediatric urolithiasis
at United States freestanding pediatric hospi-
tals. J Urol 2010; 184: 1100.
2. VanDervoort K, Wiesen J, Frank R et al:
Urolithiasis in pediatric patients: a single center
study of incidence, clinical presentation and
outcome. J Urol 2007; 177: 2300.
3. Dwyer ME, Krambeck AE, Bergstralh EJ et al:
Temporal trends in incidence of kidney stones
among children: a 25-year population based
study. J Urol 2012; 188: 247.
4. Penido MG, Srivastava T and Alon US: Pediatric
primary urolithiasis: 12-year experience in
a Midwestern children’s hospital. J Urol 2013;
189: 1493.
5. Taylor EN, Stampfer MJ and Curhan GC: Obesity,
weight gain, and the risk of kidney stones.
JAMA 2005; 293: 455.
6. Semins MJ, Shore AD, Makary MA et al: The
association of increasing body mass index and
kidney stone disease. J Urol 2010; 183: 571.
7. Siener R, Glatz S, Nicolay C et al: The role of
overweight and obesity in calcium oxalate stone
formation. Obes Res 2004; 12: 106.
8. Barlow SE: Expert committee recommendations
regarding the prevention, assessment, and
treatment of child and adolescent overweight
and obesity: summary report. Pediatrics, suppl.,
2007; 120: S164.
9. National Center for Chronic Disease Prevention
and Health Promotion: Obesity: Halting the
Epidemic by Making Health Easier. At a
Glance 2011. Available at http://www.cdc.gov/
chronicdisease/resources/publications/aag/pdf/
2011/Obesity_AAG_WEB_508.pdf. Accessed July
3, 2012.
10. Kim SS, Luan X, Canning DA et al: Association
between body mass index and urolithiasis in
children. J Urol, suppl., 2011; 186: 1734.
11. Fox J and Ost M: Is BMI associated with stone
formation in children? Single center, prospective
analysis. J Urol, suppl., 2012; 187: e340, abstract
833.
12. Kokorowski PJ, Routh JC, Hubert KC et al:
Association of urolithiasis with systemic condi-
tions among pediatric patients at children’s
hospitals. J Urol, suppl., 2012; 188: 1618.
13. Eisner BH, Eisenberg ML and Stoller ML:
Influence of body mass index on quantitative
24-hour urine chemistry studies in children
with nephrolithiasis. J Urol 2009; 182: 1142.
776 UROLITHIASIS AND OBESITY IN MIDWESTERN PEDIATRIC POPULATION
14. Sarica K, Eryildirim B, Yencilek F et al: Role of
overweight status on stone-forming risk factors
in children: a prospective study. Urology 2009;
73: 1003.
15. Data Resource Center for Child and Adolescent
Health: 2007 Childhood Obesity State Report
Cards. Available at http://www.childhealthdata.
org/browse/snapshots/obesity-2007. Accessed
July 15, 2013.
16. Kieran K, Giel DW, Morris BJ et al: Pediatric
urolithiasisddoes body mass index influence
stone presentation and treatment? J Urol, suppl.,
2010; 184: 1810.
17. Centers for Disease Control and Prevention:
Children’s BMI Tool for Schools. Available at
http://www.cdc.gov/healthyweight/assessing/bmi/
childrens_bmi/tool_for_schools.html. Accessed
July 5, 2012.
18. El Edelbi R, Lindemalm S and Eksborg S:
Estimation of body surface area in various
childhood agesdvalidation of the Mosteller
formula. Acta Paediatr 2012; 101: 540.
19. Schaffer AJ, Feng Z, Trock BJ et al: Medical
comorbidities associated with pediatric kidney
stone disease. Urology 2011; 77: 195.
20. Grover PK and Ryall RL: Urate and calcium
oxalate stones: from repute to rhetoric to reality.
Miner Electrolyte Metab 1994; 20: 361.
21. Tiwari R, Campfield T, Wittcopp C et al:
Metabolic syndrome in obese adolescents is
associated with risk for nephrolithiasis. J Pediatr
2012; 160: 615.