Professional Documents
Culture Documents
net/publication/340911888
The Validity of Urine Color as a Hydration Biomarker within the General Adult
Population and Athletes: A Systematic Review
CITATIONS READS
10 1,209
5 authors, including:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Brenda Davy on 12 May 2020.
To cite this article: Samantha B. Kostelnik, Kevin P. Davy, Valisa E. Hedrick, D. Travis Thomas
& Brenda M. Davy (2020): The Validity of Urine Color as a Hydration Biomarker within the General
Adult Population and Athletes: A Systematic Review, Journal of the American College of Nutrition,
DOI: 10.1080/07315724.2020.1750073
REVIEW
CONTACT Samantha B. Kostelnik samak94@vt.edu Department of Human Nutrition, Foods, and Exercise, Room 229 Wallace Hall, Virginia Tech,
Blacksburg, VA 24061, USA.
ß 2020 American College of Nutrition
2 S. B. KOSTELNIK ET AL.
Analyzer, Bayer’s Multistix, an optical refractometer, hand-held 1.010–1.020 ¼ dehydrated) (10,16). Armstrong et al., reported
refractometer). USG results in the general population are inter- moderate to strong correlations between USG and urine color
preted as follows: 1.013 ¼ over-hydrated or hyper-hydrated; in the athletic population (r ¼ 0.68–0.78), and in college males
l.014–1.029 ¼ euhydrated; 1.030 ¼ dehydrated (11). However, (r ¼ 0.54–0.93) (11). In older adults, correlation coefficients
these categories are different for the athletic population (i.e., between USG and urine color were substantially lower
(Table 3). Therefore, urine color may better reflect hydration
status in younger than older populations.
Table 2. Search term combinations.
Urinary terms Assessment Population
Urine color AND: Validity AND: Adult Urine osmolality (Uosm)
Urine AND: Validation Study AND: Older adult
Urinalysis AND: Reliability AND: Young adult Among urinary indicators of hydration status, Uosm is con-
Urine Osmolality AND: Reproducibility AND: College student
Plasma Osmolality AND: Athlete sidered the most valid technique in athletes and healthy
Fluid Balance AND: adults (27,28). Of the studies reviewed, eight utilized Uosm
Hydration AND:
Hydration Status AND:
as the comparator (Table 3). Moderate to high correlation
Euhydration AND: coefficients (i.e., 0.6–0.9) between urine color and Uosm
Dehydration AND: were reported in college-aged males (11) and young adult
Hypohydration AND:
athletes (15). Urine color demonstrated a high sensitivity for
Athletes n ¼ 13, 22.6 ± 4.9y, Chart; used as a predictor of environmental condition body mass loss þ urine
race unknown sweat loss during a soccer match not reported color change,
(compared to other predictors) r2 ¼ 0.57
Fernandez-Elias Olympic Combat Athletes 244 Cross-Sectional Researcher G1a Uosm (spot urine), r ¼ 0.70 (<0.001)
et al. (2014) male, 101 female, Armstrong Urine Color Chart; environmental condition
22.8 ± 4.1y, race unknown ratings compared across three not reported
b
categories of hydration G2 Uosm r ¼ 0.50 (<0.002)
c
status (G1–G3) G3 Uosm r ¼ 0.40 (<0.004)
Older Adults ( 65 years)
Mentes et al. (2006) Nursing Home Residents Cohort Armstrong Urine Color Chart Researcher Usg (spot urine), environmental r ¼ 0.48 (<0.01)
n ¼ 98, subgroup n ¼ 78, condition not reported
84 ± 8.4y, 98% Caucasian
Hooper et al. (2016) Older Adults in DRIE n ¼ 162, Cross-Sectional Human Hydration LLC; http:// Researcher Sosm (spot urine), ROC AUC ¼ 0.51,
85.8 ± 7.9y, race unknown www.hydrationcheck.com environmental condition 70%
Diagnostic accuracy of not reported
assessing dehydration
College Students
Armstrong et al. (1994) College Males n ¼ 23, 23 ± 2y, Cross-Sectional Armstrong Urine Color Chart Researcher Uosm r ¼ 0.62 (<0.01)
race unknown Usg (spot urine), r ¼ 0.54 (<0.01)
36 ± 3% humidity
College Males n ¼ 11, 23 ± 2y, Cross-Sectional Armstrong Urine Color Chart Researcher Uosm† r ¼ 0.92 (<0.001)
race unknown Usg† (spot urine), r ¼ 0.93 (<0.000001)
32.2 ± 0.4% humidity
Zhang et al. (2017) College Males in China Cross-sectional CIE Methodd (Lab); Diagnostic Researcher (using L Uosm (24-h urine), r¼ 0.56 (<0.0001)
n ¼ 68, 19.1 ± 1.1y accuracy for assessing dehydration a chromatrogram 31–35% humidity
spetrophotometer) b Uosm r ¼ 0.86 (<0.0001) For b:
97.4%/65.5%
a Uosm r¼ 0.35 (<0.0001)
Early et al. (2018) College Students n ¼ 6 male, Randomized Armstrong Urine Color Chart; Participant Experimentally induced short- Day 1: (p ¼ 0.04) Day 2:
10 female, 20.6 ± 1.2y, Cross-over ratings compared on days 1, 2, term (3 d.) hydration vs (p ¼ 0.002) Day
race unknown 3 of each study period dehydration (24-hr urine; 3: (p < 0.001)
not heat acclimated)
General Population
Malisova et al. (2016) Healthy Adults from 3 different Cohort Armstrong Urine Color Chart; Researcher Usg (24-hr urine), winter & p < 0.001
countries n ¼ 573, 39 ± 12y, ratings compared across three summer months
Greek, Spanish, German categories of hydration status
Perrier et al. (2017) Healthy French Adults Cohort Armstrong Urine Color Chart; Participant Uosm (24-hr urine), 87.8%/64.3%
n ¼ 82, 23.6 ± 2.9y Diagnostic accuracy for environmental conditions
assessing Uosm>500mOsm/kg not reported
†
pre-exercise comparison.
a
G1: Group 1 (euhydrated (250–700 mOsm kg H2O1)).
b
G2: Group 2 (dehydrated (700–1.080 mOsm kg H2O1)).
c
G3: Group 3 (severely dehydrated (1.080–1.500 mOsm kg H2O1)).
d
CIE (Commission Internationale de l’Eclarige) Method.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 5
Overall rating
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Neutral
Neutral
(using urine color >4), but lower specificity in healthy
French adults (64.3%) (29). Zhang et al. (30) reported a
high sensitivity for detecting dehydration, using a spectro-
photometer-based method to assess three aspects of urine
funding
10. Bias
unlikely
Unclear
due to
color (31).
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Two studies evaluated urine color compared to categories
of hydration status determined using Uosm; one in Olympic
combat athletes (32) and one in healthy adults (25).
Overall rating: Negative (), Neutral (ʘ), Positive (þ). Of 10 criteria: 6 or more questions rated as “no” ¼ (); Most answers including 2,3,6,7 are “yes” ¼ (þ); If 2,3,6,7 are not “yes” ¼ ʘ.
9. Conclusions
appropriate
analysis
and reliable
Unclear
Unclear
Unclear
Unclear
Yes
Yes
Yes
Yes
No
No
withdrawals
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Yes
Unclear
groups
from bias
Unclear
clearly stated
1. Research
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Armstrong, 1994
Fernandez, 2014
Malisova, 2016
Perrier, 2017
Zhang, 2017
Author, year
Early, 2018
was greater than four on the urine color chart, for three With regard to differences in urine sampling (i.e., spot vs
days. Urine color ratings were reported to be different on 24-hour urine collections), the range of associations also
days 1, 2, and 3 of each phase (all p < 0.05). appeared to be similar across studies. Most studies (6/10)
Harvey et al. evaluated predictors of sweat loss before measured spot urine, which has been reported to both over-
and after soccer matches in Premier Division soccer players and underestimate hydration status, depending on time of
(18). Predictors examined included urine color, USG, body day, relative to 24-hr urine collection (36). Two studies
mass, and hematocrit. The best prediction model included which utilized 24-hour urine samples to investigate the abil-
body mass change and urine color change (r2 ¼ 0.57). ity or urine color assessment to accurately diagnose dehy-
Based upon their findings, the investigators concluded that dration reported discrepant findings, which may be
urine color alone does not adequately assess hydration sta- attributed to differences in study population (older vs
tus. Therefore, urine color changes may be a suitable younger adults). Although Sosm is considered the gold
method for determining hydration status on a day-to-day standard for assessment of hydration status (4,27,37), only
basis in young adults, but not pre and post event in athletes. one study included in this review measured both urine color
and Sosm. However, the validity of urine color was not
assessed using Sosm (34).
Article quality assessment Three investigations that compared urine color to another
Results of the article quality assessment are provided in biomarker utilized different statistical methods such as the
Table 4. There was 100% agreement between reviewers on AIC Criterion and sensitivity/specificity (18,29,34). Findings
article quality rating criteria. Of the individual quality crite- from these studies indicated that urine color is a valid
ria, question 5 of the validity questions which addressed hydration status assessment method, although there are limi-
blinding to reduce risk of bias was most problematic across tations to using urine color as a sole hydration indicator
studies. Overall study quality was generally positive (8 of 10 (29,30). A variety of factors can influence urine color—these
studies) (24). include acute changes in body mass (38,39), certain foods,
medications and dietary supplements (40), time of day (16),
environmental conditions (4,16) and menstrual cycle phase
Discussion (15,21). Previous research has demonstrated strong associa-
tions between urine color, USG, and Uosm, and concluded
The literature reviewed generally supported the validity of
that these methods may all be used to determine hydration
urine color as a measure of hydration status in the healthy
status under normal conditions. However, due to their ease
adult population, including athletes. However, the available
of administration within real-world settings, USG and urine
research was limited to 10 studies, and very few studies
color are likely the most feasible (41).
were available in each population subgroup (i.e., 2–4; Overall, the evidence supporting the validity of urine
Athletes, 4 studies; Older adults, 2 studies; College students, color as a hydration biomarker in the general adult popula-
3 studies; General population, 2 studies. Of the articles tion and athletes was relatively consistent and positive in
included, 8 evaluated the validity of urine color compared to quality, although the limitations of this method must be
established, validated methods of hydration status. Two acknowledged. Findings appear less certain in older adults,
articles identified predictors of urine color and evaluated the given the suboptimal diagnostic accuracy of this method for
sensitivity to change in urine color based upon manipulation detecting dehydration in that study population.
of fluid intake over 3 days (25,26). There are several notable strengths of this review. First,
The color chart used was largely consistent across studies, an evidence-based framework was used to develop the
although investigations utilized a variety of reference meth- research question (i.e., PICOS) and methodology. Second,
ods. Eight of the studies used the validated urine color chart the PRISMA guidelines were utilized, which is the recom-
developed by Armstrong and colleagues (11), and one study mended approach for designing, conducting, and reporting
utilized an adapted version of this urine color chart (18). results of systematic reviews. Third, article quality ratings
Studies which utilized validated urinary indices (i.e., USG were included, and performed independently by two evalua-
and Uosm) as reference methods reported associations with tors. Lastly, focus of this review is significant in that
urine color assessment ranging from r ¼ 0.40–0.93 although urine color assessment is widely utilized in a var-
(7,11,15,30,32). Those which utilized USG reported moder- iety of settings for hydration status assessment, the validity
ate to strong correlations (i.e., r ¼ 0.48–0.93) with urine of this tool has not yet been systematically evaluated.
color (35). All but one of these studies utilized spot urine Several limitations in this body of literature were identi-
samples (vs 24-hour urine samples) for USG assessment. fied. First, a limited number of articles were available for
Five studies utilized Uosm as the reference method (in spot inclusion. Second, limitations in study methods related to
urine samples, n ¼ 2; in 24-hour urine samples, n ¼ 3), and validation methods across studies were noted. Specifically,
reported moderate to strong correlations ranging from all studies did not assess the validity of the urine color chart
r ¼ 0.40–0.92. Two articles included both USG and Uosm as using a validated hydration biomarker, and no studies uti-
reference methods; the magnitude of associations with urine lized the gold standard reference method for comparison.
color was similar between the two reference meth- Thus, research is needed to confirm the validity of urine
ods (11,15). color assessment using Sosm, across population subgroups.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 7
thirst and a water intake challenge in healthy men. Appl Physiol 32. Fernandez-Elias VE, Martinez-Abellan A, Lopez-Gullon JM.
Nutr Metab. 2014;39(2):138–44. doi:10.1139/apnm-2012-0369. Validity of hydration non-invasive indices during the weightcut-
21. Armstrong LE, Johnson EC, Munoz CX, Swokla B, Le Bellego L, ting and official weigh-in for Olympic combat sports. PloS One.
Jimenez L, Casa DJ, Maresh CM. Hydration biomarkers and 2014;9(4):e95336.
dietary fluid consumption of women. J Acad Nutr Diet. 2012; 33. Cheuvront SN, Ely BR, Kenefick RW, Sawka MN. Biological
112(7):1056–61. doi:10.1016/j.jand.2012.03.036. variation and diagnostic accuracy of dehydration assessment
22. Culp K, Mentes J, Wakefield B. Hydration and acute confusion markers. Am J Clin Nutr. 2010;92(3):565–73. doi:10.3945/ajcn.
in long-term care residents. West J Nurs Res. 2003;25(3):251–66. 2010.29490.
doi:10.1177/0193945902250409. 34. Hooper L, Bunn DK, Abdelhamid A, Gillings R, Jennings A,
23. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC,
Maas K, Millar S, Twomlow E, Hunter PR, Shepstone L, et al.
Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D,
Water-loss (intracellular) dehydration assessed using urinary
et al. The PRISMA statement for reporting systematic reviews
and meta-analyses of studies that evaluate health care interven- tests: how well do they work? Diagnostic accuracy in older peo-
tions: explanation and elaboration. PLoS Med. 2009;6(7): ple. Am J Clin Nutr. 2016;104(1):121–31. doi:10.3945/ajcn.115.
e1000100. doi:10.1371/journal.pmed.1000100. 119925.
24. Evidence Analysis Manual: Quality Criteria Checklist-Primary 35. Mukaka MM. Statistics corner: A guide to appropriate use of
Research. Acad NutrDiet: 339 EAL; 2018. correlation coefficient in medical research. Malawi Med J. 2012;
25. Malisova O, Athanasatou A, Pepa A, Husemann M, Domnik K, 24(3):69–71.
Braun H, Mora-Rodriguez R, Ortega J, Fernandez-Elias V, 36. Bottin JH, Lemetais G, Poupin M, Jimenez L, Perrier ET.
Kapsokefalou M, et al. Water intake and hydration indices in Equivalence of afternoon spot and 24-h urinary hydration bio-
healthy european adults: the european hydration research study markers in free-living healthy adults. Eur J Clin Nutr. 2016;
(EHRS). Nutrients. 2016;8(4):204. doi:10.3390/nu8040204. 70(8):904–7. doi:10.1038/ejcn.2015.217.
26. Early KS, Earnest CP, Theall B, Lemoine NP, Harrell B, 37. Armstrong LE. Assessing hydration status: the elusive gold
Johannsen NM. Free-living, continuous hypo-hydration, and car- standard. J Am Coll Nutr. 2007;26(sup5):575s–84s. doi:10.1080/
diovascular response to exercise in a heated environment. 07315724.2007.10719661.
Physiol Rep. 2018;6(8):e13672. doi:10.14814/phy2.13672. 38. O’Connell BN, Weinheimer EM, Martin BR, Weaver CM,
27. Popowski LA, Oppliger RA, Patrick Lambert G, Johnson RF, Campbell WW. Water turnover assessment in overweight adoles-
Kim Johnson A, Gisolf CV. Blood and urinary measures of cents. Obesity. 2011;19(2):292–7. doi:10.1038/oby.2010.225.
hydration status during progressive acute dehydration. Med Sci 39. Chang T, Ravi N, Plegue MA, Sonneville KR, Davis MM.
Sports Exerc. 2001;33(5):747–53. Inadequate hydration, BMI, and obesity among US adults:
28. Shirreffs SM, Maughan RJ. Urine osmolality and conductivity as NHANES 2009–2012. Ann Fam Med. 2016;14(4):320–4. doi:10.
indices of hydration status in athletes in the heat. Med Sci
1370/afm.1951.
Sports Exerc. 1998;30(11):1598–602.
40. Simerville JA, Maxted WC, Pahira JJ. Urinalysis: a comprehen-
29. Perrier ET, Bottin JH, Vecchio M, Lemetais G. Criterion values
sive review. Am Fam Physician. 2005;71(6):1153–62.
for urine-specific gravity and urine color representing adequate
41. Perrier E, Rondeau P, Poupin M, Le Bellego L, Armstrong LE,
water intake in healthy adults. Eur J Clin Nutr. 2017;71(4):
561–3. doi:10.1038/ejcn.2016.269. Lang F, Stookey J, Tack I, Vergne S, Klein A. Relation between
30. Zhang N, Du SM, Zheng MQ, et al. Urine color for assessment urinary hydration biomarkers and total fluid intake in healthy
of dehydration among college men students in Hebei, China—a adults. Eur J Clin Nutr. 2013;67(9):939–43. doi:10.1038/ejcn.
cross-sectional study. Asia Pac J Clin Nutr. 2017;26(5):788–93. 2013.93.
31. Takatsui F, Andrade M. F d, Neisser MP, Barros LAB, Loffredo 42. Hustrini NM, Siregar P, Nainggolan G, Harimurti K. Diagnostic
L. d C M. CIE Lab: comparison of digital images obtained performance of afternoon urine osmolality to assess optimal
photographically by manual and automatic modes. Braz Oral hydration status in an adult healthy population. Acta Medica
Res. 2012;26(6):578–83. doi:10.1590/S1806-83242012005000025. Indonesiana. 2017;49(2):112–7.