Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Clinical Chemistry and Laboratory Medicine (CCLM)

Published in Association with the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM)

Editor-in-Chief: Plebani, Mario

Ed. by Gillery, Philippe / Greaves, Ronda / Lackner, Karl J. / Lippi, Giuseppe / Melichar, Bohuslav / Payne, Deborah A. / Schlattmann, Peter

IMPACT FACTOR 2018: 3.638

CiteScore 2018: 2.44

SCImago Journal Rank (SJR) 2018: 1.191
Source Normalized Impact per Paper (SNIP) 2018: 1.205

See all formats and pricing
More options …
Volume 50, Issue 12


Phosphoethanolamine normal range in pediatric urines for hypophosphatasia screening

Apolline Imbard / Corinne Alberti / Priscilla Armoogum-Boizeau / Chris Ottolenghi
  • Metabolic Biochemistry Unit, Necker-Enfants Malades Hospital, APHP and Descartes University of Paris, Paris, France
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Emilie Josserand / Odile Rigal / Jean-François Benoist
Published Online: 2012-06-23 | DOI: https://doi.org/10.1515/cclm-2012-0266

Hypophosphatasia is a rare inherited disorder due to mutation in the ALPL gene coding for the renal/bone/hepatic isoform of alkaline phosphatase. Clinical presentation varies widely from antenatal forms with absence of bone mineralization, to adult late onset forms with early loss of teeth without bone symptoms. Diagnosis of hypophosphatasia can be suggested by clinical and radiographic signs of bone and/or teeth defective mineralization. Since these symptoms are not specific, laboratory assays are important for the differential diagnosis (1). Decreased total alkaline phosphatase activity and increased pyridoxal-5′-phosphate are observed in serum of affected patients, associated with increase of urinary phosphoethanolamine (PEA) and inorganic pyrophosphate. Finally, the molecular study of ALPL gene confirms the diagnosis.

As there is a lack of pediatric reference values for PEA in the literature, the contribution of this biomarker to the diagnosis remained to be investigated. Here we determine reference intervals for urinary PEA and compare concentrations in four new cases and 12 previously reported cases to our reference intervals.

We retrospectively collected the concentrations of PEA in 888 urines measured in our laboratory during a 4-year period between 2006 and 2010. Urines were collected from children with age ranging between 1 day and 19 years. All samples obtained from children with other known diseases or neonatal renal tubular immaturity were excluded. Amino acids chromatographies showing contaminant peaks were also excluded. PEA was measured in urine after deproteinization using 20% sulfosalicylic acid (10:1 v/v) by ion exchange chromatography (IEC) and ninhydrin post-column derivatization method performed on the AminoTac™ Amino Acid Analyzer (Jeol®, Croissy sur seine, France). Data acquisition and calculations were made using the Jeol Workstation software.

Age specific reference intervals for urinary PEA were estimated by a semi-parametric method described by Royston and Wright (2). A reference interval is the range of values encompassed by a pair of symmetrically placed extreme percentiles. Assuming a Gaussian distribution of the variable PEA at each age, a percentile was calculated according to a widely used formula: percentile=mean+U×SD, where mean and SD (standard deviation) are defined for a given age and U is the corresponding percentile constant of the standard Gaussian distribution (e.g., for the 5th and 95th percentiles curves, U=±1.64). The aim was to identify functions that adequately represent the changes in mean and SD with age. Initial mathematical transformation was applied, if required, to reduce the positive skewness and heteroscedasticity of the measurements of interest.

Suitable functions for the mean and SD based on age were obtained from regression splines using generalized additive models. These splines were defined with two knots to obtain a set of knots {kmin, k1, k2, kmax} among which (kmin, kmax) are placed at the extreme ends of the age distribution and k1, k2 were centile-based positions, namely 33rd and 67th centiles of age. The optimal number of knots was determined visually as the number that provided a clinically adequate curve without excessive wiggles or board effects.

Thus, the mean may be written as meanage01 (Age)2υ1 (Age)3υ2 (Age) where and (Agek)+=max(0, Agek). The same can be written for σT.

Percentile estimated and reference intervals were calculated by introducing the fitted curves of the mean and standard deviation into the percentile equation. Percentiles curves on the original scale were obtained by applying a back transformation to the calculated curves. At the end of the process, several internal validity measures were checked among which were normality probability plots, proportions of observations above and below the 90th and 95th reference intervals, Q-tests exploring the Z-score [Z-score=(measurement–mean)/SD] moments, and permutation bands (3). Statistical analysis was performed using the STATA 10.0 software for PC (Statacorp Inc.®).

Among the 888 patients included, 64% (568) were male and 36% (320) were female. The median age of the population was 3.6 years with 1 day and 19.6 years of minimum and maximum, respectively.

Urinary PEA was scattered by sex against age using a running line smoother providing a preliminary assessment of the shape of the mean separately by sex. No influence of gender was found (superposition of the curves), whereas there was an evidence of age dependency. We chose to model urinary PEA against age (Figure 1). At birth, urinary PEA has a large distribution, which decreases rapidly until the age of 3 years. Between 3 and 15 years, the urinary PEA values slowly continue to decrease to reach a stable interval starting from the age of 15 years.

Urinary PEA (mean, 5th and 95th percentiles) by age. The points represent urinary PEA concentrations obtained in 888 controls. The dotted lines represent respectively the 5th and 95th percentiles and the black line represents the mean by age according to our model.
Figure 1

Urinary PEA (mean, 5th and 95th percentiles) by age.

The points represent urinary PEA concentrations obtained in 888 controls. The dotted lines represent respectively the 5th and 95th percentiles and the black line represents the mean by age according to our model.

A preliminary exponential transformation was applied to age to take into account the constancy of urinary PEA with increasing age and resulting in a new created variable X:

Formulas to estimate the mean and SD of urinary PEA as a function of age were:

Meanage=0.6790234+14.29676 X+8.85511 v1(X) –33.63992 v2(X)

SDage=1.360018+16.35116 X+119.5439 v1(X)–133.2542 v2(X) with

Reference interval curves (mean, 5th and 95th percentiles) for urinary PEA are displayed in Figure 1, calculated, values by these formulas per age-class interval are given in Table 1.

Table 1

Urinary PEA concentrations in children with hypophosphatasia compared to the reference intervals obtained with our model.

To our knowledge, the only reference intervals for urinary PEA published were those by Licata et al. (9). These values were not detailed for children [0–15 years: (9.3–25); 15–30 years: (4.7–16.5); 31–41 years: (4.3–17.5); >45 years: (5.4–10.5) mmol/mol creatinine]. However, as previously suggested by Eastman and Bixler (11) in a smaller cohort but without any reference values estimated urine PEA concentration shows an age related distribution in children. For example, the 95th percentile decreased by 50% during the first 18 months of life. It is therefore essential to interpret the urine PEA concentration with age related reference intervals.

Table 1 also summarizes the urinary PEA values found in patients with proven hypophosphatasia. In all cases, urinary PEA was above the 95th percentile (at least upper than three times the 95th percentile), which suggests a high sensitivity for this biomarker.

Contrasting with the apparently high sensitivity of urinary PEA, the degree of its specificity has been debated. Licata et al. reported that urine PEA concentration can be elevated in other bone diseases (9). Moreover, it has been suggested that heterozygote carriers have also increased urinary PEA. Rasmussen and Licata et al. reported three carriers (12, 14 and 15 years) with urinary PEA concentrations of 22, 12 and 25 mmol/mol creatinine, respectively (7, 9). Based on our reference intervals, all these values are above the 95th percentile. This would suggest that PEA cannot distinguish between heterozygote carriers and hypophosphatasia patients.

In conclusion, PEA is an interesting biomarker for the diagnosis of the pediatric forms of hypophosphatasia because its measurement is easy and can be performed routinely by all laboratories involved in the diagnosis of inherited metabolic diseases. Further studies are necessary to confirm our conclusion that genetic investigation may not need to be performed if urine PEA is normal. Therefore, we recommend systematically screening any suspected case of hypophosphatasia by measuring both serum ALP and urine PEA before engaging into genetic analysis.

Conflict of interest statement

Authors’ conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article.

Research funding: None declared.

Employment or leadership: None declared.

Honorarium: None declared.


  • 1.

    Mornet E. Hypophosphatasia. Best Pract Res Clin Rheumatol 2008;22:113–27. Web of ScienceGoogle Scholar

  • 2.

    Royston P, Wright E. A method for estimating age-specific reference intervals based on fractional polynomilas and exponential transformation. J R Stat Soc Appl Stat 1998;161: 79–101. Google Scholar

  • 3.

    Royston P, Wright EM. Goodness-of-fit statistics for age-specific reference intervals. Stat Med 2000;19:2943–62. Google Scholar

  • 4.

    Whyte MP, McAlister WH, Patton LS, Magill HL, Fallon MD, Lorentz WB, Jr., et al. Enzyme replacement therapy for infantile hypophosphatasia attempted by intravenous infusions of alkaline phosphatase-rich Paget plasma: results in three additional patients. J Pediatr 1984;105:926–33. Google Scholar

  • 5.

    Whyte MP, Valdes R, Jr., Ryan LM, McAlister WH. Infantile hypophosphatasia: enzyme replacement therapy by intravenous infusion of alkaline phosphatase-rich plasma from patients with Paget bone disease. J Pediatr 1982;101:379–86. Google Scholar

  • 6.

    Mochizuki H, Saito M, Michigami T, Ohashi H, Koda N, Yamaguchi S, et al. Severe hypercalcaemia and respiratory insufficiency associated with infantile hypophosphatasia caused by two novel mutations of the tissue-nonspecific alkaline phosphatase gene. Eur J Pediatr 2000;159:375–9. Google Scholar

  • 7.

    Rasmussen K. Phosphorylethanolamine and hypophosphatasia. Dan Med Bull 1968;15(Suppl 2):1–112. PubMedGoogle Scholar

  • 8.

    Muller HL, Yamazaki M, Michigami T, Kageyama T, Schonau E, Schneider P, et al. Asp361Val mutant of alkaline phosphatase found in patients with dominantly inherited hypophosphatasia inhibits the activity of the wild-type enzyme. J Clin Endocrinol Metab 2000;85:743–7. Google Scholar

  • 9.

    Licata AA, Radfar N, Bartter FC, Bou E. The urinary excretion of phosphoethanolamine in diseases other than hypophosphatasia. Am J Med 1978;64:133–8. Google Scholar

  • 10.

    Macfarlane JD, Poorthuis BJ, van de Kamp JJ, Russell RG, Caswell AM. Hypophosphatasia: biochemical screening of a Dutch kindred and evidence that urinary excretion of inorganic pyrophosphate is a marker for the disease. Clin Chem 1988;34:1937–41. Google Scholar

  • 11.

    Eastman JR, Bixler D. Urinary phosphoethanolamine: normal values by age. Clin Chem 1980;26:1757–8. Google Scholar


    About the article

    Corresponding author: Apolline Imbard, Biochemistry-Hormonology Laboratory, Robert Debre Hospital, 48 bd Serurrier, 75019 Paris, France Phone: +33 1 40034722, Fax: +33 1 40034790

    Received: 2012-04-27

    Accepted: 2012-06-04

    Published Online: 2012-06-23

    Published in Print: 2012-12-01

    Citation Information: Clinical Chemistry and Laboratory Medicine, Volume 50, Issue 12, Pages 2231–2233, ISSN (Online) 1437-4331, ISSN (Print) 1434-6621, DOI: https://doi.org/10.1515/cclm-2012-0266.

    Export Citation

    ©2012 by Walter de Gruyter Berlin Boston.Get Permission

    Citing Articles

    Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

    Tomoyuki Akiyama, Takuo Kubota, Keiichi Ozono, Toshimi Michigami, Daisuke Kobayashi, Shinji Takeyari, Yuichiro Sugiyama, Masahiro Noda, Daisuke Harada, Noriyuki Namba, Atsushi Suzuki, Maiko Utoyama, Sachiko Kitanaka, Mitsugu Uematsu, Yusuke Mitani, Kunihiro Matsunami, Shigeru Takishima, Erika Ogawa, and Katsuhiro Kobayashi
    Molecular Genetics and Metabolism, 2018
    I. Gennero, F. Conte-Auriol, and J.-P. Salles
    Archives de Pédiatrie, 2017, Volume 24, Number 5, Page 5S57

    Comments (0)

    Please log in or register to comment.
    Log in