Interpretation of alkaline phosphatase (ALP) activity is essential for the diagnosis of certain diseases. ALP changes during life and may vary between different populations.
To establish reference intervals (RIs) and percentile charts for ALP activity in the Spanish population through a multicentric observational study and to compare the RIs to those defined in other countries.
A total of 662 350 ALP measurements from individuals ages 0 to 99 years from 9 Spanish tertiary care centers collected between 2020 and 2022 were analyzed. This study is the largest published on this topic in the literature to date.
Continuous percentile charts for ALP according to sex and age were established which can be used as RIs. Higher levels are reached during the first weeks of life. In puberty, a differential evolution is observed in both sexes, reaching a peak at 10 to 13 years of age in boys and remaining stable in girls at this age. Significant differences were also observed in adults, higher in men between ages 20 and 49 years and between ages 50 and 79 years in women, as reported in some countries.
ALP activity follows an age- and sex-dependent fluctuation with geographic differences. It is important to have appropriate reference values for each population in order to allow for a correct diagnostic interpretation and early diagnosis of diseases related to ALP abnormalities.
Alkaline phosphatase (ALP) is a membrane-bound phosphomonoesterase that catalyzes dephosphorylation reactions of inorganic pyrophosphate and other molecules and participates in various significant functions, such as hydrolyzing a variable spectrum of phosphate-containing physiologic compounds, contributing to DNA synthesis, bone calcification, and attenuation of inflammation.1 Four isoenzymes, with distinct functions and expressed by 4 separate genes, are known: 3 tissue-specific ALPs (intestinal, placental, and germ cell) and the ubiquitous tissue-nonspecific ALP, strongly expressed in bone, liver, and kidney, which accounts for approximately 95% of total serum ALP activity and plays a key role in the calcification of bones.2
Changes in ALP activity represent a sensitive indicator of skeletal and hepatobiliary diseases, so the establishment of reliable reference intervals (RIs) is the key to avoiding misdiagnosis.3 Recently, the discovery of an enzyme replacement therapy for hypophosphatasia (HPP),4,5 an inherited metabolic bone disease with highly variable clinical manifestations, mainly related to defective bone and tooth mineralization6 and characterized by low serum ALP and tissue-nonspecific ALP activity and increased levels of ALP substrates, makes the requirement for having ALP reference levels adapted to each population, especially in the pediatric age, more important.
Serum level of ALP changes with age and sex.7 Interpretation of ALP activity requires taking into consideration the dynamic sex- and age-specific changes in ALP activity during a lifetime. Because ALP is a marker of osteoblastic activity, its level is higher in growing children than in adults. Thus, the highest level is detected during rapid growth phases, as in childhood, especially during puberty.8–11 Some studies on bone markers in pediatric age showed that serum levels of ALP followed a course similar to that of the puberty stages, with higher levels of ALP between 10 and 12 years of age in girls, and between 13 and 15 years of age in boys, corresponding to Tanner stage 2.8,10 Physicians should be aware that many laboratories use the same adult ALP reference ranges for patients of all ages and sexes. The wide geographic differences observed in previous studies may be related to racial/ethnicity differences12 or environmental factors, which make it essential to standardize the reference values of ALP in each population.8,10,11
In 2011, standard procedures for the establishment of serum ALP RIs were published by the Committee on Reference Systems for Enzymes of the International Federation of Clinical Chemistry.3,13 However, there is a wide variation in results of ALP from various laboratories due to measurement procedure differences or diverse commercial kits, or because they use RIs adjusted from intervals that are old, that are not distinguished by sex and/or age, or that are simply obtained from the literature.3,12 Furthermore, many countries have published their RIs derived from studies of homogeneous populations according to race or ethnicity, such as Northern Europe,7,14–16 Germany,8,17,18 China,3,19 Israel,20 and Canada,21–23 among others, as shown in Table 1. Some of these and other publications have also described differentiated percentiles by age and sex in their population, whether in pediatric8,10,22,24–26 or adult populations.3,14,15,25,27 However, there are no published RIs and percentiles adjusted by sex and for all ages in the Spanish population. In addition, as shown in Table 1, our study is the largest yet published in terms of number of participants and samples analyzed using an indirect approach. Therefore, it is essential to have reliable ALP reference values adjusted by age and sex and by its population to avoid misdiagnoses.
MATERIALS AND METHODS
Study Participants
This multicentric observational study was performed during the period of January 1, 2020, to June 30, 2022, at 9 tertiary care centers in 6 regions in Spain (Table 2). These 9 Spanish hospitals belong to the same private hospital group with the same laboratory company. In this way, our results are well aligned and comparable. Measurements of ALP activity performed during clinical care of inpatients and outpatients of all ages in the hospitals included were analyzed. ALP blood measurements performed were retrieved from the participating laboratories’ databases. The ages ranged from newborns to 99 years, differentiated by sex and by their pathology. Exclusion criteria included those participants with known liver, renal, or bone diseases, or abnormal results of other tests of liver, kidney, or bone metabolism assessment, to ensure the accuracy of the data set as reference values because this would interfere with alkaline phosphatase measurement (Figure 1).
The age intervals used to measure the values of the reference percentiles (2.5, 5, 50, 75, and 97.5) and differentiated by sex are: (1) up to 3 months of age: [0, 15], (15, 30), (30, 45), (45, 60), (60, 75), and (75, 90) days; (2) from 3 months to 1 year: [0, 3], (3, 6), (6, 9), and (9, 12) months; (3) from 1 to 18 years: (1, 2), (2, 3), … up to (17, 18) years; (4) older than 18 years: (18, 20), (20, 30), (30, 40), … up to (90, 100).
The research complied with all the relevant national regulations and the tenets of the Declaration of Helsinki. The present study was approved by the Clinical Research Ethics Committee of “Ibérica de Diagnóstico y Cirugía” Salud in Catalonia, Spain (code No. 2019/32-UPN-HUGC).
Analytic Procedures
ALP activity (U/L) was determined by the enzymatic colorimetric method on the Roche/Hitachi Cobas platform, using amino-methyl-propanol (AMP) as a buffer as well as magnesium and zinc as cations, with the following automatic biochemical analyzers, all of them from Roche Diagnostics (Roche COBAS 6000 in 6 of the participating laboratories; Roche COBAS 8000 in 2 laboratories; and Roche Integra 400 in the remaining laboratory), as Table 2 shows. The same Roche platform and analyzer were used for the establishment of pediatric reference intervals by the CALIPER reference study.21 The method used for this study is standardized against the 2011 International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) procedure.
To ensure that data were comparable between the various laboratories, we chose to use the study period during which the same in vitro test for the quantitative determination of alkaline phosphatase in human serum and plasma (ALP2L: ACN 8683) on Roche COBAS systems analyzers was mostly used. In addition, quality controls and the frequency of calibrations were the same for all laboratories because they belong to the same lab company. In this way, the results obtained were more precise and consistent, and they allowed an optimal interpretation of them to ensure the accuracy of the reported results and their comparability.
The measuring range was 5 to 1200 U/L. The intra-assay percentage coefficient of variation (intra %CV) was 0.7% at concentrations between 83.7 and 84.9 U/L, 0.5% at concentrations between 221 and 223 U/L, and 0.3% at concentrations between 963 and 969 U/L. The interassay percentage coefficient of variation (inter %CV) was 2.4% at concentrations of 92.8 ± 2.2 U/L, 1.7% at concentrations of 224 ± 4 U/L, and 0.9% at concentrations of 1025 ± 9 U/L.
Statistical Analysis and Calculation of Percentile Charts
The statistical analysis of the data was performed using the statistical software R, version 4.2.2,28 in the Rstudio environment.29 The “Loess” method was used to estimate the curves that follow the percentiles (p2.5, p5, p50, p95, and p97.5) of the alkaline phosphatase level for both sexes, according to the established age groups. The “Loess” method30 is a robust and nonparametric local least squares method that does not require normality of the data, producing a smooth fit of the percentile curves.
RESULTS
We evaluated the ALP activity in blood samples taken from 662 350 outpatients and inpatients from 0 to 99 years of age from 9 tertiary hospitals in Spain. We further established the continuous percentile charts of ALP activity divided by age groups and differentiated by sex. With these 662 350 samples and after designing the database, some determinations were discarded when errors in the date of birth or date of extraction were observed that resulted in incongruous ages, accounting for 0.2% of the total samples. For patients who had several ALP measurements, median was calculated and chosen because it is the best representative of a data set. Likewise, in a first analysis to establish the percentiles (2.5, 5, 50, 75, and 97.5), data that exceeded the 97.5% percentile and had ALP values of several thousand units were considered clearly erroneous and outliers. To minimize this effect, these outliers, which represented the highest 2.5% of data, were eliminated from each age and sex group.
Finally, 644 675 samples were included in this study, 23 672 of them (3.7%) corresponding to pediatric samples (≤18 years), with a slight predominance of women (337 810 samples; 52.4%). No ethnicity differentiation was performed because the study population was mostly White.
Percentile Distribution
We created percentile charts for ALP activity in participants according to sex and age which can be used as RIs (Figures 2 through 5; Supplemental Table 1; and Supplemental Figures 1 through 6 [see the supplemental digital content containing 1 table and 6 figures at https://meridian.allenpress.com/aplm in the January 2025 table of contents]). The higher levels in both sexes during early infancy are reached during the first year, with a slight increase after the first 2 weeks of life. In puberty a differential evolution is observed in both sexes, with a marked increase in ALP activity in boys, reaching a peak at 10 to 13 years of age and with a slightly lower value between 13 and 15 years; instead, in girls ALP activity remains stable between 10 and 13 years and decreases to less than a half of the previous levels in the next 2 years.
Alkaline phosphatase percentiles for males aged 2 to 18 years, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Figure 3.Alkaline phosphatase percentiles for females aged 2 to 18 years, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Alkaline phosphatase percentiles for males aged 2 to 18 years, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Figure 3.Alkaline phosphatase percentiles for females aged 2 to 18 years, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Alkaline phosphatase percentiles for males of all ages, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Figure 5.Alkaline phosphatase percentiles for females of all ages, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% Cis. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Alkaline phosphatase percentiles for males of all ages, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% CIs. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Figure 5.Alkaline phosphatase percentiles for females of all ages, showing the 2.5th, 5th, 50th, 95th, and 97.5th percentiles, with 95% Cis. The x-axis corresponds to age in years, and the y-axis shows the ALP activity expressed in U/L.
Afterward, in men, a gradual decrease is subsequently observed until the beginning of the fourth decade, and later ALP activity remains practically unchanged until the age of 80 years. In women, after a progressive decrease up to 30 years of age, the levels remain stable for 2 decades, and subsequently an increase to 75 U/L is observed that remains from 50 to 90 years, with a minimal later increase.
DISCUSSION
Laboratory results must be interpreted based on an RI that is used to distinguish between “health” and “disease” and to assist in the diagnosis and/or prevention of disease. Generally, patients and physicians assume they will receive the same, or at least comparable, results from any laboratory; unfortunately, result interpretation—and for some methods, the result itself—may vary more than expected. In ALP interpretation the characterization of different isoenzymes and isoforms contributes to further confusion.31 There is consensus among researchers regarding age- and sex-related changes in serum ALP activity8–11,21 that should be kept in mind for a proper interpretation and, moreover, a normal total ALP activity does not rule out the presence of an abnormal isoenzyme pattern, particularly in children.31
The lack of standardization of ALP at different ages can contribute to the underdiagnosis of certain pathologies, especially when the diagnosis is based on the recognition of low levels, as in the case of HPP. A recent German study estimates that 0.52% of the adults (1:194) included show laboratory signs of HPP,32 but according to the literature the rate of recognition of low ALP activity is as low as 3%.33 Thus, more than 80% of adults with persistently low ALP activity received a diagnosis of HPP by genetic study, in both the osteoporotic and nonosteoporotic populations.34,35
The evolution of ALP activity during childhood and adolescence confirms the temporal fluctuations observed in previous studies10,11 except for the lack of an increase of ALP activity in female adolescents. ALP activity increases during the first month of life, reaching a peak at 3 months. After the first year of life ALP activity decreases and remains relatively stable until 10 years of age, succeeded by a sex-specific rise in adolescence. As previously described,36 ALP activity in our series reached higher levels in male adolescents.
The start of pubertal growth was recently established predominantly at 10 to 11 years in girls and at 12 to 13 years in boys in Catalonia, Spain.37 These data are consistent with the ALP peak level observed in 10- to 13-year-old boys in our population, but not with the adolescent girls’ results. Moreover, the peak observed in pubertal boys agrees in age with that observed in the Chinese pediatric population10 but differs from results by Zierk et al,8 because in the German population ALP activity peaks later in males (13–15 years). Nevertheless, the median and maximum ALP activity in males, 264 and 488.5 U/L (50th and 97.5th percentiles, respectively) are consistent with the published German results.8
In females, in line with ALP activity in France,11 there is not an increase in ALP activity at puberty. Strikingly, in females the maximum ALP peak reached was observed during the first year of life instead of at puberty. This observation is also consistent with data from Garnotel11 in girls. Population-based studies show a centennial decreasing trend in the age of onset of menarche among Spanish women.38–40 Because serum ALP level showed marked differences between girls with precocious and normal puberty, with a significantly higher ALP activity in 5- to 8-year-old girls with precocious puberty than in age-matched girls with typical puberty,41 we evaluate whether the lack of ALP increase levels in girls during puberty in our study could be related to an earlier menarche age. Despite the absence of pubertal ALP increase in females the median and maximum ALP activity in ages 10 to 13 years in girls (211 and 390 U/L respectively) is comparable to the German results (240 and 400 U/L).8
During adulthood, after the second decade of life, ALP activity remains stable until the eighth decade in men and the fifth decade in women, with higher levels in men from 20 to 49 years (p50: 73 U/L) than in women (65 U/L). An inversion of this trend is observed from 50 to 80 years (ALP activity in males, 77 U/L; and in females, 82 U/L) that could be partially explained by the increase of markers of bone turnover in postmenopausal women. In other words, we have observed statistically significant differences between men and women, the average being higher in men between 20 and 49 years and in women between 50 and 80 years. A Turkish study demonstrates, along these lines, significantly higher total ALP activity in postmenopausal women than those in the premenopausal period.42 Our data reinforced the positive correlation between plasma ALP and age described for both sexes43–45 and are in agreement with previous series that also found consistently higher ALP activity in women older than 65 years than in men46 and with reported ALP activity in postmenopausal Spanish women (mean ± SD, 72 ± 19 U/L).47
CONCLUSIONS
ALP activity follows an age- and sex-dependent fluctuation with geographic differences that make it essential to have appropriate reference values for each population in order to allow a correct diagnostic interpretation and early diagnosis of diseases related to ALP abnormalities. Data tables of age- and sex-specific percentile values allow the incorporation of these results into laboratory information systems. This study provides data on normal ALP values in the Spanish population, because it has been carried out with the largest number of samples published to date.
As a limitation of the study, we must point out that we could not exclude patients receiving drugs that could interfere physiologically or analytically in ALP measurements in the study population.
We thank Grupo Quirónsalud and MetabERN for their support in this study.