Common blood test results may offer an early clue to bone loss, suggesting that alkaline phosphatase levels could help identify people who might benefit from early screening for osteoporosis before fractures occur.
Study: The relationship between serum total alkaline phosphatase and the risk of osteoporosis: a cross-sectional study. Image Credit: Javier Regueiro/Shutterstock
In a recent study published in the journal Frontiers in Endocrinology, researchers investigated whether the enzyme alkaline phosphatase (ALP), routinely measured in blood, can serve as a marker of osteoporosis.
They found that higher ALP levels were consistently related to a higher likelihood of osteoporosis, with stronger associations observed among metabolically healthy, younger individuals and women, and identified a potential threshold for recommending further bone health assessments.
Osteoporosis burden and need for accessible biomarkers
Osteoporosis is characterized by reduced bone mass and structural deterioration, leading to an increased risk of fractures and substantial impacts on health and quality of life. As life expectancy increases, its prevalence increases worldwide. As the incidence of fractures increases dramatically with age, especially after age 75, there is growing interest in identifying accessible biomarkers that can help detect bone loss earlier.
FA, produced mainly by osteoblasts and bone-forming hepatocytes, plays a key role in bone mineralization by degrading pyrophosphate. Approximately half of FA in the blood originates in bone, and bone-specific FA closely tracks total FA levels in healthy and osteoporotic populations.
Total FA is inexpensive and widely available for routine medical checkups, and researchers have explored its potential as a surrogate marker of bone health. However, previous findings are inconsistent: some studies reported negative associations between FA and bone mineral density, while others found no clear pattern.
Factors such as sample size, population heterogeneity, reliance on self-reported data, and metabolic or hepatic conditions influencing AF further complicate interpretation.
Study population and clinical evaluations.
The researchers sought to clarify whether total FA can reliably indicate the risk of osteoporosis in a large, systematically evaluated population. They conducted their analysis using cross-sectional data from routine medical examination records from a large university hospital in Chongqing, China, between 2019 and 2024.
Eligible participants were adults aged 20 years or older who had completed blood AF testing and dual-energy X-ray absorptiometry (DXA) scans of the hip and spine. Records with incomplete information were excluded and when there were duplicate entries, only the most recent examination was considered.
Osteoporosis was diagnosed according to World Health Organization (WHO) criteria using DXA T scores, with modified definitions applied to younger adults. Standardized hospital procedures were used to collect anthropometric measurements, blood pressure, liver ultrasound findings, and biochemical markers including glucose, lipids, uric acid, and liver enzymes. Definitions of metabolic abnormalities followed established medical guidelines.
Statistical analyzes included descriptive comparisons, t tests, chi-square tests, and five logistic regression models that progressively adjusted for age, sex, body composition, metabolic markers, and liver function. Restricted cubic spline regression tested non-linear associations between osteoporosis and ALP, while receiver operating characteristic (ROC) analysis assessed the predictive performance of ALP and identified an optimal cut-off value using the Youden index.
Participant characteristics and baseline associations.
Among 12,835 participants, 9.5% were diagnosed with osteoporosis and almost all individuals (99%) had ALP levels within the clinical reference range. Participants with osteoporosis had significantly elevated ALP levels. Older people, women, and those with lower body weight or higher waist-to-hip ratio were more likely to have osteoporosis. Those at higher risk also showed higher levels of systolic blood pressure, fasting glucose, total cholesterol and high-density lipoprotein (HDL), while uric acid and liver enzymes were lower. No differences were observed in diastolic blood pressure, triglycerides, or low-density lipoprotein (LDL).
ALP-osteoporosis associations between statistical models
Logistic regression consistently demonstrated that each 1 IU/L increase in FA was associated with increased odds of osteoporosis, with modest but cumulative per unit effect sizes across the FA range, and this association remained strong in all adjusted models. Spline analysis showed a mostly linear relationship, but the association flattened when FA exceeded 100 IU/L. ROC analysis indicated poor to modest discrimination, with 72 IU/L emerging as the best cut-off point for predicting osteoporosis.
Subgroup differences and metabolic influences
Subgroup analyzes revealed stronger statistical associations, rather than higher absolute risk, in women, younger individuals, and those with normal liver enzymes and healthier metabolic profiles. When liver enzymes were elevated, or when glucose or lipid profiles were abnormal, the association was substantially weakened or disappeared, suggesting that metabolic and hepatic factors may distort the link between bone status and AF.
Interpretation, limitations and clinical implications.
This study found that higher serum total FA is consistently associated with a higher likelihood of osteoporosis, even within the normal reference range and after adjusting for extensive confounders.
The association was stronger in younger women and metabolically healthy individuals, probably because AF more accurately reflects bone-derived AF when liver function and metabolic status are normal. Elevated FA may represent a compensatory increase in bone turnover in response to decreased bone density, rather than a direct cause of bone loss. However, when there is liver injury or metabolic abnormalities, the liver-derived component of AF may dilute this relationship.
Strengths include the large sample, standardized clinical data, and detailed subgroup analyses. However, the study was based on a cross-sectional design, drew population from a single center, and did not include information on physical activity, thyroid status, medication use, and diet, which limits the findings.
In general, an FA level of around 72 IU/L can serve as a tentative threshold for recommending further evaluations of bone health, although longitudinal cohort studies are required to confirm its causal and predictive value.
