FI-RADS: an imaging-based scoring system for skeletal fluorosis—adjunctive clinical and biochemical analysis

Introduction

Endemic fluorosis is a globally prevalent condition closely associated with excessive fluoride intake. It is characterized by a spectrum of clinical manifestations that primarily affect dental enamel and the skeletal system, with potential impacts on other organ systems. Rasool et al. (1) reported that approximately 200 million individuals across 25 countries are affected by this pressing public health issue. Among them, India and China, the two most populous nations, bear the greatest burden. Dental fluorosis and skeletal fluorosis are the hallmark manifestations of this condition (2-4). Unfortunately, skeletal fluorosis lacks a definitive and effective cure, presenting a serious threat to human health (5,6). In recent years, the implementation of water quality improvement and fluoride mitigation programs, supported by both central and local governments, has contributed to a gradual reduction in areas with high fluoride exposure (6). Nonetheless, chronic endemic fluorosis remains a significant public health challenge worldwide, particularly in India, parts of Asia, and Africa (2,7).

Excessive fluoride exposure produces a wide range of harmful effects on osteoblasts, osteoclasts, cartilage tissue, and bone mineralization in humans (2,6). The disease presents with symptoms that vary from mild joint pain to severe deformities and disabling conditions. The diagnosis of skeletal fluorosis is primarily established by integrating epidemiological evidence with supportive radiographic findings. Radiological assessment plays a critical role in both diagnosing the disease and monitoring its progression.

Previous research has identified a range of X-ray features characteristic of skeletal fluorosis, including osteosclerosis, exostosis, and joint anomalies (8-14). However, these studies have often been constrained by small sample sizes (e.g., case reports), retrospective designs, and the absence of a standardized diagnostic framework. Although China’s Diagnostic Standard for Endemic Skeletal Fluorosis (WS/T 192-2021) (15) provides a foundational framework for radiographic diagnosis of fluorosis, challenges persist in its practical application and implementation. This system primarily relies on qualitative descriptions of features such as ligament ossification and interosseous membrane ossification, resulting in diagnostic outcomes that are highly dependent on individual physician experience. This makes it difficult to ensure intra- and inter-observer consistency. There is a clear need for a descriptive study that systematically examines the full spectrum of radiographic findings in individuals affected by endemic skeletal fluorosis. Such research would provide a deeper and more comprehensive understanding of the diseases imaging characteristics, thereby improving diagnostic accuracy and supporting the development of more effective therapeutic strategies.

This study adopted a cross-sectional study design, enrolling patients from regions with endemic fluorosis. Participants underwent X-ray imaging examinations, with the images evaluated by a team of radiologists using a standardized scoring system. In addition, data on clinical features and demographic characteristics were collected to contextualize the radiological findings. By delineating the radiological spectrum of X-ray features in endemic skeletal fluorosis, this study aimed to improve diagnostic accuracy and strengthen public health initiatives aimed at reducing both the incidence and the severe consequences of this debilitating disease. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2773/rc).

Methods

Ethical consideration

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the Affiliated Hospital of Guizhou Medical University (No. 2023 [189]), and informed consent was provided by all individual participants. Strict measures were taken to protect patient confidentiality and data security, ensuring privacy and anonymity.

Study design and participants

This study was a cross-sectional survey based on the population in the affected area. According to the classification standard for endemic fluorosis areas in China (GB 17018-2011), Baixing Town in Nayong County, Guizhou Province, was selected as the survey site in September 2023. This location is identified as an endemic fluorosis area caused by coal-burning pollution, as confirmed by the Centers for Disease Control and Prevention. With assistance from the disease control system, we continuously recruited individuals who voluntarily participated and met all inclusion and exclusion criteria locally. The inclusion criteria were as follows: (I) long-term residence in the place of origin; (II) voluntary participation in the study, willingness to provide biological samples, and signed informed consent; (III) no history of mental illness or related diseases, with normal cognitive and expressive abilities; (IV) chronic resting pain symptoms in three or more limb areas, including major joints, neck, and waist, unaffected by seasonal or climatic changes. The exclusion criteria were as follows: (I) patients with fluoride poisoning who had previously received pharmacological treatments or interventions in China; (II) patients with rheumatoid arthritis, ankylosing spondylitis, or metabolic bone diseases (Figure 1). We included all patients meeting the eligibility criteria during September 2023, comprising a total sample size of 943 individuals.

Figure 1 Flow chart for the participant enrollment of the study. FI-RADS, Fluorosis Imaging Reporting and Data System.

Data collection methods

The images were acquired using a digital radiography machine (WDW 1000 series, Wandong Medical, Beijing, China; Shuguang Town Health Center) and an X-ray machine (IMIX-2000, Oy Imix Ab, Tampere, Finland; Yangchang Town Health Center). The uniform scanning parameters were as follows: Upper extremity anteroposterior view: 50–60 kV, focus-to-detector (FDD) 110–120 cm; Tibia and fibula anteroposterior view: 60–70 kV, FDD 110–120 cm; Pelvic anteroposterior view: 70–85 kV, FDD 110–120 cm. The image files from both centers were renumbered using a unified numbering system bearing no logical relationship to the research centers. Standard projection positions were used, including the forearm, pelvis, and lower leg, all captured in the anteroposterior view. The anteroposterior X-ray images of the forearm (including the elbow joint) and lower leg (including the knee joint) were required to clearly display the interosseous membrane between the radius and ulna, as well as the interosseous membrane between the tibia and fibula.

Imaging analysis

Based on the imaging manifestations of fluorosis summarized in the classification standard for endemic fluorosis areas in China, we evaluated eight specific signs: enlargement of the radial crest with sclerotic margins and rough surfaces; ossification of the ulnar–radial interosseous membrane; loosening at the attachment of the pronator teres; ossification of the tendon of the musculus soleus; ossification of the tibiofibular interosseous membrane; ossification of the obturator foramen membrane; ossification of the sacrotuberous ligament; and ossification of the sacrospinous ligament. On this basis, we developed the Fluorosis Imaging Reporting and Data System (FI-RADS) imaging scoring evaluations system.

Diagnostic principles of FI-RADS

In fluorosis-endemic regions, assessment of suspected skeletal fluorosis should begin with confirmation of exposure, core symptoms and signs, and corroborating radiographs. The following criteria summarize the key elements that support FI-RADS-based scores.

• Diagnostic prerequisites: clear history of residence in a fluorosis-endemic area, accompanied by evident clinical symptoms, physical signs, and characteristic X-ray findings.
• Clinical symptoms: (I) persistent resting pain in three or more regions (major limb joints, neck, lower back), unaffected by seasonal or climatic changes; (II) Restricted joint mobility or secondary nerve impairment.
• Characteristic X-ray findings: enlargement of the radial crest with sclerotic, rough margins; ossification of the ulnar-radial interosseous membrane; loosening at the pronator teres attachment; ossification along the soleus tendon; ossification of the tibiofibular interosseous membrane; ossification of the obturator foramen membrane; and ossification of the sacrotuberous and sacrospinous ligaments.
• Application: these features can be systematically captured and graded using FI-RADS to standardize assessment in endemic regions.

FI-RADS scoring criteria

The FI-RADS scoring system standardizes X-ray assessment of skeletal fluorosis across multiple sites (forearm, pelvis, and lower leg) to support clinical diagnosis; it is not intended for monitoring treatment efficacy (Figure 2).

• FI-RADS 1: very low probability, extremely unlikely to be present;
• FI-RADS 2: low probability, unlikely present;
• FI-RADS 3: moderate possibility, suspicious finding;
• FI-RADS 4: high probability; may be present;
• FI-RADS 5: very high, highly likely to be present.

Figure 2 Example diagram of subjective evaluation scores for enlarged radial crest with sclerotic margins and rough surfaces. Examples of the other seven signs are provided in the Figure S1. Yellow circles and red arrows both mark the radial crest. Yellow circles in FI-RADS 1 and 2 are used to outline the anatomical location of the radial crest. Red arrows in FI-RADS 3–5 indicate the skeletal fluorosis-specific changes of the radial crest, including enlargement, sclerotic margins, and rough surfaces, as diagnostic confidence increases across the scale. FI-RADS, Fluorosis Imaging Reporting and Data System.

A FI-RADS score of 4 or 5 should be classified as skeletal fluorosis. For the detailed FI-RADS scoring of imaging features across the eight sites, please refer to Appendix 1.

All X-ray images were independently reviewed by four radiologists with 8, 10, 23, and 24 years of experience in musculoskeletal imaging, respectively. In cases of disagreement, consensus was achieved through structured consensus meetings. After the initial assessments were recorded anonymously, all readers jointly re-examined the images and discussed them according to the established radiographic diagnostic criteria until a final agreement was reached. This study randomly selected 30 cases to evaluate the reproducibility and clinical feasibility of the FI-RADS scoring system. A young radiologist specializing in musculoskeletal imaging performed blinded scoring independently according to FI-RADS criteria, without access to the expert consensus results. The radiologist’s scores were then compared with the consensus findings of four experts, and intraclass correlation coefficients (ICCs) were calculated to assess the degree of agreement.

Statistical analysis

Statistical analyses were conducted to evaluate the associations between the radiological features of skeletal fluorosis and key factors, including age and total bile acid (TBA) levels. Descriptive statistics were applied to summarize the prevalence of radiological abnormalities, presented as frequencies and percentages. Chi-squared (χ²) tests were used to assess associations between categorical variables across these groups. Employing spline-based non-linear analysis, this study investigated the association between age and TBA levels with radiological alterations associated with fluorosis. ICCs were used to assess the consistency of the two sets of scores.

According to the eight typical X-ray manifestations and diagnostic criteria, FI-RADS greater than 3 is considered positive, whereas an FI-RADS of 3 or below is considered negative by the evaluation score of the right-hand side. Univariate and multivariate logistic regression analyses were performed to examine factors associated with skeletal fluorosis (FI-RADS >3). Univariate analysis was conducted to evaluate the crude associations between each independent variable and the outcome, with results presented as odds ratios (ORs) and 95% confidence intervals (CIs). Variables with a P<0.1 in the univariate analysis, along with core variables selected based on clinical a priori knowledge (age, joint pain indicators), were included in the multivariate logistic regression model to adjust for potential confounders. Multivariate analysis was then used to identify independent predictors of the outcome, with adjusted ORs and 95% CIs reported. Statistical significance was defined as P<0.05. All analyses were conducted using standard statistical methods to ensure robustness and reliability. To explore potential non-linear relationships between continuous variables (age, TBA) and radiographic changes in skeletal fluorosis (right side), we incorporated these variables into a multivariate logistic regression model using restricted cubic spline (RCS) functions. The number of spline nodes was set to three. We assessed the statistical significance of non-linear relationships by comparing the non-linear model containing spline terms with a simplified model containing only linear terms via a likelihood ratio test.

The data were analyzed using the statistical software R (http://www.R-project.org, The R Foundation) and MedCalc version 23.3.5 (MedCalc Software, Ostend, Belgium). A P value of less than 0.05 was considered statistically significant, and the assumptions for chi-square tests were verified to ensure robustness.

This study included a total of 943 participants. Blood biochemical samples were collected from 491 individuals, of which 28 were excluded due to sample contamination. Accordingly, valid blood biochemical data were available for 463 participants.

Results

Patient characteristics

Table 1 presents the clinical characteristics of patients with endemic skeletal fluorosis. With some patients refusing to cooperate, the study cohort comprised 491 participants, with a mean age of 65.93 years [standard deviation (SD) =9.62 years] and a median age of 67.00 years [interquartile range (IQR), 59.00–72.00 years]. The mean pain level was 5. 50 (SD =1.85), with a median of 6.00 (IQR, 4.00–7.00). Biochemical parameters showed a mean total bilirubin of 12.82 µmol/L (SD =4.96 µmol/L) and direct bilirubin of 3.87 µmol/L (SD =1.88 µmol/L). Liver function tests indicated mean values of alanine aminotransferase at 22.69 IU/L and aspartate aminotransferase at 25.22 IU/L. Lipid profiles revealed a mean triglyceride level of 2.02 mmol/L, low-density lipoprotein cholesterol at 2.70 mmol/L, and high-density lipoprotein cholesterol at 1.41 mmol/L. The mean total cholesterol was 4.96 mmol/L, and uric acid averaged 323.27 µmol/L. The gender distribution consisted of 73.72% females and 26.28% males. Joint mobility was assessed for the elbow, spine, knee, and shoulder, distinguishing between normal and restricted activity levels.

X-ray imaging features in patients with suspected skeletal fluorosis in the endemic area by FI-RADS imaging scoring system

Table 2 summarizes the radiological characteristics of 942 individuals with suspected skeletal fluorosis in the endemic area, comparing findings between the left and right sides. Specific features included radial crest abnormalities (48.15%), ulnar-radial interosseous membrane ossification (6.47%), and pronator teres attachment changes (36.16%), with no significant differences observed between sides (all P>0.05). Likewise, abnormalities of the musculus soleus tendon (23.30%), tibiofibular interosseous membrane ossification (7.70%), sacrotuberous ligament (0.44%), sacrospinous ligament (1.21%), and obturator membrane changes (6.55%) were symmetrically distributed. These results underscore the bilateral and symmetrical nature of skeletal alterations in fluorosis. Among the 30 validated cases, the agreement between the FI-RADS total scores assigned by junior physicians and the expert consensus total scores was good, with an ICC of 85.24% (95% CI: 81.37–88.36%; P<0.001).

Univariate and multivariate logistic regression analyses for FI-RADS (>3) in skeletal fluorosis

Table 3 presents the results of univariate and multivariate logistic regression analyses identifying risk factors associated with a radiological score of >3 based on eight key imaging characteristics. Age was confirmed as a significant risk factor, with the OR increasing with age. Elbow joint abnormalities, knee joint abnormalities, and shoulder joint abnormalities showed weak associations in univariate analysis but did not reach statistical significance in multivariate analysis. Other skeletal features, such as spine abnormalities, were not significantly associated with FI-RADS >3 score in either analysis (P>0.05). TBA were significant risk factors in both univariate analysis (OR =1.124, 95% CI: 1.032–1.223, P=0.007) and multivariate analysis (OR =1.102, 95% CI: 1.004–1.210, P=0.040). Other biochemical markers, including alanine aminotransferase, triglycerides, and cholesterol levels, showed no significant associations in either analysis (P>0.05). These findings suggest that age and TBA are key factors influencing the severity of radiological changes in skeletal fluorosis.

The spline-based nonlinear association analysis

Nonlinear correlation analysis revealed that most typical radiological features exhibited significant nonlinear dependence on age. Among the eight assessed features, six demonstrated statistically significant nonlinear P values (P<0.05), indicating that models incorporating nonlinear spline basis functions significantly outperformed purely linear models in fitting the data. Loosening of the pronator teres insertion exhibited the strongest nonlinear association (model R²=0.154, χ²=54.38, P<0.0001). Threshold effect analysis further revealed the presence of significant age-related thresholds within the non-linear correlation characteristics, distributed across the 48–79 years age range (Table S1). Figure 3 indicates that the predictive effect value of interosseous membrane ossification in the forearm and lower limb (radioulnar and tibiofibular membranes) declined sharply between 40 and 60 years, stabilizing thereafter at 70 years and beyond. Most pelvic ligament ossifications similarly exhibited a sustained decline throughout adulthood, with the most pronounced changes occurring between 50 and 70 years. Soleus tendon ossification showed a marked decrease between 60 and 70 years.

Figure 3 RCS plots depicting the predicted effect of age on radiographic features at eight skeletal sites. The solid lines represent the predicted effect estimates, and the shaded areas indicate the 95% CIs. The x-axis shows age (years), and the y-axis represents the predicted effect size on the imaging score of each site. RCS analyses demonstrated nonlinear trends across most anatomical locations. CI, confidence interval; RCS, restricted cubic spline.

The majority of radiological manifestations in the upper and lower limbs exhibited a significant non-linear relationship with TBA levels. The strongest non-linear correlation was observed in the ulnar-radial interosseous membrane ossification (model R²=0.0723, χ²=13.59, non-linear P=0.001). Further threshold effect analysis revealed risk thresholds for these significantly non-linear indicators, distributed within the range of 1.8 to 3 µmol/L (Table S2). Figure 4 demonstrates that the effect values for ossification of the interosseous membrane of the ulna and radius, and the interosseous membrane of the tibia and fibula, exhibit a trend of initially decreasing then increasing with rising bile acid concentrations. Ossification of the sacrotuberous ligament declined sharply at lower bile acid levels before stabilizing. Ossification of the soleus tendon exhibited an approximately linear positive correlation with bile acid concentration. Overall, bile acids provided limited explanatory power for morphological variation, with R² values across all models ranging from 0.006 to 0.072.

Figure 4 RCS plots depicting the predicted effect of TBA on radiographic features at eight skeletal sites. The solid lines represent the predicted effect estimates, and the shaded areas indicate the 95% CIs. The x-axis shows TBA (µmol/L), and the y-axis represents the predicted effect size on the imaging score of each site. RCS analyses demonstrated nonlinear trends across most anatomical locations. CI, confidence interval; RCS, restricted cubic spline; TBA, total bile acids.

Discussion

This study comprehensively analyzed the factors influencing FI-RADS >3 score in skeletal fluorosis, integrating imaging characteristics, demographic data, and biochemical variables. The results demonstrated that age and TBA levels exhibit a significant nonlinear relationship with fluoride osteopathy scores (FI-RADS >3), whereas other imaging and biochemical markers showed limited associations. These findings provide valuable insights into the progression and clinical manifestations of skeletal fluorosis. The findings highlight the bilateral and symmetrical nature of skeletal fluorosis and reinforce the importance of early detection and intervention. Furthermore, the significant nonlinear associations between bile acid levels and radiological changes suggest a potential metabolic component in the pathophysiology of skeletal fluorosis, warranting further investigation into their role as diagnostic markers or therapeutic targets.

Previous radiological diagnoses have relied primarily on the subjective judgment of radiologists, without standardized quantitative criteria. This reliance can result in variability and subjectivity in diagnostic outcomes. Based on the FI-RADS scoring system, this study sought to establish a more objective and quantifiable method for evaluating the radiological features of fluorosis. Clarifying these features enhances the accuracy of fluorosis diagnosis and provides reference criteria for subsequent epidemiological research and clinical interventions.

Nonlinear analysis based on spline functions revealed significant nonlinear relationships between age and the predictive effect values of multiple anatomical structures. Overall, most structures exhibited pronounced morphological changes during middle age (40–60 years). Ossification of the interosseous membranes in the upper and lower limbs (ulno-radial membrane, tibio-fibular membrane) and pelvic ligaments (sacrospinous ligament, sacrotuberous ligament) demonstrated a marked decline in effect size during this phase. This indicates that with advancing age, the skeleton becomes more sensitive to fluoride, or the cumulative effects of fluoride become increasingly pronounced (16-18). Upon entering the elderly stage (post-70 years), the trends for most structures stabilize, indicating a slowing rate of change. This correlates with the general reduction in bone turnover rates during old age (19,20), which diminishes the interaction between fluoride and the bone matrix. Furthermore, the correlation between elevated TBA levels and the degree of bone mineralization indicates that bile acid metabolism may contribute to the progression of fluorosis. This mechanism could involve bile acid-induced oxidative stress or exacerbated inflammatory responses that intensify fluoride-induced bone damage. Fluoride exposure is known to affect multiple organs (1,21), including the liver (1,21), which may account for the observed alterations in bile acid metabolism. Elevated bile acid levels may reflect liver dysfunction secondary to fluoride toxicity. This novel finding highlights the importance of investigating metabolic markers, alongside imaging characteristics, for a more comprehensive assessment of skeletal fluorosis. Bile acids function both as detergents that facilitate lipid digestion and as hormones that activate specific receptors, thereby regulating a wide range of physiological processes, including metabolism, glucose homeostasis, inflammation, and liver regeneration (22). Recent studies have demonstrated that bile acids regulate bone homeostasis by activating bile acid receptors in osteoblasts and osteoclasts (23). Its effects primarily depend on the farnesoid X receptor (FXR) and Takeda G protein-coupled receptor 5 (TGR5). Upon activation, FXR promotes osteoblastic differentiation through Runt-related transcription factor 2 (Runx2), ERK, and β-catenin pathways while inhibiting receptor activator of nuclear factor-B ligand (RANKL)-induced osteoclastogenesis by directly regulating nuclear gene expression. Membrane-bound TGR5 primarily modulates transcriptional activity indirectly via intracellular signaling systems such as cyclic adenosine monophosphate (cAMP), protein kinase C, and cAMP-responsive element-binding protein (CREB) (24-27) (Figure 5). Therefore, we hypothesize that disruption of bile acid signaling contributes to bone homeostasis imbalance in fluorosis by exacerbating bone metabolism disorders and impairing the remodeling process. Low concentrations of bile acids may exert beneficial effects through moderate activation of protective pathways such as FXR, whereas excessively high concentrations may cause receptor saturation or signaling pathway desensitization, resulting in a plateau in biological responses. However, the R² values across all models ranged between 0.006 and 0.072, indicating that bile acids possess limited explanatory power regarding bone morphogenesis. Their clinical utility requires integration with other clinical or radiographic markers for comprehensive assessment.

Figure 5 The modulatory effects of BAs on bone metabolism by targeting various cells. CA, DCA, and LCA predominantly influence macrophages within the bone marrow by binding to the TGR5 receptor and inhibiting the AMPK signaling pathway. LCA also inhibits the expression of the OCN and RANKL genes by binding to the VDR, which helps regulate the remodeling process of bone. CDCA positively affects BMMSCs by increasing the expression of osteoblast genes such as BSP, OCN, OPN, and ALP through the FXR receptor. At the same time, it reduces the expression of adipocyte genes such as PPARγ, adipoQ, leptin, and C/EBPα. Furthermore, CDCA activates the Runx2, ERK, and β-catenin signaling pathways, promoting osteoblast differentiation, which is beneficial for bone health. UDCA has a positive effect on BMMSCs and macrophages, promoting Runx2 expression by up-regulating SP7 gene, increasing the expression of osteoblast markers such as MMP9, and inducing RANK expression to regulate bone remodeling process. TUDCA promotes osteoblast formation by influencing the differentiation of BMMSCs, and accelerates new blood vessel formation and bone formation by modulating AKT and eNOS signaling associated with ERK1/2 and JNK activation. adipoQ, adiponectin; AKT, protein kinase B; AMPK, AMP-activated protein kinase; BA, bile acid; BMMSC, bone marrow mesenchymal stem cell; BSP, bone sialoprotein; C/EBPα, CCAAT/enhancer-binding protein alpha; CA, cholic acid; DCA, deoxycholic acid; eNOS, endothelial nitric oxide synthase; ERK, extracellular signal-regulated kinase; ERK1/2, extracellular signal-regulated kinases 1 and 2; FXR, farnesoid X receptor; JNK, c-Jun N-terminal kinases; LCA, lithocholic acid; MMP9, matrix metalloproteinase 9; OCN, osteocalcin; OPN, osteopontin; PPARγ, peroxisome proliferator-activated receptor gamma; RANKL, receptor activator of nuclear factor Kappa-B ligand; Runx2, Runt-related transcription factor 2; SP7, Sp7 transcription factor; TGR5, Takeda G-protein coupled receptor 5; TUDCA, tauroursodeoxycholic acid; UDCA, ursodeoxycholic acid; VDR, vitamin D receptor.

This study has several limitations. First, the cross-sectional design restricts the ability to determine causality between age, bile acid levels, and the progression of skeletal fluorosis, underscoring the need for longitudinal studies to confirm these associations. Second, the study concentrated on specific imaging features without examining other potential skeletal or soft tissue changes, which may limit the comprehensiveness of the findings. Third, although significant associations were identified between bile acid levels and radiological abnormalities, the study did not investigate the underlying mechanisms or account for potential confounders such as variations in fluoride exposure, nutritional deficiencies, or comorbidities, which may have influenced the results. These unmeasured confounders may simultaneously influence both bile acid metabolism and skeletal health, and thus their potential confounding effects on the observed association between TBA and FI-RADS scores cannot be ruled out. Fourth, selection bias may exist, as blood biochemical data were only available for 463 of the 943 participants (49.1%). Although the missing data were due to sample contamination and classified as missing completely at random, the possibility of residual selection bias cannot be entirely ruled out. Addressing these limitations in future research will provide a more comprehensive understanding of skeletal fluorosis.

Conclusions

Skeletal fluorosis is a progressive systemic disease characterized by bilateral and symmetrical radiological changes. Age and TBA levels were non-linear predictors of various radiographic features of skeletal fluorosis. The integration of our newly developed X-ray-based FI-RADS grading system—which standardizes and quantifies key diagnostic features—with relevant demographic and biochemical factors may enhances early detection, objective risk stratification, and clinical decision-making in fluorosis-endemic regions. Prospective validation, reliability testing, and performance evaluation of this integrated approach are required. The associations with bile acid levels indicate metabolic involvement in the disease’s pathophysiology, warranting further investigation into bile acids as diagnostic markers or therapeutic targets. Future longitudinal studies should validate FI-RADS performance and elucidate mechanisms underlying joint-specific and metabolic changes in skeletal fluorosis.

Acknowledgments

We would like to thank all the participants for their cooperation, the staff at our university for their support, and the members of the StudyForBetter team for their dedicated research efforts throughout the project.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2773/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2773/dss

Funding: This work was supported by the National Key Research and Development Program (No. 2022YFC2503003, to Peng Luo) and the Guizhou Provincial Major Scientific and Technological Program (No. [2024]015, to Peng Luo).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2773/coif). Peng Luo received grants from the National Key Research and Development Program (No. 2022YFC2503003) and the Guizhou Provincial Major Scientific and Technological Program (No. [2024]015). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the Affiliated Hospital of Guizhou Medical University (No. 2023 [189]), and informed consent was taken from all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Rasool A, Farooqi A, Xiao T, Ali W, Noor S, Abiola O, Ali S, Nasim W. A review of global outlook on fluoride contamination in groundwater with prominence on the Pakistan current situation. Environ Geochem Health 2018;40:1265-81. [Crossref] [PubMed]
2. Veneri F, Iamandii I, Vinceti M, Birnbaum LS, Generali L, Consolo U, Filippini T. Fluoride Exposure and Skeletal Fluorosis: a Systematic Review and Dose-response Meta-analysis. Curr Environ Health Rep 2023;10:417-41. [Crossref] [PubMed]
3. Solanki YS, Agarwal M, Gupta AB, Gupta S, Shukla P. Fluoride occurrences, health problems, detection, and remediation methods for drinking water: A comprehensive review. Sci Total Environ 2022;807:150601. [Crossref] [PubMed]
4. Foat A, Stevens C, George G, Massawe J, Mhina A, Gray WK, Mmbaga BT, Rwakatema DS, Sallis P, Jarvis H, Haule I, Benedict D, Walker R. Prevalence of Skeletal Fluorosis in Northern Tanzania: A Follow-Up Study. Glob Health Sci Pract 2023;11:e2200342. [Crossref] [PubMed]
5. Till C, Grandjean P, Martinez-Mier EA, Hu H, Lanphear B. Health Risks and Benefits of Fluoride Exposure During Pregnancy and Infancy. Annu Rev Public Health 2025;46:253-74. [Crossref] [PubMed]
6. Zhou J, Sun D, Wei W. Necessity to Pay Attention to the Effects of Low Fluoride on Human Health: an Overview of Skeletal and Non-skeletal Damages in Epidemiologic Investigations and Laboratory Studies. Biol Trace Elem Res 2023;201:1627-38. [Crossref] [PubMed]
7. Mittal HC, Bajpai R, Singh RK, Sardana D, Mohanty U. Prevalence of Dental Fluorosis in India: A Systematic Review and Meta-Analysis. J Public Health Dent 2025;85:410-8. [Crossref] [PubMed]
8. Sellami M, Riahi H, Maatallah K, Ferjani H, Bouaziz MC, Ladeb MF. Skeletal fluorosis: don't miss the diagnosis! Skeletal Radiol 2020;49:345-57. [Crossref] [PubMed]
9. Assefa G, Shifera G, Melaku Z, Haimanot RT. Clinical and radiological prevalence of skeletal fluorosis among retired employees of Wonji-Shoa sugar estate in Ethiopia. East Afr Med J 2004;81:638-40. [PubMed]
10. Wang Y, Yin Y, Gilula LA, Wilson AJ. Endemic fluorosis of the skeleton: radiographic features in 127 patients. AJR Am J Roentgenol 1994;162:93-8. [Crossref] [PubMed]
11. Gerges MM, Ngo VT, Tu J, Colletti PM, Wassef H. Skeletal Fluorosis. Clin Nucl Med 2023;48:1127-30. [Crossref] [PubMed]
12. Meena L, Gupta R. Skeletal Fluorosis. N Engl J Med 2021;385:1510. [Crossref] [PubMed]
13. Mayer MJ, Gliedt JA, Shaker JA, Symanski JS. Case 298: Skeletal Fluorosis Secondary to Huffing. Radiology 2022;302:484-8. [Crossref] [PubMed]
14. Boivin G, Chavassieux P, Chapuy MC, Baud CA, Meunier PJ. Skeletal fluorosis: histomorphometric analysis of bone changes and bone fluoride content in 29 patients. Bone 1989;10:89-99. [Crossref] [PubMed]
15. National Health Commission of the People’s Republic of China. (2021). WS/T 192-2021 Diagnostic standard for endemic skeletal fluorosis. Available online: https://www.nhc.gov.cn/wjw/s9500/202107/39a79e1ef16149868b9918743328de83.shtml
16. Turner CH, Boivin G, Meunier PJ. A mathematical model for fluoride uptake by the skeleton. Calcif Tissue Int 1993;52:130-8. [Crossref] [PubMed]
17. Grynpas MD. Fluoride effects on bone crystals. J Bone Miner Res 1990;5:S169-75. [PubMed]
18. Ayele BA, Godebo TR, Tekle-Haimanot R, Yifru YM. Neuro-medical manifestations of fluorosis in populations living in the Main Ethiopian Rift Valley. Environ Geochem Health 2022;44:1129-36. [Crossref] [PubMed]
19. Fan L, Lin H. Influence of estrogen on bone mass during different physiological periods. Chinese Journal of Osteoporosis 2011;17:5.
20. Macey R, Tickle M, MacKay L, McGrady M, Pretty IA. A comparison of dental fluorosis in adult populations with and without lifetime exposure to water fluoridation. Community Dent Oral Epidemiol 2018;46:608-14. [Crossref] [PubMed]
21. Perera T, Ranasinghe S, Alles N, Waduge R. Effect of fluoride on major organs with the different time of exposure in rats. Environ Health Prev Med 2018;23:17. [Crossref] [PubMed]
22. de Aguiar Vallim TQ, Tarling EJ, Edwards PA. Pleiotropic roles of bile acids in metabolism. Cell Metab 2013;17:657-69. [Crossref] [PubMed]
23. Xiang T, Deng Z, Yang C, Tan J, Dou C, Luo F, Chen Y. Bile acid metabolism regulatory network orchestrates bone homeostasis. Pharmacol Res 2023;196:106943. [Crossref] [PubMed]
24. Cho SW, An JH, Park H, Yang JY, Choi HJ, Kim SW, Park YJ, Kim SY, Yim M, Baek WY, Kim JE, Shin CS. Positive regulation of osteogenesis by bile acid through FXR. J Bone Miner Res 2013;28:2109-21. [Crossref] [PubMed]
25. Zheng T, Kim NY, Yim M. Fexaramine Inhibits Receptor Activator of Nuclear Factor-κB Ligand-induced Osteoclast Formation via Nuclear Factor of Activated T Cells Signaling Pathways. J Bone Metab 2017;24:207-15. [Crossref] [PubMed]
26. Chiang JYL, Ferrell JM. Discovery of farnesoid X receptor and its role in bile acid metabolism. Mol Cell Endocrinol 2022;548:111618. [Crossref] [PubMed]
27. Li Z, Huang J, Wang F, Li W, Wu X, Zhao C, Zhao J, Wei H, Wu Z, Qian M, Sun P, He L, Jin Y, Tang J, Qiu W, Siwko S, Liu M, Luo J, Xiao J. Dual Targeting of Bile Acid Receptor-1 (TGR5) and Farnesoid X Receptor (FXR) Prevents Estrogen-Dependent Bone Loss in Mice. J Bone Miner Res 2019;34:765-76. [Crossref] [PubMed]