Article Text
PDF
Abstract
Objectives This study aimed to investigate the prevalence of diabetes mellitus in patients with Fabry disease using a nationwide population-based dataset. We hypothesised that patients with Fabry disease would have a higher prevalence of diabetes mellitus compared with the general population.
Design A cross-sectional study.
Setting Taiwan.
Participants We identified a study sample from Taiwan’s LHID2010 Database. There were 9408 sampled patients in this study, 2352 study patients with Fabry disease and 7056 propensity-score-matched comparison patients.
Primary outcome measures Multiple logistic regression analyses were conducted to explore the association between diabetes mellitus and Fabry disease after taking the variables of age, sex, geographic location, monthly income category, urbanisation level of the patient’s residence, hyperlipidaemia and hypertension into consideration.
Results The results revealed significantly higher prevalence of diabetes mellitus among patients with Fabry disease than among comparison patients (35.8% vs 29.6%, p<0.001). The difference was observed, irrespective of gender, males 37.0% vs 30.6%, (p<0.001) and female patients, 34.8% vs 28.7% (p<0.001). The adjusted OR of diabetes mellitus for patients with Fabry disease was 1.424 (95% CI=1.276~1.591) when compared with comparison patients. Classified by sex, the adjusted ORs for diabetes were 1.428 (95% CI=1.213~1.682) for males and 1.424 (95% CI=1.226~1.655) for females. Additionally, among patients aged <65 years, the adjusted OR was 1.489 (95% CI=1.277~1.737), and among patients aged 65 or older, the adjusted OR was 1.361 (95% CI=1.161~1.596).
Conclusions Our findings suggest the importance of screening for diabetes mellitus in patients with Fabry disease. Physicians should be conscious of this association and be prepared to manage both conditions concurrently.
- Epidemiology
- DIABETES & ENDOCRINOLOGY
- Diabetes Mellitus, Type 2
Data availability statement
Data may be obtained from a third party and are not publicly available. The data that support the findings of this study are available from HWDC, Department of Statistics, Ministry of Health and Welfare, Taiwan, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This study used population-based patient data from a large national database (Taiwan’s LHID2010).
Our study’s retrospective observational design, bolstered by the use of propensity score matching, refines the reliability of our findings by minimising selection and misclassification biases.
The study’s cross-sectional design does not permit inference of temporal relationships, notably causality between Fabry disease and diabetes mellitus.
The diagnosis of Fabry disease and diabetes mellitus was based on ICD codes, but ICD codes do not represent disease severity or the full clinical picture of the patients.
Introduction
Fabry disease is an uncommon lysosomal storage disorder that is X-linked and arises due to mutations in the GLA gene. This genetic anomaly leads to a shortage of alpha-galactosidase A enzyme, which results in the buildup of globotriaosylceramide and related glycosphingolipids within different tissues, ultimately causing gradual organ deterioration.1 2 The clinical manifestations of Fabry disease are diverse and include renal, cardiac and cerebrovascular disease, neuropathic pain and gastrointestinal symptoms.3 4
Diabetes mellitus, a chronic metabolic disorder characterised by hyperglycaemia, is a major global health concern due to its high prevalence and associated morbidity and mortality. It is well-established that diabetes mellitus can lead to various complications, including cardiovascular disease, kidney disease and neuropathy.5 6
A growing body of evidence suggests a potential link between Fabry disease and diabetes mellitus.7 Several case reports and small studies have reported the cooccurrence of Fabry disease and diabetes mellitus, raising the possibility of a shared pathophysiological mechanism. More recently, a study using a screening-based approach with expanded inclusion criteria—encompassing patients with chronic kidney disease attributed to hypertension and diabetes—found a notable cooccurrence of Fabry disease and diabetes mellitus, reinforcing the need to explore this association in broader clinical settings.8 Despite these observations, large-scale epidemiologic data quantifying the prevalence of diabetes among patients with Fabry disease remain scarce. Therefore, in this study, we aimed to investigate the prevalence of diabetes mellitus in patients with Fabry disease using a nationwide population-based dataset, based on the hypothesis that patients with Fabry disease have a higher prevalence of diabetes mellitus compared with the general population.
Methods
Database
Taiwan implemented its single-payer National Health Insurance (NHI) programme in 1995, a mandatory social insurance programme providing comprehensive medical care coverage for all Taiwanese citizens. We identified the study sample for this cross-sectional study from Taiwan’s Longitudinal Health Insurance Database 2010 (LHID2010). The LHID2010 consists of registration files of 2,000,000 NHI beneficiaries and claims files with data on a registry for contracted medical facilities, the Registry for beneficiaries, details of inpatient orders, ambulatory care expenditures by visit, ambulatory care visits made, diagnoses, details of ambulatory care orders and prescriptions dispensed at NHI-contracted pharmacies. Several Taiwanese researchers have utilised deidentified datasets offered by the LHID2010 to publish epidemiological investigations on diseases and treatments.8
The study was approved by the institutional review board (IRB) of Taipei Medical University (TMU-JIRB NN202308030). Patient consent was deemed not necessary as the study involved the use of deidentified data released for research purposes by the National Health Research Institute. The IRB of Taipei Medical University waived the informed consent requirement. All methods were performed in accordance with the Declaration of Helsinki.
Identification of cases and comparison patients
This cross-sectional study compared a study group with a demographically and medically similar comparison group. For the cases (study group), we identified all persons aged ≥20 years old with a diagnosis of Fabry disease (ICD-9-CM code 272.7 or ICD-10-CM code E75.21) during an ambulatory care visit between January 2016 and December 2019, totaling 2352 patients. Although the vast majority of patients with Fabry disease in Taiwan are classified as having the late-onset form, the ICD coding system does not permit differentiation between the classic and late-onset phenotypes. In Taiwan, Fabry disease is usually first suspected by cardiologists, especially in later-onset cases with unexplained cardiomyopathy, but a definitive diagnosis requires a comprehensive evaluation. Genetic confirmation (identification of a pathogenic GLA variant) is mandatory for all patients. In males, markedly reduced α-galactosidase A (α-GAL A) activity in leukocytes or plasma (<5% of normal) provides strong supportive evidence. In females, however, α-GAL A activity is an unreliable marker because heterozygous carriers may show enzyme levels ranging from severely reduced to near normal, making genetic analysis indispensable. Furthermore, Taiwan’s NHI programme designates around 30 categories of “catastrophic illnesses” (including cancers, haemophilia, autoimmune disorders, chronic renal failure and rare genetic diseases) to mitigate the financial burden on affected families. Fabry disease is an officially recognised rare disorder (code A3-03, ICD-10-CM E75.21) and thus qualifies for this catastrophic illness certification. Patients meeting the stringent diagnostic criteria for Fabry disease can obtain a Catastrophic Illness Certificate (CIC), which grants them full exemption from copayments for Fabry-related medical care. To receive a CIC for Fabry disease, the patient’s case must be reviewed and approved by a specialised NHI committee, with documentation of the diagnostic criteria described above (such as enzyme assay results, genetic test reports and even biopsy findings in atypical cases). This rigorous review process ensures a high level of diagnostic certainty—only bona fide Fabry disease cases are granted a CIC. Consequently, the validity of Fabry disease diagnoses in Taiwan is exceptionally high, reflecting the robust confirmation protocols in place.
We obtained the comparison patients by selecting LHID2010 enrollees who were not included in the Registry of Beneficiaries. We first eliminated all enrollees who had a diagnosis of Fabry disease in medical encounters before January 2016. After that, we used propensity score-based matching to identify comparison patients, three comparison patients for every patient with Fabry disease, a ratio of 3:1. The matching variables used in this study included socio-demographics (sex, age, geographic location, monthly income category and urbanisation level of the patient’s residence) as well as medical comorbidities (hyperlipidaemia and hypertension). The aforementioned variables were incorporated into a multivariable logistic regression model as predictors in order to determine the propensity score for Fabry disease in each enrollee. Given that precise score-matched comparison patients could not be obtained for all cases, we resorted to utilising an accepted alternative approach to identify controls—the nearest neighbour within the callipers method (with a predetermined value of ±0.02 set for the callipers). As a result, we identified a total of 9408 study sample patients, 2352 patients with Fabry disease and 7056 comparison patients.
Measures of outcome
A diagnosis of diabetes mellitus was the outcome of interest. Patients with diabetes mellitus, including type 1 and type 2 diabetes, were identified by ICD-9-CM code 250 or ICD-10-CM codes E10-E14. To improve the reliability of the diagnosis, we only included patients with at least two different medical encounter claims showing a diagnosis of diabetes mellitus.
Statistical analysis
The SAS system was used for statistical analyses. The t-tests and Chi-square were performed to assess differences in sociodemographic characteristics and medical comorbidities among patients with Fabry disease versus comparison patients. We employed multiple logistic regression analyses to assess the association between diabetes mellitus and Fabry disease after taking the variables of sex, age, geographic location, monthly income, urbanisation level of the patient’s residence, hypertension and hyperlipidaemia into consideration.
Results
Of 9408 sample patients, the mean age was 59.6 (±15.1 years, SD). After using propensity-score matching, table 1 shows no statistically significant difference between patients with Fabry disease and comparison patients on sex (p>0.999), urbanisation level (p=0.999), monthly income (p=0.998), geographic region (p=0.948), prevalence of hypertension (p>0.999) and hyperlipidaemia (p>0.999).
Demographic characteristics of study patients stratified by the presence of Fabry disease in Taiwan (n=9408)
Table 2 shows the prevalence of diabetes mellitus among the sample patients. A total of 2927 (31.1% of 9408 patients) had a diagnosis of diabetes mellitus. Chi-square test showed a statistically significant difference in diabetes prevalence between the Fabry disease group and comparison group (35.8% vs 29.6%, p<0.001), which was sustained among both sexes, 37.0% vs 30.6% among male patients and female patients, 34.8% vs 28.7% (both p<0.001). We also found that the adjusted OR of diabetes for patients with Fabry disease relative to the comparison group was 1.424 (95% CI=1.276~1.591) after adjusting for the variables of sex, monthly income, age, geographic location, residential urbanisation level and the presence of hyperlipidaemia and hypertension. Classified by sex, the adjusted ORs for diabetes were 1.428 (95% CI=1.213~1.682) for males and 1.424 (95% CI=1.226~1.655) for females.
Crude and covariate-adjusted ORs of having diabetes mellitus among patients with Fabry disease relative to comparison patients
Table 3 presents the results of adjusted analysis after categorising the study sample on age (<65 and ≥65 years old). Among patients aged <65 years, the adjusted OR was 1.489 (95% CI=1.277~1.737) and among patients aged 65 or older, the adjusted OR was 1.361 (95% CI=1.161~1.596).
Crude and covariate-adjusted ORs of having diabetes mellitus among patients with Fabry disease and comparison patients classified by age
Discussion
In this population-based study, we find a significantly higher prevalence of diabetes mellitus among individuals diagnosed with Fabry disease compared with those without a diagnosis of this rare genetic disorder among the total sample and within the male and female patient groups as well as within older and younger patient groups. Our findings suggest that Fabry disease may be associated with an increased risk of developing diabetes mellitus. Our findings have implications for the clinical care of patients with Fabry disease, suggesting that clinicians should consider regular screening for diabetes mellitus in this population.
Understanding the higher risk of diabetes mellitus among patients with Fabry disease is crucial as each condition is independently associated with substantial morbidity and mortality, and the coexistence of these conditions could exacerbate the clinical presentation and complicate the management of both diseases. Moreover, it helps physicians to better anticipate and manage diabetes mellitus among patients with Fabry disease, potentially improving patient outcomes.
Few studies have investigated the prevalence of diabetes mellitus in patients with Fabry disease. Most of the available literature consists of case reports and small-sample studies on the cooccurrence of the two conditions. In 2006, Fischer et al procured renal biopsies from patients diagnosed with Fabry disease and scrutinised clinical data alongside the results of light and electron microscopy. Among their cohort comprising 11 cases, a septuagenarian woman presented with concomitant type II diabetes mellitus, exhibiting an augmented mesangial matrix accompanied by early nodule formation and peri-glomerular fibrosis.9 While little data on this phenomenon is available from epidemiological studies, an early study by Kohler et al investigated plasma levels of lysosomal enzymes, including α-galactosidase, in 110 patients with insulin-dependent diabetes mellitus and found significantly decreased enzyme activity levels compared with healthy controls. This finding suggests that lysosomal enzyme deficiencies may not only be a hallmark of genetic disorders like Fabry disease but could also manifest as a secondary phenomenon in diabetes, potentially linking the two diseases through common metabolic disruptions.10 An animal study with Chinese hamsters also demonstrated that the α-galactosidase activity was significantly lower in the kidneys of diabetic animals than among nondiabetic animals.11 Another study investigated enzymatic activities of certain lysosomal enzymes in the plasma of insulin-dependent diabetic patients and found support for a direct connection between the lysosomal apparatus and insulin-controlled metabolic pathways.12 A subsequent study reported that the serum lysosomal enzyme levels reflected the glucose metabolic control status, supporting that these enzymes may be a proxy for the overall metabolic state connected with insulin actions.13 Serum lysosomal enzymes have been demonstrated as valid indicators of the metabolic control of the disease in each diabetic patient.14 Collectively, these studies—spanning both human and animal models—provide indirect but compelling evidence supporting a biological link between Fabry disease and diabetes mellitus. For example, Chang demonstrated in diabetic Chinese hamsters that renal α-galactosidase activity was significantly depressed, implicating diabetes-related metabolic changes in lysosomal dysfunction.11 In a study using insulin-dependent diabetic patients, Burlina et al reported altered plasma levels of lysosomal enzymes during artificial pancreas treatment, reinforcing the hypothesis that lysosomal enzymes reflect broader metabolic control disturbances.12 These findings underscore the plausibility of shared or intersecting pathogenic mechanisms, such as impaired lysosomal degradation or glycosphingolipid metabolism, which may contribute to the observed association in our study. However, more biomedical research is needed to identify the exact mechanisms mediating the higher prevalence of diabetes mellitus among individuals diagnosed with Fabry disease.
In our study, the overall prevalence of diabetes mellitus was 35.8% among Fabry patients, significantly higher than 29.6% in the matched comparison group. Stratified analysis revealed a higher prevalence in males (37.0%) than in females (34.8%), mirroring patterns of sex differences in diabetes prevalence in the general population. However, the adjusted ORs for both sexes were nearly identical (1.428 for males and 1.424 for females), suggesting that Fabry disease elevates diabetes risk across both sexes at similar magnitudes.
With regard to age, we observed a diabetes prevalence of 29.0% in Fabry patients under 65 years, compared with 23.1% in matched controls, yielding an adjusted OR of 1.489. Among those aged 65 years and older, diabetes prevalence rose to 46.6% in Fabry patients, again higher than 39.8% in controls, with an adjusted OR of 1.361. These findings indicate that younger patients with Fabry disease may face a relatively greater excess risk of diabetes compared with their older counterparts. This may have clinical implications for early screening strategies.
Although robust epidemiological data remain scarce, a few small-scale studies have suggested potential demographic and clinical risk factors associated with diabetes development in Fabry patients. Factors such as age, male sex, renal dysfunction and higher body mass index have been variably reported as contributors to glucose metabolism abnormalities in this population.11–15
Our findings align with these observations and further quantify the risk using a large matched cohort. Future studies incorporating detailed clinical metrics such as BMI, insulin resistance markers and longitudinal follow-up will be essential to confirm these associations and elucidate causal pathways.
Our study contributes to the limited body of research regarding a potential association between Fabry disease and diabetes mellitus development within this population by offering population-based evidence of a substantially higher prevalence of diabetes mellitus among individuals with Fabry disease. While renal involvement is recognised as a cardinal feature of Fabry disease since patients with characteristic albuminuria have been found and broadly reported, the findings of our study may help improve the outcomes of patients with Fabry disease by alerting clinicians to the potential for early detection of cooccurring diabetes and more aggressive management of diabetes. Such proactive efforts may mitigate or delay renal complications among patients with Fabry disease by minimising the potential additive renal impact of undiagnosed diabetes, potentially improving the overall prognosis of patients with Fabry disease.16
Our findings underscore the clinical importance of recognising the elevated risk of diabetes mellitus in patients with Fabry disease. Given the substantial morbidity associated with both conditions, their cooccurrence warrants early detection strategies and integrated management plans. Regular screening for diabetes in Fabry patients may enable timely interventions that reduce the risk of complications, particularly renal involvement, which is common to both diseases. Heightened clinical vigilance can therefore contribute to more comprehensive and effective care for this patient. Our study has significant strengths, notably, the use of population-based patient data from a large national database (Taiwan’s LHID2010) and the setting of a single-payer, NHI system. There are some study limitations. First, the study’s cross-sectional design does not permit inference of temporal relationships, notably causality between Fabry disease and diabetes mellitus. Although we found a substantially higher prevalence of diabetes among patients with Fabry disease, we cannot conclude whether Fabry disease causes a higher risk of diabetes, or if the presence of diabetes exacerbates the progression of Fabry disease. Second, the diagnosis of Fabry disease and diabetes mellitus was based on ICD codes documented in the LHID2010. While the use of ICD codes to identify disease status is the norm in epidemiological studies based on administrative claims data, ICD codes do not represent disease severity or the full clinical picture of the patients. Third, our study was conducted in a Taiwanese population, which might limit the generalisability of the findings to other ethnicities. Genetic and environmental factors, and potentially healthcare practices, differ from population to population, which may influence the prevalence of Fabry disease and its cooccurrence with diabetes mellitus. Fourth, our study lacks data on potential confounding variables, such as lifestyle factors (diet, physical activity, smoking status), family history of Fabry disease or diabetes and clinical parameters (eg, kidney function status, lipid profile, HbA1c level). These factors could influence the association between Fabry disease and diabetes mellitus. In addition, although this study used propensity score matching to reduce confounding on account of covariates by balancing the case and control groups on observable covariates, we did not account for the outcomes of Fabry disease treatment. Lastly, our study did not differentiate between type 1 and type 2 diabetes mellitus. The pathophysiology of these two types of diabetes is significantly different, and the relationship between Fabry disease and diabetes could potentially differ between these types. Future studies should consider exploring this aspect.
While our study provides a new insight on the prevalence of diabetes mellitus in patients with Fabry disease, further research is required to validate our findings and to document the underlying mechanisms that link these two conditions. Longitudinal studies with more detailed clinical data, and studies conducted among ethnically diverse populations are warranted. Future studies should also investigate the impact of this association on clinical outcomes and explore potential strategies for managing diabetes in patients with Fabry disease.
In conclusion, our findings highlight the importance of screening for diabetes mellitus in patients with Fabry disease. Physicians should be conscious of this association and be prepared to manage both conditions concurrently.
Data availability statement
Data may be obtained from a third party and are not publicly available. The data that support the findings of this study are available from HWDC, Department of Statistics, Ministry of Health and Welfare, Taiwan, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available.
Ethics statements
Patient consent for publication
References
Footnotes
Y-FC, H-CL and C-SC contributed equally.
Contributors Conceptualization: S-HH, H-CL, Y-FC and C-SC. Data curation: C-SC. Formal analysis: C-SC. Methodology: C-SC and H-CL. Project administration: H-CL. Software: C-SC. Validation: C-SC. Roles/writing–original draft: S-HH, SX, Y-FC, C-SC and H-CL. Writing–critical review for content and editing: S-HH and SX. All authors reviewed the manuscript. Guarantor: C-SC.
Funding This research is supported by the Taipei Medical University Wan-Fang Hospital Research Grant 114-wf-eva-14.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.