Epidemiology of Marfan syndrome: a retrospective cohort study based on the Korean population:
A retrospective cohort study
Article information
Abstract
Purpose
To determine the risk of incidence, prevalence, survival, and death risk for Marfan syndrome (MFS) in Korea.
Methods
MFS (10th revision of the International Statistical Classification of Diseases [ICD-10] code: Q87.4) data were collected from the Korea National Health Insurance Service between 2006 and 2017. Simple and multiple Cox proportional hazards analyses were performed. The death data of Koreans from 2006 to 2018 were used.
Results
The mean age of patients with MFS was 27.2± 16.8 years, 26.5± 15.9 years for males and 28.2± 18.1 years for females (P < 0.01), and the proportion of males was 59.5%. The age-standardized incidence and prevalence in 2017 were 0.51 and 5.70 per 100,000 persons, respectively. The 10-year survival rate was 92.7%. Adjusted hazard ratio increased significantly with age, male sex, heart failure, ischemic stroke, hemorrhagic stroke, chronic kidney disease, malignant neoplasm, aortic dissection, and abdominal aortic aneurysm.
Conclusion
Although the overall 10-year survival rate of MFS was over 90%, the adjusted hazard ratio was significantly higher in the older age group, male sex, and individuals with heart failure, stroke, chronic kidney disease, malignant neoplasm, aortic dissection, and abdominal aortic aneurysm. These results suggest the need for meticulous surveillance of aortopathy in patients with MFS, as it may portend a worse prognosis in aortic dissection and aneurysm, and in older age groups with MFS.
INTRODUCTION
Marfan syndrome (MFS) is an autosomal dominant disorder caused by mutations in the fibrillin-1 gene on chromosome 15. Without proper management, MFS can lead to early or sudden death due to its impact on the circulatory system, which may result in mitral valve prolapse, aortic regurgitation, dilated aorta, aortic aneurysm, or aortic dissection. The symptoms and prognosis of each patient differ depending on the extent of the aortic and/or heart damage. However, the temporal trends in the prevalence, incidence, and mortality of MFS have not been well established in East Asian populations. Our study revealed that the overall age-standardized cumulative prevalence of MFS increased between 2006 and 2011, with the most notable rise seen in males aged 15–19 and female aged 10–14 [1]. Building on theses findings, we analyzed Korean National Health Insurance Service (KNHIS) data from 2006 to 2017 to assess age-standardized incidence, age-standardized prevalence, survival rate (SR), causes of death, and death risk associated with MFS.
METHODS
Study population
The KNHIS database for health insurance beneficiary and medical aids beneficiary excluding foreigners consists of the following four main components: (1) a qualification database with demographic information such as age, sex, type of beneficiary, and income rank; (2) a medical check-up database, including results from health examination and lifetime transition period medical check-ups at 40 and 66 years of age; (3) a medical care institution database; and (4) a treatment database containing disease information and 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes. The treatment database is divided into medicine, dentistry, oriental medicine, and pharmacy; our analysis focuses on medicine data only. We utilized variables from the qualifications database in conjunction with treatment data [2-4].
Diagnosis of MFS
Newly diagnosed MFS data were collected from KNHIS records from 2006 to 2017 (n= 2,944). The data consisted of primary and secondary diagnoses related to MFS, classified according to ICD-10. Newly diagnosed MFS was operationally defined using the ICD-10 code Q87.4. To include only newly diagnosed patients with MFS from the KNHIS database, those diagnosed before the corresponding year were excluded.
Definition of variables
Age was categorized as follows: 0–9, 10–19, 20–29, 30–39, 40–49, 50–59, 60–69, and 70–79 years. Socioeconomic position was defined according to the income level derived from the national health insurance premium, and categorized as upper, medium, and lower. According to the household insurance premium, members of the same household are included in the same income category depending on whether they are employee-insured or self-employed.
Comorbidities
We described comorbidities from the primary and secondary diagnoses. These included hypertension (ICD-10: I10, I11, I12, I13, I15); diabetes mellitus (ICD-10: E10, E11 E12, E13, E14); dyslipidemia (ICD-10: E78); myocardial infarction (ICD-10: I21, I22, I25.2); heart failure (ICD-10: I11.1, I50, I97.1); atrial fibrillation (ICD-10: I48); paroxysmal tachycardia (ICD-10: I47); other arrhythmias including atrioventricular and left bundle branch block (ICD-10: I44), other conduction disorders (ICD-10: I45), and cardiac arrhythmias; (ICD-10: I49); syncope (ICD-10: R55); ischemic stroke (ICD-10: I63, I64); hemorrhagic stroke (ICD-10: I60, I61, I62); transient ischemic attack (ICD-10: G45); chronic kidney disease (ICD-10: N18, N19); malignancy neoplasm (ICD-10: C00–C97); infective endocarditis (ICD-10: I38, I39); pneumothorax (ICD-10: J93); other lens disorders (ICD-10: H27); other eye diseases including retinal detachment with retinal break (ICD-10: H33), disorders of vitreous body (ICD-10: H43), and visual disturbance and blindness and low vision (ICD-10: H53, H54); scoliosis (ICD-10: M41); dissection of aorta (ICD-10: I71.0); thoracic aortic aneurysm (ICD-10: I71.1, I71.2); abdominal aortic aneurysm (ICD-10: I71.3, I71.4); thoracoabdominal aortic aneurysm (ICD-10: I71.5, I71.6); and aortic aneurysm of unspecified site (ICD-10: I71.8, I71.9).
Death data
We obtained the death data of Korean people from the Statistics Korea between 2006 and 2018. In accordance with official protocols, we received only information on the cause and date of death [3].
Cause of death
MFS is a rare disease, and the use and disclosure of health data are restricted due to privacy concerns surrounding personal information. To assess the primary cause of death in our study cohort, we relied on ICD-10 codes; certain infections and parasitic diseases (ICD-10: A00–B99); malignant neoplasm (ICD-10: C00–C97); endocrine, nutritional, and metabolic diseases (ICD-10: E00–E90); mental and behavioral disorders (ICD-10: F01–F99); diseases of the nervous system (ICD-10: G00–G98); diseases of the circulatory system (ICD-10: I00–I99); diseases of the respiratory system (ICD-10: J00–J98); diseases of the digestive system (ICD-10: K00–K92); diseases of the musculoskeletal system and connective tissue (ICD-10: M00–M99); congenital malformation, deformations and chromosomal abnormalities (ICD-10: Q00–Q99); symptoms, signs, and abnormal clinical and laboratory findings, not elsewhere classified (ICD-10: R00–R99); injury, poisoning, and certain other consequences of external causes (ICD-10: S00–T98); and not provided.
Statistical methods
Differences in characteristics by sex were analyzed using Student’s t-test for continuous variables and the chi-square test for categorical variables. The age-standardized incidence and prevalence of MFS were determined through a direct method, based on beneficiaries of health insurance from the 2006 to 2017 KNHIS yearbook and the estimated Korean population in 2015 [1]. SR among MFS patients, stratified by age and sex, were assessed using the Kaplan–Meier method, with comparisons made through log-rank tests. Simple and multiple Cox proportional hazards analyses were performed using age, sex, income level, and comorbidities such as hypertension, diabetes mellitus, heart failure, ischemic stroke, hemorrhagic stroke, chronic kidney disease, malignant neoplasm, pneumothorax, other disorders of lens, dissection of aorta, thoracic aortic aneurysm, thoracoabdominal aortic aneurysm, abdominal aortic aneurysm, and aortic aneurysm at an unspecified site [5].
Ethics statement
This study, which involved human participants, was reviewed and approved by the Institutional Review Board of Samsung Medical Center (IRB file no. 2017-02-032). As we used secondary processed national health big data, the requirement for informed consent was waived, given that the study posed no more than minimal risk to participants. Exemption from consent did not adversely affect the rights or well-being of the study subjects. Given the timeframe, obtaining individual consent was impractical, with no reasonable basis to assume participants refusal. Furthermore, the study posed a very low level of risk to participants, even without explicit consent.
RESULTS
Population, comorbidities, cause of death, and death risk
The mean age of patients of newly diagnosed with MFS was 27.2± 16.8, 26.5± 15.9 years for males and 28.2± 18.1 years for females (P = 0.007), with the proportion of males being 59.5%. Patients with MFS aged < 40 years made up 74.8% (77.8% of males and 70.6% of females, P < 0.001). The percentage of comorbidities with MFS was 14.2% for hypertension, 1.73% for diabetes mellitus, 4.04% for dyslipidemia, 0.31% for myocardial infarction, 2.99% for heart failure, 2.88% for arrhythmias, 0.54% for ischemic stroke, 0.14% for hemorrhagic stroke, 0.03% for transient ischemic attack, 0.17% for chronic kidney disease, 0.41% for malignant neoplasm, 5.54% for eye problems, 2.85% for scoliosis, 10.2% for aortic dissection, and 7.4% for aortic aneurysm. Approximately 6% of the patients with MFS died between 2006 and 2017. The most common cause of death was disease of the circulatory system (Table 1). Supplementary Table 1 presents the yearly distribution of general and clinical characteristics of patients with newly diagnosed MFS over a decade.
Distribution of general characteristics, socioeconomic position, and causes of death based on sex at incidence
The adjusted hazard ratios (HR) and 95% confidence interval (CI) for MFS were higher in individuals aged 30–39 years (2.30; 95% CI, 1.22 to 4.31), 40–49 years (3.80; 95% CI, 2.07 to 7.00), 50–59 years (3.47; 95% CI, 1.76 to 6.86), 60–69 years (10.7; 95% CI, 5.40 to 21.5), and 70–79 years (11.9; 95% CI, 4.38 to 32.6); in males (1.75; 95% CI, 1.25 to 2.45); and in those with heart failure (2.59; 95% CI, 1.52 to 4.43), ischemic stroke (3.61; 95% CI, 1.44 to 9.04), hemorrhagic stroke (10.3; 95% CI, 2.51 to 42.8), chronic kidney disease (16.7; 95% CI, 4.58 to 61.3), malignant neoplasm (3.95; 95% CI, 1.21 to 12.8), dissection of aorta (2.52; 95% CI, 1.70 to 3.72), and abdominal aortic aneurysm (4.50; 95% CI, 2.21 to 9.16) (Table 2).
All-cause death risk in patients with Marfan syndrome (2006–2017, n=2,944)
Incidence and prevalence
Overall, the age-standardized incidence of MFS was 0.48 per 100,000 persons in 2006 and 0.51 per 100,000 persons in 2017 (Fig. 1A, B and Supplementary Table 2). Additionally, the age-standardized prevalence increased from 1.93 per 100,000 persons in 2006 to 5.70 per 100,000 persons in 2017 (Fig. 1C, D and Supplementary Table 3). Additionally, Fig. 1E showed distribution of MFS by age group and sex during a decade.
Distribution, age-standardized incidence, and prevalence of Marfan syndrome (MFS) per 100,000 persons between 2006 and 2017. (A) Incidence number and age-standardized incidence: overall, by sex (male: upper; female: lower), and by year. (B) Age-standardized incidence: overall, by age group, and by year. (C) Prevalence number and age-standardized prevalence: overall, by sex (female: upper; male: lower), and by year. (D) Age-standardized prevalence: overall, by age group, and by year. (E) Distribution of MFS by age group and sex.
Survival in MFS
The 1-, 3-, 5-, and 10-year SRs of MFS were 98.1%, 96.9%, 95.7%, and 92.6%, respectively. Females exhibited SRs of 98.4%, 97.3%, 96.5%, and 94.5%, respectively, while males showed 97.9%, 96.6%, 95.2%, and 91.4% (P= 0.024). In the 60 to 69 age group, the SRs were 93.9%, 87.9%, 80.5%, and 68.3%, respectively, and in the 70 to 79 age group, they were 96.0%, 90.6%, 71.1%, and 56.8% (P< 0.001) (Fig. 2 and Supplementary Table 4).
Survival curve of Marfan syndrome (MFS) in Korea. (A) MFS survival rates overall and by sex (P=0.024). (B) MFS survival rates by age group (P<0.001).
DISCUSSION
Aortic dissection and aortic abdominal aneurysms are associated with a significantly higher risk of death in patients with MFS. However, few studies have examined the HR for these conditions in MFS using big data on a nationwide scale. Nevertheless, our findings are consistent with a previous study that reported an HR of approximately three for type A aortic dissections among 172 patients with MFS in China, following aorta replacement surgery [6].
In our study, the proportion of initial aortic dissection in MFS was 10.2%, which is similar to the 10% estimate reported by a Taiwanese study of 2,329 patients, conducted from 2000 to 2012 using a nationwide database [7]. However, a Danish nationwide cohort from 1994 to 2014 showed a proportion of up to 36.4% [8], which was higher than ours. In an additional analysis based on age group, we found that the proportion of aortic dissection in MFS increased with age (Supplementary Table 5). This result is similar to that of a study of 4,286 individuals from the Taiwan National Database, which found that freedom from dissection was 99%, 80%, and 66% for groups in their 20s, 40s, and 50s, respectively [7]. However, we cannot make an accurate comparison with the Taiwanese study owing to differences in statistical analysis methods. Nonetheless, the study also found that the SR in aortic dissection decreased with age [7].
Our study showed a higher HR for abdominal aortic aneurysms in MFS, which is consistent with the findings of a few previous studies. However, a study of 268 patients with MFS enrolled in the Euro Heart Survey on adult congenital heart disease found that an increased diameter of the descending aorta was associated with a higher risk of aortic events in patients without previous dissection or aortic root replacement, independent of the diameter of the aortic root [9]. Our study also demonstrated that the adjusted HR for MFS increased with age and in male sex [8,10]. Furthermore, the adjusted HRs for MFS were significantly higher with heart failure, stroke, chronic kidney disease, and malignant neoplasms. However, no studies with comparable findings could be found.
In this study, the overall 10-year SR of MFS was greater than 90% and deaths were rarely observed in patients aged 0–19 years. However, the 10-year SR was found to decrease with increasing age, which aligns with a nationwide study of 627 Danish patients with MFS conducted between 1977 and 2014 [11]. The most common cause of death in the patients was disease of the circulatory system, accounting for over 40% of the cases. This finding is in close agreement with that of an MFS cohort in Norway from 2003 to 2004 (n = 84), in which cardiovascular disease was the most common cause of death, accounting for 68.8% [12].
The distribution of MFS was the highest among teenagers of both sexes, with males having a higher distribution than females. The age-standardized incidence and prevalence rates of MFS were also high in teenagers, which is consistent with the Taiwanese study [7]. These results suggest that MFS is often diagnosed in teenagers because it grows quickly during this stage of development [1], as supported by the higher distribution rate among males in the Danish national registry [11].
This study had several limitations. First, the KNHIS data may have missed potential patients with MFS who did not use medical services or who covered medical expenses independently. Consequently, the incidence, prevalence, or SR of MFS in our findings might be inaccurately estimated. Second, because of limited data, we were unable to include key variables in the HR analysis, such as aorta dilation size, type A or B aortic dissection, aortic surgery status, and use of circulatory system medications such as β-blockers and/or angiotensin II receptor blockers.
In conclusion, the age-standardized incidence and prevalence rates of MFS in 2017 were 0.5 and 5.7 per 100,000 persons, respectively. Although the overall 10-year SR of MFS was over 90%, the adjusted HRs for MFS were significantly higher in older age groups, the male sex, and patients with aortic dissection and aneurysm. These findings suggest a poorer prognostic factor and worse prognosis in older patients, especially those with aortic dissection and aortic abdominal aneurysms. Therefore, we recommend careful and regular surveillance of aortopathy in patients with MFS.
Supplementary materials
Distribution of general and clinical characteristics of patients newly diagnosed with Marfan syndrome between 2006 and 2017
Age-standardized incidencea) and 95% CI of Marfan syndrome overall and by sex (per 100,000 persons)
Age-standardized prevalencea) and 95% CI of Marfan syndrome, overall and by sex (per 100,000)
Survival rate (%) and 95% CI of Marfan syndrome, overall, by sex, and age group
Distribution of general characteristics of MFS caused by AD (2006–2017, n=2,944)
Notes
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: SYJ, DKK, TKP.
Acquisition, analysis, or interpretation of data: SYJ, JH, DKK, TKP.
Drafting the work or revising: SYJ, JH, DKK, TKP.
Final approval of the manuscript: SYJ, JH, DKK, TKP.
Acknowledgements
This study used data from the National Health Insurance Service (research management number NHIS-2019-1-147); however, the study results are not related to the National Health Insurance Service.