Psychological well‐being in patients with aneurysms‐osteoarthritis syndrome

Abstract Aneurysms‐osteoarthritis syndrome (AOS) is characterized by arterial aneurysms and dissection in combination with early‐onset osteoarthritis, which can impact quality of life. We describe the subjective quality of life and investigate anxiety and depression in 28 AOS patients aged 15–73 years. Three questionnaires were used: 36‐Item Short Form Survey (SF‐36), hospital anxiety and depression scale (HADS) and Rotterdam disease specific questionnaire. Results of the SF‐36 and HADS were compared to a reference Dutch cohort and the SF‐36 questionnaire also to patients with Marfan syndrome. Compared to the general population, AOS patients scored significantly lower on the following SF‐36 domains: physical functioning, vitality, social functioning, bodily pain, and general health. Physical functioning was also lower than in Marfan patients. Patients with AOS scored higher on the HADS depression scale, while anxiety did not show a significant difference compared to the general population. No difference in SF‐36 and HADS domain scores were found between patient with and without orthopaedic symptoms and patients with or without previous aortic surgery. Additionally, we found that patients' worries for their future and heredity of their disease are important factors for anxiety, which should be addressed in clinical practice.


| INTRODUCTION
An aneurysm or dilatation of the thoracic aorta can cause aortic dissection, which is a potentially life threatening event as over half of all patients with an acute thoracic aortic dissection die within 30 days (Melvinsdottir et al., 2016). In 20% of the patients an aortic aneurysm results from a heritable thoracic aortic disease (HTAD) (Caglayan & Dundar, 2009;Hannuksela, Stattin, Johansson, & Carlberg, 2015). In 2011, a new HTAD was described, the so-called "aneurysmsosteoarthritis syndrome (AOS)," caused by a pathogenic variant in the SMAD3 gene (van de Laar et al., 2011), which is part of the TGF-β pathway. Aneurysms-osteoarthritis syndrome has many similarities with 2 | MATERIALS AND METHODS

| Study population
All carriers of a pathogenic variant in the SMAD3 gene undergoing follow-up in our tertiary center per in-house protocol since January 2009 were invited for this study. Family members which were 50% risk carriers with obvious AOS related symptoms (aortic dilatation or osteoarthritis at an early age) were also included. Demographic and clinical data were obtained from the electronic patient files. Diabetes mellitus was defined as current use of medication. As part of our protocol, all patients underwent echocardiography and whole-body computed tomography angiography (CTA). The aortic measurements of the sinus of Valsalva, ascending aorta, aortic arch, and descending aorta were measured using the inner edge-to-inner edge method on the most recent CTA. Aneurysms and dissections were categorized by the following locations and definition: head and neck, thoracic, coronary, abdominal, leg and/or arm or pulmonary artery. Information on the following valvular, ventricular and arrhythmic abnormalities was collected: bicuspid aortic valve, aortic stenosis (Vmax >2.5 m/s), aortic regurgitation (at least moderate) (Lancellotti et al., 2013), valvular disease other than from the aortic valve, congenital heart disorders, ventricular hypertrophy (septal wall>13 mm), left ventricular dilatation (diastolic diameter >60 mm), and atrial fibrillation (former, paroxysmal or current). The study complied with the Declaration of Helsinki and was approved by the medical ethical committee of the Erasmus Medical Centre (MEC17-057). Written informed consent was provided by all patients.

| Questionnaires
All participants received three questionnaires: the Short Form-36 Health Survey (SF-36) (Ware & Sherbourne, 1992), the hospital anxiety and depression scale (HADS) (Snaith, 2003) and the Rotterdam disease specific questionnaire. The questionnaires were sent at first on the 14th of November 2017 and were collected until the 1st of February 2018. If participants did not respond at first, they received a maximum of two reminders. The SF-36 was used to determine patient-reported quality of life. It covers the following eight domains: physical functioning, role limitations due to physical health, bodily pain, general health, mental health, role limitations due to mental health, vitality (energy or fatigue related), and social functioning. The scale ranges from 0 through 100 points.
A lower score per subcategory reflects a lower quality of life on that life domain. In addition, two sum scores, the mental component summary (MCS) and physical component summary (PCS), were calculated (Ware JE, 1994). These summary scores are standardized according to the general Dutch population (Aaronson et al., 1998), which means that all scores above and below 50 are above and below the average in the Dutch population. The HADS questionnaire determines the levels of anxiety and depression on two subscales with a total score ranking from 1 to 21.
A score of 0-7 for either subscale is in the normal range, a score of 8-10 is possible abnormal and 11 or higher indicates the probable presence of anxiety or depression. The "Rotterdam disease specific questionnaire" was developed by our multidisciplinary team in the Erasmus Medical Center (Supporting information) to investigate the impact of having AOS related aortic aneurysm on daily life, work participation, sexual functioning, pregnancy wish, and sports participation. Patients received 18 statements and were asked to grade how they felt on a continuous scale from 0 to 10, 0 being "I completely disagree" and 10 being "I completely agree".

| Comparison with the general population and other aortic disease patients (Marfan syndrome)
For the HADS and SF-36 questionnaires we compared our data to the reference values of the age-matched general Dutch population (Aaronson et al., 1998;Spinhoven et al., 1997). For the SF-36, a cohort of age and sex matched Marfan patients was also available, which allowed us to compare AOS with another syndromic HTAD (Rand-Hendriksen et al., 2010). For the results of the Rotterdam disease specific questionnaire, there are no reference values available yet, because this questionnaire was newly developed for this study's aim.

| Statistical analysis
Continuous variables with a normal distribution were reported as mean with ± SD and the median and interquartile range was reported in case of non-normal distribution. Categorical variables were summarized as frequencies and percentages. Data distribution was checked using histograms. Because of the non-normally distribution of the values in the domains of the SF-36 and HADS questionnaires, the median and interquartile ranges are presented in table 1 and the p-value of the onesample Wilcoxon signed rank test was presented in the text and figures.
With the one-sample Wilcoxon signed rank test the median of a continuous variable in our cohort was compared with a hypothesized median of a reference group. Since our reference article showed their values only with mean ± SD, we assumed that the variables were distributed normally. Because mean and median are comparable in normally distributed variables, we used the mean of the reference as hypothesized median in our nonparametric test. To visually compare our data with reference data presented as mean ± SD, also our data were presented as mean ± SD in the figures although we could not proof normal distribution. However, only small differences were found between the calculated mean and

| Study population
Of the 31 patients with AOS in our center, 28 patients (90%) agreed to participate and returned the questionnaires. The other three patients were approached two times, but could not be reached (n = 2) or decided not to participate due to time-constraints (n = 1). There were no major differences between the 28 responders and 3 nonresponders. The baseline characteristics of the 28 study patients are presented in Table 2. Our cohort contained 23 participants with a confirmed SMAD3 mutation representing 10 different genetic mutations with the most common heterozygous mutation (R287W, 859C>T, SMAD3 ex 9) present in 9 patients. The mean age was 44.0 ± 17.3 years with an age range from 15 to 73 years. Of the 28 patients, 17 (61%) were women. Cardiac or vascular abnormalities were present in 22 (79%) patients and orthopaedic symptoms were present in 24 (86%) patients. In 18 (64%) of the 28 patients both cardiovascular manifestations and orthopaedic symptoms were reported.

| Quality of life, anxiety, and depression
The median values with interquartile range of the domains from the SF-36 and HADS questionnaires are presented in Table 1. Our cohort scored significantly lower compared to the age-matched reference group of 1,742 Dutch citizens (Aaronson et al., 1998) (Aaronson et al., 1998). The median PCS score of our patients with pathogenic variants in the SMAD3 gene was 34.3, which means a much lower physical health than the general population, while the median MCS score (50.4) showed that mental health was comparable with the general population. However, on two subscales of mental health, namely vitality and social functioning, AOS patient did score lower than the general population. We know that middle aged Dutch patients with congenital heart disease show similar or even more favorable levels compared to normative data using the SF-36 survey (Opic et al., 2016 (Loeys et al., 2010;. This assumption is supported by data from patients with Ehlers-F I G U R E 1 Comparison of eight SF-36 domains between AOS patients, Marfan patients and healthy control. Data is shown as mean (incl. 95% CI). The SF-36 scale ranges from 0 through 100 points. The lower the point count per subcategory, the more prevalent it is that the individual has a negative effect of that sub scale's premise. For social functioning and general health one patient was missing, because he forgets to fill in one page of the questionnaire. * Significant lower compared to the mean of the general population (One-sample Wilcoxon signed rank test). † Significant lower compared to both the mean of the general population and the mean of patients with Marfan syndrome (One-sample Wilcoxon signed rank test). AOS, aneurysms-osteoarthritis syndrome; BP, bodily pain; GH, general health; HADS, hospital anxiety and depression scale; MH, mental health; PF, physical functioning; RM, role limitations due to mental health; RP, role limitations due to physical health; SF, social functioning; VT, vitality F I G U R E 2 Comparison of two HADS domains between AOS patients and healthy control. Data is shown as mean (incl. 95% CI). The anxiety and depression sub scales have a point system of 0 through 21 points total. The higher the point count for any sub scale, the more likely that the individual suffers from anxiety or depression. * Significant higher compared to the general population (one-sample Wilcoxon signed rank test) Danlos syndrome, a connective tissue disorder which presents with extreme musculoskeletal symptoms including hypermobility (De Paepe & Malfait, 2012). Ehlers-Danlos patients report even lower physical function score (39.6) and general health score (26.8) compared to AOS patients (Berglund et al., 2015;Bovet, Carlson, & Taylor, 2016). We did not find a difference between patients with and without orthopaedic symptoms.
However, this should be tested in larger cohorts, since our cohort might be too small to prove the association between orthopaedic symptoms and reduced quality of life. In conclusion, AOS patients and probably also patients with other heritable thoracic aortic disease, have lower quality of life than the general population, but there seem to be some differences between syndromes. By assessing quality of life, anxiety, and depression, we found unfavorable outcomes and impairments, warranting attention and in some cases treatment. In our population of patients with pathogenic variants in the SMAD3 gene, 4% scored above the cut-off for clinical anxiety, while 7% scored in the range for clinical depression. These percentages are comparable to or even slightly lower than the prevalence's in the general German population, which are 5.2% for anxiety and 9.6% for depression based on the HADS questionnaire (Hinz & Brahler, 2011 questionnaires, such as the Rotterdam disease specific questionnaire, in larger cohorts we will be able to identify the cause of physiological burden in a disease more precisely in the future. Nevertheless, before using these questionnaires in clinical practice, they should be validated.

| Limitations
In this study, we used questionnaires to measure self-reported quality of life, anxiety, and depression, which may have caused documentation of more complaints than patients would have mentioned spontaneously.
Also, it may cause some information bias, although we assume that the high response rate of 90% reduced the chance of bias. Because of our single center design, we included a small cohort, which prevented us from extensive identification of factors associated with quality of life, anxiety, or depression.

| Conclusion
In conclusion, our population of AOS patients showed reduced quality of life in comparison with the general population on physical functioning, role limitations due to physical functioning, bodily pain, general health, F I G U R E 3 Results of the Rotterdam disease specific questionnaire. This figure shows the influence of (possible) aortic dilatation or dissection on different aspects of life. Data shown as median with interquartile range 25-75% vitality, and social functioning. Physical functioning was also lower than in Marfan patients. Although the prevalence of depression was similar to the general population, patients with AOS scored significantly higher on the depression scale, which physicians must be aware of to provide good medical care. Additionally, we found that patients' worries for their future and heredity of their disease are important factors for anxiety, which should be addressed in clinical practice.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of this article.