Oral symptoms and oral health related quality of life in people with Ehlers-Danlos syndromes (EDS): a questionnaire-based cross-sectional study


 Background: To date, there have only been a few studies on the oral health-related quality of life (OHRQoL) of people with Ehlers-Danlos syndromes (EDS) and the oral symptoms. The aim of this study was, therefore, to analyse the OHRQoL of people with EDS from their own point of view as well as obtain information about their age at the time of the diagnosis, the period of time until diagnosis, and the presence of oral symptoms (if any) and their association with oral health quality.Methods:The study was designed as an anonymous questionnaire-based cross-sectional study. We conducted a descriptive analysis of the Oral Health Impact Profile-14 (OHIP-14) scores, age of the participants, age at diagnosis, and the time period between the first signs of the disease and the diagnosis of EDS. To verify the differences in OHIP-14 scores between patients with and without oral symptoms, a Mann-Whitney U test was performed.A multivariate quantile (median) regression analysis was performed to evaluate the effect of different general characteristics (gender, age, and the presence of oral symptoms) on the OHIP‑14 scores. Furthermore, using a Mann-Whitney U test, the influence of different oral symptoms was verified by testing the differences between patients without any oral symptoms and patients with a specific diagnosis.Results:A total of 79 evaluable questionnaires from 66 female (83.5%) and 13 male (16.5%) participants were analysed. On average, after the first symptom, it takes 18.36 years before EDS are correctly diagnosed.Oral symptoms were described by 69.6% of the participants. The OHIP-14 score was 8 (10) points for patients without oral symptoms and 19 (15) for patients with oral symptoms. The quantile regression for OHIP-14 scores, depending on oral symptoms, was highly significant (p = 0.0007). OHIP-14 scores for dysgnathia, periodontitis, TMD, a high-arched palate, malocclusion, and the anomaly of tooth formation were significantly different between the participants with and the participants without oral symptoms.Conclusion:Long diagnostic pathways seem to be a typical problem in patients with EDS. Oral symptoms associated with the underlying disease occurred regularly and showed a negative correlation with OHRQoL.


Conclusion:
Long diagnostic pathways seem to be a typical problem in patients with EDS. Oral symptoms associated with the underlying disease occurred regularly and showed a negative correlation with OHRQoL.

Background
Ehlers-Danlos syndromes (EDS) are a clinically and genetically heterogeneous group of hereditary connective tissue disorders involving joint hyperlaxity, cutaneous hyperelasticity, and tissue fragility.
[1, 2] A revised classification containing 13 subtypes was published by the International EDS Consortium, with a number of clinical symptoms to guide and improve the diagnosis of each subtype. [2] In the European Union, a disease is considered 'rare' if it affects fewer than 1 in 2000 people. [3] In the case of EDS, the prevalence varies, depending on the type: between 1:30,000 (classical type) and < 1:1,000,000 (arthrochalasia type), whereas for the rarest form (the periodontitis type), the prevalence is unknown. [1] EDS' oral symptoms have been described as periodontitis, [2,4,5] temporomandibular dysfunctions, [2,6] bleeding tendencies, [7,8] enamel hypoplasia, [2,9] shape abnormalities of the teeth and changes in the number of teeth, [2,8] a high palate, [2,10] dysgnathia, and malocclusion, [2,11] as well as decreased effects of local anaesthesia. [12,13] To date, there have been only a few studies on the oral health-related quality of life (OHRQoL) of people with EDS who report reduced OHRQoL. [6,14,15] A proven method for evaluating OHRQoL is the Oral Health Impact Profile-14 (OHIP-14) questionnaire, which measures the frequency of 14 different functional and psychosocial factors influencing OHRQoL. [16] This study aimed to analyse the OHRQoL of people with EDS from their own point of view as well as obtain information on their age at the time of the diagnosis, the period of time until diagnosis, and presence of oral symptoms (if any) and their association with oral health quality.

Study design
The study was designed as an anonymous questionnaire-based cross-sectional study involving people 4 with EDS to evaluate their respective OHRQoL.
A questionnaire consisting of free text questions about age, gender, oral symptoms (if any), and the period between the first symptom of the disease and its diagnosis was developed for this purpose. In addition, the standardised German version of the OHIP-14 questionnaire was added to the evaluation of the OHRQoL [17].
Each of the 14 questions in the OHIP-14 questionnaire was assigned standardised numerical values: 0 = never, 1 = hardly ever, 2 = sometimes, 3 = often and 4 = very frequently. It was possible to score a total of zero and a maximum of 56 points. The higher the score, the worse the OHRQoL. The questions referred to experiences registered over the past month.
The study was presented at the annual meeting of the 'Ehlers-Danlos Initiative e.V.' self-help group on September 22nd, 2018, in Bad Kissingen, Germany. The participants were asked to send the completed questionnaire, which was provided online to the members by the self-help group either by email or by post to the study organisers. Responses were taken into account until March 22nd, 2019. Vienna, Austria) were used for statistical analysis. All the analyses were exploratory, not confirmatory, and adjustments for multiple testing were not performed.
A descriptive analysis of the OHIP-14 scores, participants' ages, age at diagnosis, and time period between the first signs of the disease and the diagnosis of EDS was conducted by determining the mean, median, minimum, maximum, standard deviation (STD), and interquartile range (IQR) for both the males and the females. To verify if there were differences between the OHIP-14 scores of patients 5 with and patients without oral symptoms, a MannWhitney U test was performed. A multivariate quantile (median) regression analysis was performed to evaluate the effect of different general characteristics (gender, age, and the presence of oral symptoms) on the OHIP-14 scores.
Furthermore, using a Mann-Whitney U test, the influence of different oral symptoms was verified by testing if there were differences between patients without any oral symptoms and patients with a specific diagnosis.
The researchers compared participants who reported having EDS-related symptoms to participants who reported having none; this practice was in favour of comparing two subgroups that declared being symptomatic or asymptomatic. The evaluation was completed this way, as the respondents may have had different combinations of symptoms and the control group would have been too heterogeneous (i.e., including participants having no symptoms and participants having other oral symptoms). However, comparisons between individual symptoms were not useful for the present cohort.

Results Participants
From all participants, a total of 79 evaluated questionnaires were received. Participants consisted of 66 female (83.5%) and 13 male (16.5%). The gender-independent median (IQR) of age was 38 (25) years (range: 16-81 years). The median (IQR) age for the female participants was 38 (26) years and the median (IQR) age of the male participants was 42 (29) years.

Age at the time of diagnosis and the time taken for diagnosis
The gender-independent median (IQR) age at which EDS was diagnosed was 28.5 (26) years (range: 0-79 years). The median age at diagnosis was 29 (24) years for females and 28 (36) years for males.

Reported oral symptoms
When asked about oral symptoms, 79 participants provided responses, 55 (69.6%) of which described oral symptoms. Thirty-seven respondents cited oral symptoms as various forms of skeletal 6 malocclusion and growth aberrations (e.g., high palate, micrognathia, macrognathia, microgenia or 'jaws do not fit together'), while a high palate was explicitly mentioned by 18 respondents. Toothrelated dysgnathia (malocclusions) were also described by 16 participants. There were also combined forms of skeletal and dentoalveolar dysgnathia among the respondents. Craniomandibular dysfunctions occurred in 21 of the respondents, according to the data. Gingivitis/periodontitis was also mentioned as oral symptoms by 17 respondents, shape abnormalities of the teeth were described by 11 respondents, and a decrease in the number of teeth (hypodontia/ oligodontia) was described by a total of 6 respondents. Figure 1 provides an overview of the symptoms described by the respondents. Figure 2 shows the OHIP-14 scores and their distribution based on the manifestation of oral symptoms. In the group of participants without any oral symptoms, a median score of 8 (10) points was found. In contrast, in the group of participants with reported oral symptoms, the median of the OHIP-14 score was 19 (15) points (p < 0.001).

OHIP-14 scores
Furthermore, the influence of gender, age and the presence of oral symptoms were analysed by multivariate quantile regression. However, age and gender do not seem to influence the OHIP14 score. The resulting coefficients can be found in Table 1. Influence of OHIP-14 score on the attendance of oral symptoms All the oral symptoms that were named by more than 10 respondents were further analysed. The OHIP-14 scores for these symptoms and for the group of participants without oral symptoms are plotted in Fig. 3.
It is remarkable that for all the symptoms that were considered, the median score was significantly higher than the score for participants without oral symptoms (all the p-values were sufficiently small).
In Table 2, the median, mean, and p-values of the described comparisons are given for all the symptoms considered.

Discussion
The literature has described a fundamental problem concerning the period of time until the diagnosis of rare diseases, in this case, seven years on average. [18] In our group of respondents, a period of 18.36 years between the onset of the first symptom and correct diagnosis of the disease was described by those affected with EDS, regardless of gender. Those affected were on average 28.5 years old. We could identify only one study covering the age of diagnosis of people affected by EDS. [19] The study found that 28 years passed between the first symptom and the diagnosis of EDS, after consulting 16 different doctors. Our results thus confirm the problem of significant delays in diagnosing EDS. It, therefore, seems obvious that those affected by EDS will have to endure a period of unnecessary suffering before finally being diagnosed.
The partial overlap of different phenotype features between the 13 subtypes of EDS contributes to the difficulty of diagnosis. [20] The symptoms are also sometimes misinterpreted as 'psychosomatic' or 'nonexistent'. [21] To speed up the diagnostic process in the future, medical education, publications, and guidelines must inform medical doctors and dentists about the disease and its symptoms.
More than two-thirds of the respondents described oral symptoms. Especially in rare diseases, the oral symptoms, such as those described in EDS, [4,5,6,7,8,9,10,11] may provide an important indication of the underlying disease. Dentists who correctly identify oral symptoms can significantly contribute to earlier diagnosis of the underlying disease. 8 The oral symptoms described by the respondents in this study are manifestations that are already known, which may occur in EDS. However, the correct terminology cannot always be expected because the symptoms have not been clinically verified. When categorising the symptoms, the information provided by the participants was translated and assigned a technical term accordingly.
For example, some study participants referred to the symptom 'upper jaw too small', which was recently assigned the term 'micrognathia' in the 'dysgnathia' category. Other participants only stated the symptom 'the jaws don't fit together', which was also assigned to the 'dysgnathia' category. The detailed assignment of oral symptoms must, therefore, be regarded as having certain limitations.
In general, our study showed that OHIP values among participants who described oral symptoms and those who did not report oral symptoms were statistically noticeable. For example, the OHIP values for those with oral symptoms were, on average, 21.35 points. While for those with no oral symptoms, the average was 10.13 points.
On average, the OHIP-14 overall score of the participants was 18.25 points, regardless of gender. A Swedish study [14] reported an average overall score of 11.1 points regardless of gender and a total score of 11.8 points among female participants who made up the majority of the study. Thus, the values in the present study were above the results obtained by Berglund, [14] who reported worse OHRQoL among people with EDS compared to the standard population.
In a study by John et al. [22], an average OHIP-14 score of 4.09 points was determined for the German standard population. Thus, these results show that, as also determined by Berglund  This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors' contributions
MH, JK and JKL conceived the study. MH, JK, MBL, and DS helped in the acquisition and interpretation of data. JK conducted the statistical analysis. MH, MM, JK, MBL, LB and DS participated in the literature review, design, and drafting of the manuscript. All authors read and approved the final manuscript.