Health-related quality of life and its influencing factors in adult patients with localized scleroderma - a cross-sectional study.

Background Localized Scleroderma (LoS) is an autoimmune connective tissue disease that affects skin and less commonly subcutaneous tissues. The illness occurs in children and adults, and may have a serious impact on health-related quality of life (HRQoL). The goal of this study was to explore what factors might deteriorate scores on HRQoL measures in adult LoS patients. Methods Detailed information on the demographic and clinical features of LoS patients was collected. The HRQoL was assessed using Skindex-29 and Short form 36 (SF-36) questionnaires. Results Thirty three women and seven men with LoS were enrolled. Female gender negatively influenced scores for the emotion subscale of Skindex-29. Multiple-factor linear regression analysis confirmed, as with single-factor analysis, that the causes of low SF-36 physical component score (PCS) were subjective symptoms (pruritus, pain, paresthesia), musculoskeletal manifestations and older age at the time of survey, while rural area of residence negatively affected the SF-36 mental component score. Additionally, single-factor analysis showed that the SF-36 PCS was related to the LoS cutaneous assessment tool (LoSCAT) summary score. Conclusions Apart from a clinical presentations, sociodemographic characteristics, including older age, female gender and living conditions, may impair HRQoL of LoS patients. Further studies that will examine the role of these factors for physical and mental functioning of adults with LoS are needed.


Introduction
Localized scleroderma (LoS) is an autoimmune disorder characterised by inflammation and sclerosis of the skin and less commonly subcutaneous fat, in some cases only affecting the fasciae, muscles and bones. It is not a lifethreatening illness, but health related quality of life (HRQoL) of patients is often negatively influenced [1]. The early skin lesions are erythematous plaques that become progressively indurated. After several months to years they may resolve but skin and/or deeper tissues atrophy and hyper/hypopigmentation remain. Additionally, linear and generalised subtypes are usually associated with functional impairment i.e. arthralgia and/or arthritis, joint contractures of the affected limb, limb length discrepancy and physical disability [1,2]. Neurologic symptoms (e.g. seizures/epilepsy, headaches/migraines, stroke) and/or ophthalmology conditions such as uveitis may occur in LoS of the head or face (en coup de sabre subtype, progressive hemifacial atrophy) [3]. Although the linear subtype, musculoskeletal, neurologic and ophthalmologic involvement predominate among children, there are several reports of low health related quality of life (HRQoL) in adults patients [4][5][6][7][8][9][10][11]. Klimas et al. [5] reported that skin-specific HRQoL of adult LoS patients was worse than those with non-melanoma skin cancer, vitiligo, and alopecia. What is more, LoS impacted in the emotions and mental health domains of HRQoL, similarly to rheumatoid arthritis or systemic lupus erythematosus [5]. The clinical variables most influencing HRQoL were pruritus or pain related to LoS lesions and extracutaneous manifestations [4][5][6][7]11]. Nonetheless, there is still known very little about the physical, mental and emotional functioning of patients with LoS. Demographic features and socioeconomic status, that constitute a group of factors strongly affecting the HRQoL [12,13], were not investigated in most researches on LoS.
The objective of this study was to evaluate the HRQoL and relate it to socio-demographic and clinical aspects of adult onset LoS.

Patients
This cross-sectional study was conducted on patients with LoS representing a range of disease severity (mild to severe), gender (female and male) and age (over 18 years of age). Study participants were recruited from the Dermatology Department at the Medical University of Silesia in Katowice. The patients were diagnosed with LoS according to the classification of Kreuter et al. [14] with a biopsy specimen when necessary. Data were collected by a physician using a structured questionnaire for demographic (age, gender, place of residence, educational level, employment status, marital status) and clinical information (disease subtype, age at the disease onset and at the diagnosis, recurrences, extracutaneous manifestations, subjective symptoms e.g. itch, pain, burning or pricking sensation, coexisting disorders, Creactive protein level and erythrocyte sedimentation rate). Educational level was measured using 4 possible responses classified as primary, lower secondary, upper secondary and tertiary level, representing 8, 11, 12-13 and minimum of 15 years of formal education, respectively. A physical examination was performed in all patients and all of them were evaluated by consultant ophthalmologist and neurologist and those with arthralgia also by a rheumatologist. Activity of the disease and tissue damage was assessed with LoS cutaneous assessment tool (LoSCAT). This tool is a validated physicianderived measure that consists of two components: the modified LoS skin severity Index (mLoSSI) and the LoS skin damage index (LoSDI) [15][16][17]. With the use of the mLoSSI disease activity is scored from 0 to 162 based on the evaluation of the intensity of erythema (ER) and skin thickness (ST), appearance of new lesions or enlargement of an existing lesions within 1 month (N/E) in eighteen anatomic sites. Similarly, LoSDI quantifies the tissue damage based on the evaluation of dermal atrophy (DAT), subcutaneous atrophy (SAT) and dyspigmentation (DP) severity. A new LoSCAT composite score (based on 2 above-cited indexes) was used to classify the disease severity. To calculate this firstly introduced global index of skin involvement -LoSCAT summary score, scores of the mLoSSI and LoSDI were added together.
The Short Form Health Survey (SF-36v2) is a generic tool widely used to assess the level of HRQoL [22]. From this questionnaire, a physical component score (PCS) and a mental component score (MCS) can be calculated. The questionnaire can also be used to measure eight domains to give more specific information about a patient's HRQOL: physical functioning (PF), role limitations due to physical health problems (RP), bodily pain (BP), general health perceptions (GH), vitality (VT), social functioning (SF), role limitations due to emotional problems (RE), and mental health (MH). The results ranged from 0 to 100, with higher values reflecting better HRQoL. The score of 50 corresponds with the average HRQoL, the scores below 50 reflect lower HRQoL, and the scores above 50-higher up to excellent HRQoL [23].

Statistical analysis
Descriptive statistics were expressed mainly as medians and interquartile ranges (and also means and standard deviations for clinical data) for continuous variables, and as numbers and percentages for categorical variables. The Shapiro-Wilks test revealed that the distribution of variables was often non-normal, so Spearman's correlation, Kruskall-Wallis's and U Mann-Whitney's tests were used when necessary to explore the relation of each individual variable on the HRQoL. The calculated Spearman's rank correlation coefficient (r) was evaluated using the Student's t-test. All variables that showed a relationship (defined as P < 0.05) at a univariate level were then included in multivariable linear regression models. To remove "worst" predictors early and to leave only "important" predictors, a backward elimination method was used. Categorical variables such as subjective symptoms and musculoskeletal manifestations were defined as dummy variables, taking the value of 1 if found, and 0 if not. For the place of residence, the rural area of residence was considered the reference category.

Relationships of demographic and clinical factors with
HRQoL are presented in Tables 3, 4    Among clinical factors associated with worse SF-36 PCS, LoSCAT summary score (r = − 0.33, p = 0.037), subjective symptoms (U = 97, p = 0.03) and presence of extracutaneous manifestations (U = 310, p = 0.002) were found. Except for the age at the LoS onset and LoSCAT summary score, the other included variables remained significant (p < 0.05) as independent factors affecting HRQoL in the multiple linear regression analysis: the older age at survey, symptoms and musculoskeletal manifestations for the PCS, and the rural residence for the MCS.

Discussion
In our study analyzing sociodemographic factors of LoS patients and their relationship with HRQoL, older age and female gender were the ones that most influenced this variable, followed by rural place of residence. Being older at survey and at disease onset negatively influenced the PCS of SF-36, which seems to be typical trait of physical functioning impairment with age [13]. According to Klimas et al. [5], patients may have difficulty discerning the impact of LoS in the context of multiple health concerns and coexistent diseases may have a significant effect across all the HRQoL measures (SF-36, Skindex-29 and DLQI). In contrast however, HRQoL of our patients was not impaired by the presence of comorbidities. It should be highligted that detailed information about LoS and its natural course was prepared and this information was made available to patients in the present study. Consistent with general population studies and earlier researches on LoS, female gender of our patients was associated with worse HRQoL [4,9,10]. At this point it should be added that LoS may result in a permanent skin and subcutaneous tissue damage causing aesthetic defect, affecting self-esteem and mental state particularly in women. Indeed, women participating in the presented study had higher scores of Skindex-29 emotion subscale than men as previously reported by Szramka-Pawlak et al. [9]. The rural place of residence as a predictor for the lower SF36 MCS among patients with LoS is firstly reported. On contrary, Zagozdzon et al. [12] found better mental health of Polish women aged 45-60 living in rural areas compared to those from urban settings. According to the authors [12], area of residence is strongly associated with environmental and psychosocial factors determining HRQoL in the spheres of physical and mental health. Poor subjective health, chronic diseases and low socioeconomic status may negatively affect the HRQoL of rural populations [12,13]. Other sociodemographic factors evaluated firstly were education, marital and employment status, and none of them obtained significant results, neither in SF-36 nor in Skindex-29. Likewise, the effect of selected socioeconomic variables (race, regional income status and health insurance status) on HRQoL was investigated by Klimas et al. [5] and no significance for the functioning of LoS patients was shown. Out of the clinical features of LoS assessed in our patients, only the LoSCAT summary score, subjective symptoms (itching, burning sensation, skin tightness or pain) and musculoskeletal involvement influenced the SF-36 PCS. Activity of the disease per se did not affect HRQoL of study subjects. Similarly to Klimas et al. results [5], the mLoSSI was more closely linked with Skindex-29 scores and SF-36 PCS than the LoSDI but no significant correlations were found, as previously described [17,24]. Therefore, scores of the mLoSSI and LoSDI were added together and the relationship of the summary score with HRQoL was assessed. The sum of both indexes-the mLoSSI and LoSDI-seems to reflect general severity of skin lesions, as in the case of changes in the activity of LoS the signs of the damage usually occur. However, despite the correlation between the SF-36 PCS and the LoSCAT summary score, no significant impact of this score on Skindex-29 was revealed. The small sample size probably hindered the identifying of significant relationships from the data. Finding the itch, pain or other subjective symptoms as well as functional impairment among the factors influencing the HRQoL in adult LoS patients confirmed similar results from earlier reports [4][5][6][7]11]. The SF-36 PCS was significantly affected by LoS accompanying complaints and extracutaneous features, while they did not significantly correlated with Skindex-29 score of the study subjects. Nonetheless, one third of them (i.e. 20% mildly, 10% moderately to severely) had impaired functioning domain.
There was no correlation between the LoS subtype and HRQoL in the present study. What is more, percentage of body surface area involved influenced neither skin disease-specific nor universal questionare scores. Some researchers, including the group from Poland signalized that Skindex-29 or DLQI score may not be linked with LoS subtype [4,9]. However, it should be underlined that in the lager studies, patients with generalized LoS had significantly poorer HRQOL than those with the plaque or other subtype of the disease [5,10].
LoS was also reported to have negative impact on HRQoL with the greater number of affected body segments or with greater lesion number [11]. Furthermore, neither disease duration nor number of recurrences correlated with HRQoL of our patients. Similar results for HRQoL and disease duration were described by Kroft et al. and Das et al. [4,6]. Most importantly, further research should include larger groups of patients representing different characteristics.
Researches on HRQoL seem to be very important to the proper management of LoS. Some authors signalized the need for creating a specific LoS HRQoL instrument in order to meet the complexity of patient concerns within this disease [15,25]. Our results demonstrated that the SF-36 is well-suited for application across health states in LoS. Skindex-29 -a skin disease specific instrument -did not capture all these aspects, whilst they are covered by the SF-36. The major limitation of this study, however, eluding to cross-sectional study design, small sample size, generalizability of the study population characteristics and predictor selection methods, was the lack of a control group. Nevertheless, this is the first study that showed relationship between the SF36 MCS of LoS patients and the place of residence. Therefore, we would like to highlight that the HRQoL in rural populations deserves further investigation because it may be an important factor in a development of a support strategies for patients. Unlike to earlier studies, the sum of both indexes-the mLoSSI and LoSDI, was correlated with the HRQoL scores. The LoSCAT summary score was linked to SF-36 PCS of our patients, apart from a subjective symptoms and musculoskeletal manifestations. This implies, that the global index may be an important instrument for assessment of skin involvement in LoS, if we are to come up with holistic approach to the patient.

Conclusions
Older age, global skin involvement, presence of a subjective symptoms and extracutaneous manifestations may negatively affect the perceived physical health of adult patients with LoS. Female gender and rural residence may have negative impact on emotional or mental health, but further research is needed to clarify factors related to impaired HRQoL in these spheres.