Skin symptoms and their impact on quality-of-life in adults with osteogenesis imperfecta 

Winnie Liu  1 , Jonathan Xu  1 , Apoorva Salvi  1 , Suephy Chen  2 , Lars Folkestad  3   4   5 , Jorg Oliver Semler  6 , Eric Orwoll  1

Affiliations Expand

• PMID: 42694270

• PMCID: PMC13538963

• DOI: 10.1210/jendso/bvag182

Abstract

Keywords: brittle bone disease; extracellular matrix; osteogenesis imperfecta; skin; type 1 collagen.

Similar articles

• Systematic review of health related-quality of life in adults with osteogenesis imperfecta.

  Mc Donald D, Mc Donnell T, Martin-Grace J, Mc Manus G, Crowley RK.Orphanet J Rare Dis. 2023 Feb 22;18(1):36. doi: 10.1186/s13023-023-02643-3.PMID: 36814291Free PMC article.

• The IMPACT Survey: The Humanistic Impact of Caring for an Individual with Osteogenesis Imperfecta.

  Westerheim I, Rauch F, Hart T, Wekre LL, van Welzenis T, Raggio C, Mulhall H, Battersby A, Prince S, Semler O.Adv Ther. 2025 Nov;42(11):5811-5829. doi: 10.1007/s12325-025-03372-8. Epub 2025 Sep 26.PMID: 41004075Free PMC article.

• COL1A1- and COL1A2-Related Osteogenesis Imperfecta.

  Rodriguez Celin M, Steiner RD, Basel D.2005 Jan 28 [updated 2025 May 29]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews^® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026.PMID: 20301472Free Books & Documents.Review.

• Respiratory impairment impacts QOL in osteogenesis imperfecta independent of skeletal abnormalities.

  Yonko EA, Emanuel JS, Carter EM, Sandhaus RA, Raggio CL.Arch Osteoporos. 2020 Oct 2;15(1):153. doi: 10.1007/s11657-020-00818-0.PMID: 33009598

• Quality of life of pediatric and adult individuals with osteogenesis imperfecta: a meta-analysis.

  Wehrli S, Rohrbach M, Landolt MA.Orphanet J Rare Dis. 2023 May 24;18(1):123. doi: 10.1186/s13023-023-02728-z.PMID: 37226194Free PMC article.Review.

 Osteogenesis imperfecta (OI) is a heritable connective tissue disease primarily caused by mutations in a variety of genes that result in defects in the synthesis of type I collagen. Despite the disorder's genetic heterogeneity, 85% of OI cases result from dominantly inherited mutations in the COL1A1 and COL1A2 genes, which encode the α-polypeptide chains of type I collagen [1]. The remaining 15% represent diverse genetic etiologies stemming from noncollagen genes. These less common OI types (types V–XXII) implicate genes involved in bone mineralization, osteoblast differentiation, and the post-translational modification, folding, and cross-linking of collagen molecules [2].

The phenotypic presentations of OI are broad, with clinical features ranging from mild skeletal fragility to perinatal lethality. The clinical classification of OI follows the Sillence system, which outlines five primary classes (types I–V) that encompass most of the OI population. Type I (nondeforming with blue sclerae) is the mildest form with increased fracture risk and minimal deformity; type II (perinatally lethal) is the most severe, with in utero fractures and death in the neonatal period; type III (progressively deforming) is the most severe form of OI in individuals surviving the neonatal period. Type IV (common variable with normal sclerae) presents with intermediate severity; and type V features unique calcification of interosseous membranes [3].

Although OI is characterized by bone fragility, the qualitative and quantitative collagen abnormalities in OI are systemic and manifest in other collagen-containing tissues such as the respiratory, cardiovascular, and integumentary systems [4]. Collagen is the most abundant protein in the body and plays a key role in the structure and function of the extracellular matrix (ECM). In skin, type 1 collagen accounts for 85% of the dermal ECM and provides strength, structural support, and elasticity [5].

Small studies have demonstrated structural and functional skin abnormalities in OI, including reduced dermal collagen content, altered collagen-to-elastic fiber ratios, disorganized collagen fibrils with variable diameter, and reduced elasticity with increased distensibility on mechanical testing. Survey-based studies and case reports document increased skin symptoms in OI patients, including fragile skin, delayed wound healing, easy bruising, and overlap syndromes with Ehlers-Danlos and elastosis perforans serpiginosa [6‐10]. While the existing literature suggests abnormalities in skin structure, function in individuals with OI, the clinical scope and prevalence of skin symptoms and their impact on quality-of-life (QOL) have not been adequately explored.

Purpose

This study aimed to assess the prevalence of common skin symptoms and their impact on QOL in individuals with OI using questionnaire-based symptomatic, emotional, and functional indices. We hypothesized that skin symptoms are more prevalent in individuals with OI compared to comparators and that these symptoms significantly impair QOL.

Methods

Study participants and survey design

This cross-sectional study examined dermatologic symptoms in adults with OI compared with controls and differences between OI types. A suite of surveys was developed and distributed to volunteers from two cohorts: (1) individuals with OI who were participants in patient registries created by the Osteogenesis Imperfecta Foundation and Osteogenesis Imperfecta Federation Europe; and (2) participants from the Research Volunteer Registry and Biorepository—a registry of healthy volunteers at OHSU who are interested in and consented to be contact about research. These participants were recruited through health fairs and community events; direct advertisements (OHSU Research Opportunities webpage, flyers, brochures) and referrals from other OHSU IRB-approved studies. Participants were between 18 and 89 years, generally healthy, willing to be contacted periodically to update their information, and able to provide consent. We excluded participants with active infectious disease (eg, HIV, hepatitis), serious medical condition or medications that would interfere with their ability to participate (such as active cardiac or pulmonary disease, diabetes mellitus, clinically significant cancer diagnosis, active infectious disease, and active inflammatory diseases).

The survey suite consisted of four questionnaires that were presented in the following sequence. First, we administered a questionnaire that included general demographic items, such as age, sex, and race/ethnicity. Respondents with OI were asked to report their perceived OI severity as mild, moderate, severe or known/prefer not to answer. Second, we administered a questionnaire of skin symptoms (The General Skin Symptoms Questionnaire [GSSQ]). Finally, if respondents answered “yes” to the symptom question “Does your skin bother you?”, respondents were prompted to complete the Skindex-29 questionnaire [11]. If respondents answered “yes” to the symptom question “Does your skin itch?”, they were prompted to complete ItchyQoL [12].

The GSSQ is a 20-item, yes/no assessment of skin symptoms (Table S1) [13]. The GSSQ was developed collaboratively by the authors, including physicians who treat patients with OI, with input from persons with OI and their advocates. Skindex-29 and ItchyQoL are clinically validated questionnaires used to quantitatively measure the burden of skin disease on QOL [11, 12]. Skindex-29 is a 30-question survey that characterizes the effects of skin disease by dividing its impact on QOL into symptomatic (eg, “My skin hurts”), emotional (eg, “My skin condition makes me feel depressed”), and functioning (eg, “My skin condition affects my social life”) categories. There is also a Skindex-29 composite score that includes the three QOL domains. It provides scaled scores for each domain, ranging from 0 to 100, with 0 representing no effect and 100 representing constant affliction [11]. ItchyQoL is a 22-item questionnaire that includes the same three QOL categories but instead measures both frequency and severity on a pruritus-specific basis (eg, “My skin hurts because of my itchy skin condition”). Similar to Skindex-29, scoring of ItchyQoL consists of three domain scale scores (symptomatic, emotional, and functional) from of 1-5, with 1 representing a frequency of never and a severity of not bothered, and 5 representing a frequency of “all the time” and a severity of “severely bothered” [12].

Survey data were securely collected using REDCap (Research Electronic Capture) electronic data capture tools [14, 15]. The study was approved by the Institutional Review Board at OHSU and all participants provided informed consent. The survey was open from September 2024 to December 2024

Statistical analyses

The primary outcome measures were responses to the skin symptoms survey, expressed as the number of participants in the comparator and OI groups stratified by sex and OI severity who answered “Yes” to each question, and the domain scale scores from Skindex-29 and ItchyQoL. To assess significant differences in responses to the skin symptom survey, independent samples t-tests and chi-squared tests of independence were utilized to compare comparator and total OI cohorts, and across OI severity subgroups, respectively. To examine Skindex-29 and ItchyQoL domain scale scores stratified by OI severity, one-way analysis of variance was used. Primary analyses focused on prespecified, hypothesis-driven comparisons using validated QOL instruments. Since questionnaire subscales represent conceptually related and correlated domains rather than independent endpoints, strict correction for multiple testing was not performed, and secondary and subgroup analyses were interpreted cautiously.

All analyses were performed at OHSU using Stata (StataCorp LLC, USA) version 18 software, with P-values <.05 considered statistically significant.

Results

Survey invitations were emailed to 985 individuals from the Research Volunteers Repository contact list and 274 (27.8%) of them responded and completed the surveys. Five hundred and eighty nine individuals with OI responded and completed the surveys. We cannot calculate a response rate since the survey link was distributed through email lists, websites, and social media posts. Of the 589 people with OI who initially responded, 42 were excluded due to reporting an age of less than 18 years. Ultimately, 821 (66.6% OI respondents) surveys were analyzed. The mean age was similar between respondents with OI and the comparator group (49.1 15.8 vs 50.6 16.9 years,

 respectively, P = .2). Most respondents were female (77.1% in the OI group and 74.5% in the comparator group, P = .52) and both groups were predominantly White (>83%). There was a statistically significant difference in the racial distribution, with slightly more Hispanic/Latino individuals in the OI group. Among the 547 respondents with OI, 302 (55.2%) described their disease severity as mild, 179 (32.7%) as moderate, and 46 (8.4%) as severe (Table 1).

General skin symptoms questionnaire

Sex-stratified analyses

Women with OI reported more skin symptoms than women in the comparator group in 15 out of the 20 items in the GSSQ. The same pattern was observed for males with OI, who reported more symptoms on 11 of the 20 items (Fig. 1 and Table S1) [13].

Female OI vs male OI

Within the OI cohorts, females reported significantly more skin symptoms than males, with the largest differences observed for fragile nails (31.9%), stretch marks (31.2%), and easy bruising (28.4%). Overall, this pattern of sex differences in skin symptoms was similar to that of the comparator population (Fig. 1).

Mild, moderate, and severe OI

Individuals with mild, moderate, and severe OI answered similarly to 10 out of 20 skin questions. A higher proportion of those with severe OI than those with mild OI reported bothersome skin, itchy skin, and frequent rashes. On the other hand, a group of symptoms was significantly more common among those with milder OI, including bruising, bleeding, slow healing, scarring, stretch marks, and brittle nails

Skindex-29

Over half of participants (53%) with OI reported bothersome skin and completed the Skindex-29, compared with 31% of comparators. Among those with bothersome skin, individuals with OI had significantly higher Skindex-29 symptomatic scores than comparators (40.1 ± 18.5 vs 31.5 ± 19.0; P < .001); nevertheless, both mean scores fall within the severe category and there were no differences in functional or emotional scores. The composite scores were also higher in OI respondents than in comparators (30.0 ± 17.6 vs 25.0 ± 15.1, respectively; P = .02).

Men with OI had significantly higher symptomatic scores (34.0 ± 17.8 vs 20.8 ± 16.9, P = .011) and composite scores (24.6 ± 17.1 vs 15.7 ± 10.3, P = .049) than male comparators, whereas emotional and functional subscales did not differ. Women with OI had higher symptomatic scores than comparators (41.3 ± 18.5 vs 34.3 ± 18.7, P = .008). Women with and without OI responded similarly for emotional, functional, and composite scores (Table 2).

There were no significant differences among mild, moderate, and severe OI respondents in any of the Skindex-29 composite and subscale scores.

ItchyQoL

More than half (58%) of participants with OI reported itchy skin and completed the ItchyQoL instrument, compared to a much smaller proportion of the reference participants (28.6%) (Table 2). The OI group had a statistically significantly higher symptomatic score than the reference group (1.48 ± 0.74 vs 1.17 ± 0.81, P = .003). There were no significant score differences between groups in either the functional or emotional subscale scores or in the overall frequency categories for these domains (Table 2).