Atopic eczema significantly associated with excessive scarring in new UK analysis

The influence of comorbidities on disfigurement keloid and hypertrophic scars (excessive scarring) was the subject of a comprehensive new analysis of data from the UK Biobank, with the results highlighting the need for further research into the pathophysiology of excessive scarring. In particular, atopic eczema was found to have significant associations with excessive scarring.

Although they arise from similar causes –in both cases, lesions and skin irritations – keloid scars persist while hypertrophic scars may dissipate over time as a result of skin wounds. The investigators of this population-based cross-sectional multicenter cohort study published in JAMA Dermatology used data on several established comorbidities of excessive skin scarring—hypertension, uterine leiomyoma, vitamin D deficiency, and atopic eczema—to validate previous findings and potentially uncover new comorbidities.

“Understanding disease comorbidities has both biological and clinical benefits, such as uncovering novel mechanisms and providing opportunities for targeted and early clinical intervention,” the researchers wrote. “Previous studies of keloid and hypertrophic scar comorbidities were limited to candidate diseases, based on assumed biological or demographic similarities.”

Their population of 230,078 patients included a study group of 972 with excessive scarring identified by diagnostic codes (n=740, keloid scar; n=110, hypertrophic scar; n=177, either other type of scar) and a control group with 229,106 healthy people. The study group had more female patients (65% versus 55%) but fewer patients of white ethnicity (86% versus 95%). The overall mean (SD) age was 64 (8) years.

The self-reported ethnicities of the entire patient cohort were Asian or Asian British (2.1%), Black or Black British (1.1%), Chinese (0.3%), mixed (0.5% ), white (94.9%) and other (0.7%). Among the 3 most represented ethnicities, black patients had the highest prevalence of excessive scarring compared to Asian patients and white patients: 2.4% vs 1.1% vs 0.4%.

Among all participants, the most common comorbidity was hypertension in 37.2% of the scar group and 34.3% of the control group. This was followed by uterine leiomyoma in 14.5% and 11.2%, respectively; atopic eczema in 10.2% and 5.8%; and vitamin D deficiency in 5.1% and 2.4%.

“All previously studied comorbidities (hypertension, uterine leiomyoma, vitamin D deficiency, atopic eczema) were more common in those with excessive scarring,” the authors wrote. They added that no differences were observed in these primary comorbidities within the group with excessive scarring between those who did or did not receive treatment.

Two models were used to investigate independent associations with excessive scarring:

• The “minimal model” adjusted for age, gender (uterine leiomyoma was restricted to female participants), and ethnicity
• The age-corrected “full model”; gender (except uterine leiomyoma); ethnic group; the influences of body mass index (BMI), Townsend deprivation index (TDI) and smoking status on hypertension, uterine leiomyoma and vitamin D deficiency; and the influences of BMI, TDI, allergic rhinitis and asthma on atopic eczema

All 4 comorbidities were universally associated with excessive scarring. With the minimal model, the risks were 20% higher (odds ratio [OR], 1.20; 95% CI, 0.96-1.51) uterine leiomyoma, 24% (OR, 1.24; 95% CI, 1.08-1.43) hypertension, 42% (OR, 1 .42; 95% CI, 1.05-1.93) of vitamin D deficiency, and 78% (OR, 1.78; 95% CI, 1.44-2.19). With the full model, the corresponding higher risks were 19% (OR, 1.19; 95% CI, 0.95-1.49), 11% (OR, 1.11; 95% CI, 0 .96-1.30), 47% (OR, 1.47; 95% CI, 1.08-1.99) and 68% (OR, 1.68; 95% CI, 1.36-2.07 ).

However, with the minimal model, statistically significant associations were observed only between excessive scarring and hypertension (P = .002) or atopic eczema (P < .001). And with the full model, only the presence of atopic eczema lent itself to a significant association (P < .001).

The authors’ sub-analysis of self-reported data on ethnicity yielded the following results:

• Among black participants, there were nominally significant associations between excessive scarring and hypertension (OR, 2.05; 95% CI, 1.13-3.72; P = 0.02) or uterine leiomyoma (OR, 1.93; 95% CI, 1.00-3.71; P = .05)
• Among Asian participants, there was a significant association between excessive scarring and vitamin D deficiency (OR, 2.24; 95% CI, 1.26-3.97; P = .006)
• For atopic eczema: among white participants, there was a highly significant association with excessive scarring (OR, 1.68; 95% CI, 1.34-2.12; P <0.001); among Asian participants, a nominally significant association (OR, 2.17; 95% CI, 1.01-4.67; P = 0.048); and among black participants, an association that tended towards statistical significance (OR, 1.89; 95% CI, 0.83-4.28; P = .13)

A final discovery analysis of the results in the excessive scarring group, a phenome-wide association study, examined 1518 phecodes in 17 disease states. In this subanalysis, investigators found the strongest associations between excessive scarring and sebaceous cyst (P = 9.45 × 10⁻³⁰), non-epithelial skin cancer (P = 2.03 × 10⁻²⁵), hair/hair follicle diseases (P = 1.50 × 10⁻²²), and skin/subcutaneous tissue infections, seborrheic keratosis, actinic keratosis, acne, and atopic/contact dermatitis (all, P < 1.0 × 10⁻¹¹).

“Using the clinical codes currently available within UK Biobank, we could only establish heterogeneous cases of excessive scarring based on potentially subjective clinical assessment, with no clinicopathologic correlation,” the authors concluded. “Nevertheless, our results add to what is already known in the literature, as evidenced by the detection of both previously reported and novel associations.”

Generalizability of these findings to other patient populations should be investigated in future studies, they stressed.

Reference

Ung CY, Warwick A, Onoufriadis A, et al. Comorbidities of keloid and hypertrophic scars in UK Biobank participants. JAMA Dermatol. Published online January 4, 2023. doi:10.1001/jamadermatol.2022.5607