Atopic dermatitis is associated with an increased risk of keloids: A case-control study

To the Editor: Keloids are pathologic scars extending beyond the boundaries of the original cutaneous injuries that disproportionately affect patients with skin of color.1Hajdarbegovic E. Bloem A. Balak D. Thio B. Nijsten T. The association between atopic disorders and keloids: a case-control study.Indian J Dermatol. 2015; 60: 635Crossref PubMed Scopus (6) Google Scholar,2Young W.G. Worsham M.J. Joseph C.L. Divine G.W. Jones L.R. Incidence of keloid and risk factors following head and neck surgery.JAMA Facial Plast Surg. 2014; 16: 379-380Crossref PubMed Scopus (21) Google Scholar The molecular pathophysiology of keloid formation is inadequately understood.3Kwon H.E. Ahn H.J. Jeong S.J. Shin M.K. The increased prevalence of keloids in atopic dermatitis patients with allergic comorbidities: a nationwide retrospective cohort study.Sci Rep. 2021; 11: 23669Crossref PubMed Scopus (5) Google Scholar However, keloids and the inflammatory condition atopic dermatitis (AD) both involve responses mediated by T-helper (Th) 2 cytokines, with imbalances of pro-fibrotic and anti-fibrotic signals.3Kwon H.E. Ahn H.J. Jeong S.J. Shin M.K. The increased prevalence of keloids in atopic dermatitis patients with allergic comorbidities: a nationwide retrospective cohort study.Sci Rep. 2021; 11: 23669Crossref PubMed Scopus (5) Google Scholar,4Diaz A. Tan K. He H. et al.Keloid lesions show increased IL-4/IL-13 signaling and respond to Th2-targeting dupilumab therapy.J Eur Acad Dermatol Venereol. 2019; 34: e161-e164PubMed Google Scholar Keloids were associated with AD in Korean and Taiwanese populations, but an association was not found in a population of European ancestry.1Hajdarbegovic E. Bloem A. Balak D. Thio B. Nijsten T. The association between atopic disorders and keloids: a case-control study.Indian J Dermatol. 2015; 60: 635Crossref PubMed Scopus (6) Google Scholar,3Kwon H.E. Ahn H.J. Jeong S.J. Shin M.K. The increased prevalence of keloids in atopic dermatitis patients with allergic comorbidities: a nationwide retrospective cohort study.Sci Rep. 2021; 11: 23669Crossref PubMed Scopus (5) Google Scholar,5Lu Y.Y. Lu C.C. Yu W.W. et al.Keloid risk in patients with atopic dermatitis: a nationwide retrospective cohort study in Taiwan.BMJ Open. 2018; 8e022865Crossref Scopus (19) Google Scholar Given commonality in keloid and AD pathogenesis, our goal was to determine the risk of keloid diagnosis in patients with AD and stratify these outcomes with respect to Black vs White patients. Institutional review board approval was not needed as deidentified data were used. Using the TriNetX Research Network, a global health network comprising deidentified medical data, patients with AD (defined as ≥2 instances of diagnosis with AD [International Classification of Diseases 10th Revision code L20]) aged ≥18 years from 2010 to 2022 were identified. They were 1:1 propensity score matched by age, sex, and race to controls without dermatitis or eczema (L20-L30, Table I). Patients with diagnoses prior to use of International Classification of Diseases 10th Revision were classified using General Equivalence Mappings. One- and 5-year relative risks with 95% CIs for diagnosis of keloids (L91.0, hypertrophic scar) were measured, with significance at P < .05.Table IDemographic data for patients with atopic dermatitis and matched control patientsPatients with atopic dermatitisMatched control patientsP valueAge, mean ± SD (y)40.7 ± 21.940.8 ± 21.9.7664Sex, n (%) Female60,890 (62.046%)60,794 (61.948%).6553 Male37,223 (37.93%)37,319 (38.027%).6553Race, n (%) White50,800 (51.765%)50,704 (51.667%).6646 Black24,843 (25.315%)24,939 (25.412%).6185 Asian5099 (5.196%)5099 (5.196%)1.0000 Unknown16,844 (17.164%)16,844 (17.164%)1.0000 Open table in a new tab Consistent with previous literature, Black patients without AD had increased rates of keloids at 1 and 5 years versus White patients without AD (3.5 and 2.9-fold higher relative risk, respectively). Black patients with AD had 2.5-fold and 2.3-fold higher relative risk of keloids at 1 and 5 years, respectively, vs White patients with AD. Black and White patients with AD both had increased relative risk of keloids at 1 and 5 years vs matched patients without AD. The relative risk of keloids at 5 years was 1.931 in Black patients with AD vs Black controls and 3.184 in White patients with AD vs White controls (Table II).Table IIRelative risk of keloids at 1 and 5-y across varying groupsAD (#)AD + keloids (#)Risk (%)Control (#)Control + keloids (#)Risk (%)Relative risk (95% CI)P valueAD vs Control∗Denotes race-matched group. patients, 1-yr98,1373550.36298,1371140.1163.114 (2.522-3.845)<.0001AD vs Control∗Denotes race-matched group. patients, 5-y98,1377530.76798,1373270.3332.303 (2.023-2.621)<.0001Black AD (#)Black AD + keloids (#)Risk (%)Black control (#)Black control + keloids (#)Risk (%)Relative risk (95% CI)P valueBlack AD vs Black control patients, 1-y24,8431540.6224,843600.2422.567 (1.906-3.457)<.0001Black AD vs Black control patients, 5-y24,8433071.23624,8431590.641.931 (1.596-2.336)<.0001White AD (#)White AD + keloids (#)Risk (%)White control (#)White control + keloids (#)Risk (%)Relative risk (95% CI)P valueWhite AD vs White control patients, 1-y50,8001120.2450,800290.057%4.207 (2.807-6.306)<.0001White AD vs White control patients, 5-y50,8002770.545%50,800870.1713.184 (2.503-4.05)<.0001Black AD (#)Black AD + keloids (#)Risk (%)White AD (#)White AD + keloids (#)Risk (%)Relative risk (95% CI)P valueBlack AD vs White AD patients, 1-y24,7301510.62324,730610.2472.525 (1.878-3.394)<.0001Black AD vs White AD patients, 5-y24,7303051.23324,7301310.532.328 (1.899-2.855)<.0001Black control (#)Black control + keloids (#)Risk (%)White control (#)White control + keloids (#)Risk (%)Relative risk (95% CI)P valueBlack control vs White control patients, 1-y1,251,30626370.2111,251,3067640.0613.452 (3.185-3.741)<.0001Black control vs White control patients, 5-y1,251,30660180.4811,251,30620780.1662.896 (2.755-3.044)<.0001All groups of patients were age and sex-matched. “Control” signifies patients that did not have documentation of atopic dermatitis.AD, Atopic dermatitis.∗ Denotes race-matched group. Open table in a new tab All groups of patients were age and sex-matched. “Control” signifies patients that did not have documentation of atopic dermatitis. AD, Atopic dermatitis. These findings support an association between AD and keloids. Aberrant T-helper 2-mediated responses, with increased interleukin 4/IL-13 signaling as in AD, are a major inflammatory component in keloids.4Diaz A. Tan K. He H. et al.Keloid lesions show increased IL-4/IL-13 signaling and respond to Th2-targeting dupilumab therapy.J Eur Acad Dermatol Venereol. 2019; 34: e161-e164PubMed Google Scholar Furthermore, both conditions feature mast cell dysregulation, integral to both vascular homeostasis and innate and adaptive immune responses.1Hajdarbegovic E. Bloem A. Balak D. Thio B. Nijsten T. The association between atopic disorders and keloids: a case-control study.Indian J Dermatol. 2015; 60: 635Crossref PubMed Scopus (6) Google Scholar While consistent with recent studies showing up to a 7-fold increased incidence of keloids in Black vs White individuals, our results suggest that both Black and White patients with AD have an increased risk of keloids compared to controls without AD. Although the absolute risk is greater in Black patients with AD, the relative risk compared to controls without AD is greater in White patients with AD. These differences may be attributable to genetic polymorphisms and/or social and environmental factors that drive health disparities, predisposing to keloid formation.3Kwon H.E. Ahn H.J. Jeong S.J. Shin M.K. The increased prevalence of keloids in atopic dermatitis patients with allergic comorbidities: a nationwide retrospective cohort study.Sci Rep. 2021; 11: 23669Crossref PubMed Scopus (5) Google Scholar,5Lu Y.Y. Lu C.C. Yu W.W. et al.Keloid risk in patients with atopic dermatitis: a nationwide retrospective cohort study in Taiwan.BMJ Open. 2018; 8e022865Crossref Scopus (19) Google Scholar Limitations include possible misdiagnosis and the possibility that patients with AD are more likely to see a dermatologist and receive additional diagnoses, including keloid, vs populations without AD. The genetic and environmental determinants of keloid formation as well as the ability of systemic medications prescribed for T-helper 2-mediated conditions like AD to prevent or treat keloids warrants further investigation. None disclosed.