Reproductive and Hormonal Factors and Risk of Incident Rosacea among US White Women.
Wen-Hui Wu,Hao Geng,Eunyoung Cho,A Heather Eliassen,Aaron M Drucker,Tricia Y Li,Abrar A Qureshi,Wen-Qing Li
DOI: https://doi.org/10.1016/j.jaad.2021.06.865
IF: 15.487
2021-01-01
Journal of the American Academy of Dermatology
Abstract:To the Editor: Rosacea is a common skin inflammatory disease.1Two A.M. Wu W. Gallo R.L. Hata T.R. Rosacea: part I. Introduction, categorization, histology, pathogenesis, and risk factors.J Am Acad Dermatol. 2015; 72 (quiz 759-760): 749-758Abstract Full Text Full Text PDF PubMed Scopus (246) Google Scholar A connection between hormonal and reproductive factors and rosacea has long been proposed. Previous studies reported that pregnancy and oral contraceptive use may affect the appearance and severity of rosacea symptoms.2Bechstein S.K. Ochsendorf F. Hautarzt. 2017; 68: 111-119Crossref PubMed Scopus (8) Google Scholar Female hormone imbalance has been associated with an increased risk of rosacea.3Rainer B.M. Fischer A.H. Luz Felipe da Silva D. Kang S. Chien A.L. Rosacea is associated with chronic systemic diseases in a skin severity-dependent manner: results of a case-control study.J Am Acad Dermatol. 2015; 73: 604-608Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar Sex hormones have myriad effects on immune function4Moulton V.R. Sex hormones in acquired immunity and autoimmune disease.Front Immunol. 2018; 9: 2279Crossref PubMed Scopus (299) Google Scholar and dysregulated innate and adaptive immunity are important for rosacea initiation.1Two A.M. Wu W. Gallo R.L. Hata T.R. Rosacea: part I. Introduction, categorization, histology, pathogenesis, and risk factors.J Am Acad Dermatol. 2015; 72 (quiz 759-760): 749-758Abstract Full Text Full Text PDF PubMed Scopus (246) Google Scholar We therefore hypothesized that hormonal and reproductive factors would be associated with a risk of developing rosacea. We conducted the first comprehensive investigation assessing the associations between hormonal and reproductive factors and incident rosacea in a retrospective analysis of 89,873 White women based on the Nurses' Health Study-II. In 2005, participants were asked if they ever had clinician-diagnosed rosacea and diagnosis year in time intervals (before 1991, 1991-1994, 1995-1998, 1999-2002, or 2003-2005). Two small-scale validation studies supported the validity of rosacea diagnosis in Nurses' Health Study-II.5Li S. Chen M.L. Drucker A.M. et al.Association of caffeine intake and caffeinated coffee consumption with risk of incident rosacea in women.JAMA Dermatol. 2018; 154: 1394-1400Crossref PubMed Scopus (24) Google Scholar Menstrual, reproductive and exogenous hormonal factors were assessed at cohort inception (1989) and updated biennially (Supplemental Methods and Table I available via Mendeley at https://doi.org/10.17632/jrcjjnxfsx.1). Information on rosacea and hormonal and reproductive factors was collected independently for this cohort of health professionals. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for associations with risk of incident rosacea. We did not correct for multiple comparisons because analyses were based on predefined hypothesis and adjusted for multivariables. We identified 5248 incident rosacea cases during follow up (1991-2005). Onset of rosacea was less common among postmenopausal women than premenopausal women (HR = 0.73, 95% CI, 0.58-0.92) (Table I). No significant associations were found for other menstrual factors (Supplemental Table II). We found increased incident rosacea risk associated with use of menopausal hormone therapy (MHT, HR = 1.32; 95% CI, 1.08-1.61) and OC (HR = 1.15,; 95% CI, 1.05-1.25). In addition, rosacea risk was increased with longer duration of MHT use (HR = 1.78; 95% CI, 1.26-2.51 for ≥10 years), and with MHT containing oral conjugated estrogen (HR = 1.29; 95% CI, 1.05-1.59) (Table I).Table IHazard ratios of rosacea according to menopause status and exogenous hormone use in the Nurses' Health Study II (1991-2005)∗Hazard ratios were calculated by further adjusting for body mass index, alcohol intake, physical activity, smoking, age at menarche, age at menopause, type of menopause, parity, menopausal hormone use, oral contraceptive use, and personal histories of chronic diseases. Sensitivity analyses additionally adjusting for ultraviolet exposure during adulthood did not materially change the findings.No. of casesNo. of person-yearsAge-adjusted HR (95% CI)Multivariate HR (95% CI)Menopause status Premenopausal3294956,7971.001.00 Postmenopausal733160,2390.73 (0.66-0.80)0.73 (0.58-0.92)MHT use status†Analyses among postmenopausal women only. Never14833,4881.001.00 Ever593121,7541.36 (1.13-1.62)1.32 (1.08-1.61) Past15326,9351.19 (0.95-1.49)1.17 (0.93-1.48) Current44094,8191.43 (1.19-1.73)1.41 (1.14-1.75)<5 y26860,0451.37 (1.12-1.67)1.38 (1.12-1.71)5-9 y11624,3191.37 (1.07-1.74)1.45 (1.09-1.91)≥10 y6911,3351.66 (1.25-2.20)1.78 (1.26-2.51) P for trend.05.26Most recent MHT type†Analyses among postmenopausal women only. Never14834,4881.001.00 E27860,4211.46 (1.19-1.78)1.61 (1.22-2.12) E + P15940,6661.17 (0.93-1.46)1.16 (0.92-1.46) Others539,0951.45 (1.06-1.98)1.47 (1.06-2.03) E or E + P446102,6821.34 (1.11-1.61)1.29 (1.05-1.59)OC use status Never651172,3271.001.00 Ever45551,031,5841.15 (1.06-1.24)1.15 (1.05-1.25) Past4069921,9821.13 (1.04-1.23)1.14 (1.05-1.24) Current486109,6021.29 (1.15-1.45)1.21 (1.07-1.36)<5 y11323,9611.47 (1.20-1.79)1.40 (1.14-1.71)5-9 y15236,2571.34 (1.12-1.60)1.27 (1.06-1.52)≥10 y22149,1741.19 (1.02-1.39)1.10 (0.94-1.28) P for trend.07.02CI, Confidence interval; E + P, estrogen + progestin combination; E, unopposed estrogen; HR, hazard ratio; OC, oral contraceptive; MHT, menopausal hormone therapy.∗ Hazard ratios were calculated by further adjusting for body mass index, alcohol intake, physical activity, smoking, age at menarche, age at menopause, type of menopause, parity, menopausal hormone use, oral contraceptive use, and personal histories of chronic diseases. Sensitivity analyses additionally adjusting for ultraviolet exposure during adulthood did not materially change the findings.† Analyses among postmenopausal women only. Open table in a new tab CI, Confidence interval; E + P, estrogen + progestin combination; E, unopposed estrogen; HR, hazard ratio; OC, oral contraceptive; MHT, menopausal hormone therapy. For reproductive factors, we observed increased risk of incident rosacea among nulliparous women (HR = 1.22; 95% CI, 1.14-1.30) and an inverse association between number of births and risk of incident rosacea (P-trend = .008). Women with a late age at first or last birth also had increased risk of incident rosacea (Table II).Table IIHazard ratios of rosacea according to reproductive factors in the Nurses' Health Study II (1991-2005)∗Hazard ratios were calculated by further adjusting for the covariates listed in Table I footnote.No. of casesNo. of person-yearsAge-adjusted HR (95% CI)Multivariate HR (95% CI)Nulliparous No4034961,7611.001.00 Yes1203248,7741.21 (1.13-1.29)1.22 (1.14-1.30)Parity 1755185,5281.001.00 23279776,2340.99 (0.91-1.07)0.98 (0.90-1.06) 22040464,6181.04 (0.95-1.13)1.02 (0.94-1.11) 3929231,9150.93 (0.84-1.02)0.92 (0.84-1.01) ≥431079,7010.89 (0.78-1.01)0.90 (0.79-1.03)P for trend.006.008Age at first birth <22 y525126,2071.10 (0.99-1.23)1.14 (1.02-1.26) 22-24 y811218,2951.001.00 25-29 y1684407,2811.13 (1.04-1.23)1.12 (1.03-1.22) ≥30 y1018211,8031.25 (1.14-1.37)1.23 (1.12-1.36)P for trend.0007.003Age at last birth <25 y36588,6491.04 (0.92-1.17)1.05 (0.93-1.18) 25-29 y1151303,6851.001.00 30-34 y1542359,3881.11 (1.03-1.20)1.14 (1.05-1.23) ≥35 y826170,0861.13 (1.03-1.24)1.18 (1.07-1.29)P for trend.01.003Breastfeeding duration Never730175,5181.001.00 1-6 mos547142,3670.95 (0.85-1.06)0.94 (0.84-1.05) 7-11 mos718164,5811.07 (0.96-1.18)1.05 (0.95-1.16) 12-23 mos992219,5981.09 (0.99-1.20)1.10 (1.00-1.22) ≥24 mos868196,1761.04 (0.95-1.15)1.12 (1.01-1.25)P for trend.63.047CI, Confidence interval; HR, hazard ratio; No., number.∗ Hazard ratios were calculated by further adjusting for the covariates listed in Table I footnote. Open table in a new tab CI, Confidence interval; HR, hazard ratio; No., number. In our study, risk of incident rosacea was significantly decreased in postmenopausal women and increased in women taking MHT and OCs. Decreased estrogen caused by menopause may explain the decreased risk of rosacea among postmenopausal women. The complex interplays of hormonal factors during pregnancy associated with rosacea requires further study. Recall bias is a potential limitation and laboratory efforts are warranted to explain the mechanisms underlying observed findings. Our study suggests menstrual, reproductive and exogenous hormone factors may be associated with the risk of incident rosacea in women, which may shed light on rosacea etiology and inform dermatologists and general practitioners to answer patients' inquiries. Further studies are required to assess the associations for rosacea subtypes and to extend our findings to other population groups. None disclosed.