Advertisement

Association between Diet and Seborrheic Dermatitis: A Cross-Sectional Study

Open ArchivePublished:August 18, 2018DOI:https://doi.org/10.1016/j.jid.2018.07.027
      Current treatments for seborrheic dermatitis provide only temporary relief. Therefore, identifying modifiable lifestyle factors may help reduce disease burden. The objective of this study was to determine whether specific dietary patterns or total antioxidant capacity are associated with seborrheic dermatitis. Participants of the Rotterdam Study with a skin examination and a food frequency questionnaire were included. Total antioxidant capacity was assessed on the basis of ferric reducing antioxidant potential of each food item. Dietary patterns were identified with principal component analysis (PCA). Multivariable logistic regression analyses were used to assess the association between total antioxidant capacity, dietary pattern-derived PCA factors, and seborrheic dermatitis adjusted for confounders. In total, 4,379 participants were included, of whom 636 (14.5%) had seborrheic dermatitis. The PCA identified vegetable, Western, fat-rich and fruit dietary patterns. The fruit pattern was associated with a 25% lower risk (quartile 1 vs. quartile 4: adjusted odds ratio = 0.76, 95% confidence interval = 0.58–0.97, P = 0.03), and the Western pattern with a 47% increased risk (quartile 1 vs. quartile 4: adjusted odds ratio = 1.47; 95% confidence interval = 0.98–2.20, P = 0.03), but only for females. Other factors were not associated with seborrheic dermatitis. In conclusion, a high fruit intake was associated with less seborrheic dermatitis, whereas high adherence to a “Western” dietary pattern in females was associated with more seborrheic dermatitis.

      Abbreviations:

      FBSE (full body skin examination), FFQ (food frequency questionnaire), FRAP (ferric-reducing ability of plasma), PCA (principal component analysis), Q (quartile), RS (Rotterdam Study)

      Introduction

      Seborrheic dermatitis is a common chronic relapsing skin disease (
      • Gupta A.K.
      • Bluhm R.
      Seborrheic dermatitis.
      ). Although there is no known cause for seborrheic dermatitis, evidence does suggest that certain risk factors may predispose people to seborrheic dermatitis. As well as a genetic predisposition, these factors include male sex, light skin color, winter season, and high abundance of Malassezia yeasts on the skin (
      • Sanders M.G.H.
      • Pardo L.M.
      • Franco O.H.
      • Ginger R.S.
      • Nijsten T.
      Prevalence and determinants of seborrhoeic dermatitis in a middle-aged and elderly population: the Rotterdam Study.
      ,
      • Sanders M.G.H.
      • Pardo L.M.
      • Uitterlinden A.G.
      • Smith A.M.
      • Ginger R.S.
      • Nijsten T.
      The genetics of seborrheic dermatitis: a candidate gene approach and pilot genome-wide association study.
      ). Because treatment options for seborrheic dermatitis such as topical or oral antifungals and topical corticosteroids provide at most temporary relief, it is essential to identify modifiable lifestyle factors that may reduce the burden of this condition.
      Several studies suggested that nutrition may influence inflammatory skin diseases such as acne vulgaris and, to a lesser extent, psoriasis (
      • Kwon H.H.
      • Yoon J.Y.
      • Hong J.S.
      • Jung J.
      • Park M.S.
      • Suh D.H.
      Clinical and histological effect of a low glycaemic load diet in treatment of acne vulgaris in Korean patients: a randomized, controlled trial.
      ,
      • Millsop J.W.
      • Bhatia B.K.
      • Debbaneh M.
      • Koo J.
      • Liao W.
      Diet and psoriasis, part III: role of nutritional supplements.
      ). However, there are no studies investigating the effect of diet on seborrheic dermatitis. Numerous components in our diet may affect skin health in vitro. For instance, antioxidants, which are commonly found in fruits and vegetables, may be beneficial for inflammatory skin diseases (
      • Bickers D.R.
      • Athar M.
      Oxidative stress in the pathogenesis of skin disease.
      ,
      • Carlsen M.H.
      • Halvorsen B.L.
      • Holte K.
      • Bohn S.K.
      • Dragland S.
      • Sampson L.
      • et al.
      The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
      ). Other dietary components that might affect skin diseases include vitamin A (i.e., involved in keratinization, immunomodulation, and the regulation of sebaceous gland activity), omega-3 fatty acids (i.e., anti-inflammatory properties), and psoralen in citrus fruits (i.e., photocarcinogenic properties) (
      • Beckenbach L.
      • Baron J.M.
      • Merk H.F.
      • Loffler H.
      • Amann P.M.
      Retinoid treatment of skin diseases.
      ,
      • Dugrand A.
      • Olry A.
      • Duval T.
      • Hehn A.
      • Froelicher Y.
      • Bourgaud F.
      Coumarin and furanocoumarin quantitation in citrus peel via ultraperformance liquid chromatography coupled with mass spectrometry (UPLC-MS).
      ,
      • Millsop J.W.
      • Bhatia B.K.
      • Debbaneh M.
      • Koo J.
      • Liao W.
      Diet and psoriasis, part III: role of nutritional supplements.
      ). The effect of a single dietary component on a disease is often too small to detect in individuals, and the single components are part of an overall dietary pattern. Therefore, assessing dietary patterns and the overall antioxidant capacity may be more suitable for identifying potential nutritional risk factors for seborrheic dermatitis (
      • Hu F.B.
      Dietary pattern analysis: a new direction in nutritional epidemiology.
      ).
      In a recent observational study in the Rotterdam Study (RS), we showed that seborrheic dermatitis occurrence was associated with male sex, light skin color, dry skin, and winter season (
      • Sanders M.G.H.
      • Pardo L.M.
      • Franco O.H.
      • Ginger R.S.
      • Nijsten T.
      Prevalence and determinants of seborrhoeic dermatitis in a middle-aged and elderly population: the Rotterdam Study.
      ). In this cross-sectional study of a middle-aged and elderly population, we aimed to determine whether the total dietary antioxidant intake or a specific a posteriori defined dietary pattern is associated with seborrheic dermatitis.

      Results

      Study population

      In total, 5,498 participants underwent a full-body skin examination (FBSE), and 4,379 of them (RS-1: 801, RS-II: 1,441, RS-III: 2,137) had complete nutrition data. Of the 4,379 eligible participants, 636 had seborrheic dermatitis lesions found during the FBSE (point prevalence = 14.5%). The median age of all participants was 68.9 years (interquartile range = 62.6–77.4), and the proportion of woman was 57.6%. The distribution of the demographic characteristics and possible confounding factors between people with and without seborrheic dermatitis at time of FBSE are presented in Table 1.
      Table 1Characteristics of participants by having or not having seborrheic dermatitis
      The chi-square test was used for categorical variables and the t test for continuous variables.
      CharacteristicSDNo SDP-value
      Participants, n (%)636 (14.5)3,743 (85.5)
      Male sex, n (%)378 (59.4)1,477 (39.5)<0.01
      Age in years, median (IQR)69.9 (63.9–77.5)68.8 (62.4–77.4)0.14
      BMI in kg/m2, n (%)0.23
       <25179 (28.1)1,176 (31.4)
       25–30310 (48.7)1,714 (45.8)
       >30147 (23.1)845 (22.6)
       Missing0 (0.0)8 (0.2)
      Skin color, n (%)<0.01
       Very white–white567 (89.2)3,103 (82.9)
       White–olive61 (9.6)524 (14.0)
       Light brown–black8 (1.3)116 (3.3)
      Education, n (%)0.45
       Low54 (8.5)357 (9.5)
       Average378 (59.4)2,267 (60.6)
       High195 (30.7)1,068 (28.5)
       Missing9 (1.4)51 (1.4)
      Xerosis cutis, n (%)0.07
       No214 (33.6)1,421 (38.0)
       Extensor-side extremities325 (51.1)1,850 (49.4)
       Generalized93 (14.6)440 (11.8)
       Other4 (0.6)31 (0.8)
       Missing0 (0.0)1 (0.03)
      Physical activity in METhours/week, median (IQR)40.1 (15.0–75.8)43.0 (17.5–81.6)0.04
       Missing, n (%)24 (3.8)191 (5.1)
      Season, n (%)<0.01
       Winter187 (29.4)861 (23.0)
       Spring137 (21.5)800 (21.4)
       Summer77 (12.1)605 (16.2)
       Autumn235 (36.9)1,477 (39.5)
      Current smoker, n (%)546 (85.8)3,262 (87.1)0.37
      Alcohol in glasses/week, median (IQR)8.3 (1.4–21.0)6.8 (0.9–17.3)0.04
      Energy intake in kcal/day, median (IQR)2,118.1 (1,721.3–2,626.2)2,077.2 (1,670.9–2,542.3)0.07
      Weekly use of supplements, n (%)296 (46.5)1,686 (45.0)0.48
      Abbreviations: IQR, interquartile range; MET, metabolic equivalent; SD, seborrheic dermatitis.
      1 The chi-square test was used for categorical variables and the t test for continuous variables.

      Total antioxidant capacity

      The median ferric-reducing ability of plasma (FRAP) score at an intake of 2,000 kcal was 24.3 mmol/day (interquartile range = 17.3–30.5). The crude and multivariable logistic regression between the FRAP score of the diet and seborrheic dermatitis is shown in Table 2. We did not find evidence for an association between antioxidant intake and seborrheic dermatitis (FRAP score, quartile 1 vs quartile 4: adjusted odds ratio = 0.94; 95% confidence interval = 0.73–1.19, P = 0.88).
      Table 2Multivariable logistic regression between the total antioxidant capacity of the diet and the risk of having seborrheic dermatitis
      FRAPCrude OR
      Crude OR: odds ratio adjusted for age and sex.
      95% CI Lower95% CI UpperP-ValueAdjusted OR
      Adjusted OR: odds ratio adjusted for age, sex, total energy intake, skin color, smoking, alcohol, body mass index, season, physical activity education, and supplement use.
      95% CI Lower95% CI UpperP-Value
      Q1 (ref)
      Q20.810.631.030.090.800.631.030.08
      Q31.080.861.370.511.040.821.320.75
      Q40.960.751.220.740.930.731.190.57
      P for trend0.680.91
      Abbreviations: CI, confidence interval; FRAP, ferric-reducing ability of plasma; OR, odds ratio; Q, quartile.
      1 Crude OR: odds ratio adjusted for age and sex.
      2 Adjusted OR: odds ratio adjusted for age, sex, total energy intake, skin color, smoking, alcohol, body mass index, season, physical activity education, and supplement use.

      Dietary patterns

      The principal component analysis (PCA) yielded four independent components of interest, explaining 26.2% of the total variation of the diet patterns. The first component was characterized as a vegetables dietary pattern; the second as a Western pattern, characterized by meat, potato, and alcohol consumption; the third as a fruit pattern; and the fourth as a fat pattern, which most correlated with consuming olive oil and with other healthy and unhealthy fats (Table 3).
      Table 3Principal component analysis of 34 food groups
      Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization. Factor loadings with a low predictive value (–0.2 ≤ x ≤ 0.2) are presented in regular type. Factor loadings with a high predictive value (>0.35) are presented in boldface type.
      Food GroupVegetablesWesternFruitFat
      Greenleaf vegetables0.8460.0390.0400.047
      Vegetables, other0.8250.1100.0130.030
      Yellow leaf vegetables0.760–0.0860.1520.000
      Meat unprocessed0.0480.7270.0060.021
      Meat processed–0.0450.670–0.0420.070
      Potatoes0.1640.4780.0120.096
      Alcoholic drinks (not wine)–0.0620.366–0.0670.040
      Citrus fruits0.050–0.0130.9030.002
      Fruits (non-citrus)0.134–0.0210.883–0.011
      Olive oil0.111–0.0990.0750.758
      Unhealthy fats–0.0610.086–0.0790.692
      Healthy fats–0.0130.144–0.0090.596
      Soups and sauces0.1350.2730.0280.330
      Sweets–0.0650.0560.1230.112
      Savory snacks0.0180.152–0.1170.027
      Refined grains0.1430.113–0.1630.105
      Nuts and seeds0.110–0.0610.0630.013
      Fatty fish0.145–0.0050.0390.010
      Lean fish0.113–0.0080.0070.029
      Shellfish–0.0120.1420.029–0.010
      Whole grains0.0750.032–0.0840.222
      Yogurt0.110–0.0440.175–0.192
      Mineral water0.088–0.0060.060–0.006
      Poultry0.1970.1850.025–0.029
      Black tea0.0210.1230.079–0.017
      Coffee0.0870.148–0.007–0.045
      Soy0.127–0.1130.033–0.004
      Pulses0.0280.178–0.037–0.054
      Herb tea0.140–0.1610.1100.042
      Wine0.073–0.022–0.027–0.026
      Cheese0.0510.1600.0700.023
      Soft drinks–0.0300.130–0.031–0.081
      Milk0.063–0.0800.0880.010
      Eggs–0.0660.295–0.0180.076
      1 Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization. Factor loadings with a low predictive value (–0.2 ≤ x ≤ 0.2) are presented in regular type. Factor loadings with a high predictive value (>0.35) are presented in boldface type.
      Table 4 shows the effect of adherence to the diet patterns on the risk of having seborrheic dermatitis. In the crude model, there seems to be a negative effect of the Western pattern, a positive effect of fruit consumption (borderline significant), and no effect of vegetable or fat consumption. In the adjusted model, adherence to the Western pattern seems to be associated with a higher risk for seborrheic dermatitis, but this was significant only for the highest quartile (quartile 1 vs. quartile 4: adjusted odds ratio = 1.34, 95% confidence interval = 1.03–1.75, P = 0.07). Adherence to the fruit pattern was associated with a lower risk for seborrheic dermatitis (quartile 1 vs. quartile 4: adjusted odds ratio = 0.75, 95% confidence interval = 0.58–0.97, P = 0.03), and adherence to the other patterns did not influence seborrheic dermatitis risk.
      Table 4Multivariable logistic regression between adherence to the dietary patterns and the risk of having seborrheic dermatitis
      PatternCrude OR
      Crude OR: odds ratio adjusted for age and sex.
      95% CI Lower95% CI UpperP-Value
      Boldface indicates P-values < 0.05.
      Adjusted OR
      Adjusted OR: odds ratio adjusted for age, sex, skin color, smoking, body mass index, season, total energy intake, physical activity, education, and supplement use.
      95% CI Lower95% CI UpperP-Value
      Boldface indicates P-values < 0.05.
      Vegetables
       Q1 (ref)
       Q21.040.821.330.751.0280.811.310.82
       Q31.130.891.440.311.1340.891.450.31
       Q41.120.871.430.371.1220.871.450.38
      P for trend0.280.28
      Western
       Q1 (ref)
       Q21.291.001.660.051.271.111.440.07
       Q31.190.921.540.181.160.891.500.27
       Q41.351.051.730.021.341.031.750.03
      P for trend0.050.07
      Fruit
       Q1 (ref)
       Q20.870.691.090.220.830.740.930.12
       Q30.800.631.020.070.770.600.990.04
       Q40.800.631.030.080.750.580.970.03
      P for trend0.060.03
      Fat
       Q1 (ref)
       Q21.260.991.610.061.281.001.630.05
       Q30.960.751.240.760.980.771.240.85
       Q41.200.941.530.141.22.9451.580.14
      P for trend0.470.47
      Abbreviations: CI, confidence interval; OR, odds ratio; Q, quartile; ref, reference.
      1 Crude OR: odds ratio adjusted for age and sex.
      2 Boldface indicates P-values < 0.05.
      3 Adjusted OR: odds ratio adjusted for age, sex, skin color, smoking, body mass index, season, total energy intake, physical activity, education, and supplement use.
      In the additional analysis, we tested for interaction between the dietary outcomes and all other variables. A significant interaction was found between the Western dietary pattern and sex (P = 0.013). Therefore, we decided to stratify this dietary pattern. For males, there was no significant association between the Western pattern and seborrheic dermatitis. However, for females, a higher adherence to this pattern was associated with an increased risk of seborrheic dermatitis (Table 5).
      Table 5Multivariable logistic regression between the Western dietary pattern and seborrheic dermatitis, stratified for sex
      This stratified multivariable logistic regression was conducted because of the significant interaction between the Western dietary pattern and sex (P = 0.013).
      QuartileMaleFemale
      OR
      Odds ratio adjusted for age, sex, skin color, smoking, body mass index, season, total energy intake, physical activity, education and supplement use.
      95% CI Lower95% CI UpperP-ValueOR
      Odds ratio adjusted for age, sex, skin color, smoking, body mass index, season, total energy intake, physical activity, education and supplement use.
      95% CI Lower95% CI UpperP-Value
      Boldface indicates P-values < 0.05.
      Q1 (ref)
      Q21.190.991.440.351.260.881.810.21
      Q30.870.601.270.471.501.042.180.03
      Q41.180.831.670.371.470.982.200.06
      P for trend0.670.03
      Abbreviations: CI, confidence interval; OR, odds ratio; Q, quartile; ref, reference.
      1 This stratified multivariable logistic regression was conducted because of the significant interaction between the Western dietary pattern and sex (P = 0.013).
      2 Odds ratio adjusted for age, sex, skin color, smoking, body mass index, season, total energy intake, physical activity, education and supplement use.
      3 Boldface indicates P-values < 0.05.

      Discussion

      In the study, we found that participants with a dietary pattern characterized by high fruit intake had lower odds of having seborrheic dermatitis after adjustment for confounders. Furthermore, we found that a Western dietary pattern was associated with higher odds of having seborrheic dermatitis, but only for females. We did not find an association between the presence of seborrheic dermatitis and dietary patterns characterized by vegetables or fat, nor did we find an association between seborrheic dermatitis and total dietary antioxidant capacity.
      The consumption of fruits might reduce the likelihood of having seborrheic dermatitis via consumption of a wide variety of vitamins and other compounds (e.g., flavonoids, antioxidant) that have been shown to reduce inflammation in several diseases (
      • Cepeda A.M.
      • Del Giacco S.R.
      • Villalba S.
      • Tapias E.
      • Jaller R.
      • Segura A.M.
      • et al.
      A traditional diet is associated with a reduced risk of eczema and wheeze in Colombian children.
      ,
      • He F.J.
      • Nowson C.A.
      • MacGregor G.A.
      Fruit and vegetable consumption and stroke: meta-analysis of cohort studies.
      ). Also, fruits contain several nutrients that can serve as methyl donors, which can prevent the expression of inflammatory genes (
      • Perdigoto C.N.
      • Valdes V.J.
      • Bardot E.S.
      • Ezhkova E.
      Epigenetic regulation of epidermal differentiation.
      ). Another possible hypothesis by which fruits affect skin health involves psoralen. Psoralen is highly present in citrus fruits and increases the sensitivity of the skin to UVR (
      • Wu S.
      • Han J.
      • Feskanich D.
      • Cho E.
      • Stampfer M.J.
      • Willett W.C.
      • et al.
      Citrus consumption and risk of cutaneous malignant melanoma.
      ). This increased sensitivity to UVR could have a positive effect on seborrheic dermatitis because this dermatosis is less frequently present in summer months (
      • Pirkhammer D.
      • Seeber A.
      • Honigsmann H.
      • Tanew A.
      Narrow-band ultraviolet B (ATL-01) phototherapy is an effective and safe treatment option for patients with severe seborrhoeic dermatitis.
      ,
      • Sanders M.G.H.
      • Pardo L.M.
      • Franco O.H.
      • Ginger R.S.
      • Nijsten T.
      Prevalence and determinants of seborrhoeic dermatitis in a middle-aged and elderly population: the Rotterdam Study.
      ).
      Western diets and diets high in meat and processed food consumption have often been associated with markers of inflammation (
      • Giugliano D.
      • Ceriello A.
      • Esposito K.
      The effects of diet on inflammation: emphasis on the metabolic syndrome.
      ,
      • Ozawa M.
      • Shipley M.
      • Kivimaki M.
      • Singh-Manoux A.
      • Brunner E.J.
      Dietary pattern, inflammation and cognitive decline: the Whitehall II prospective cohort study.
      ). Omega-6 fatty acids are one of the markers that can significantly change because of a diet and have been suggested to induce chronic inflammation. However, intervention studies with omega-6 supplementation did not substantiate this (
      • Innes J.K.
      • Calder P.C.
      Omega-6 fatty acids and inflammation.
      ). The stratified analysis of our data showed that females with a high adherence to the Western pattern seem to have higher odds of having seborrheic dermatitis. However, higher adherence to the Western pattern was not associated with seborrheic dermatitis in males. Previous dietary intervention studies showed that there are differences in response to diet between males and females. For example, a healthy diet improved insulin homeostasis in males but not in females; sex differences in body fat distribution might be one explanation for this (
      • Bedard A.
      • Riverin M.
      • Dodin S.
      • Corneau L.
      • Lemieux S.
      Sex differences in the impact of the Mediterranean diet on cardiovascular risk profile.
      ). Also, it is known that immune response in females differs from that in males and that females are more susceptible to autoimmune and inflammatory diseases (
      • Klein S.L.
      • Flanagan K.L.
      Sex differences in immune responses.
      ). This study underlines the importance of possible interactions between sex and nutrition, in which adding sex as a confounder in the final model is not sufficient.
      Because seborrheic dermatitis is a chronic inflammatory disease, and because reactive oxygen species may promote chronic inflammation or aggravate inflammatory skin diseases (
      • Trouba K.J.
      • Hamadeh H.K.
      • Amin R.P.
      • Germolec D.R.
      Oxidative stress and its role in skin disease.
      ), we expected individuals with a high total antioxidant intake to have a lower prevalence of skin disease. Moreover, two previous studies, one of the total antioxidant level in serum and one of the antioxidant levels of scalp scrapes, suggested that oxidative stress might be higher in seborrheic dermatitis patients (
      • Emre S.
      • Metin A.
      • Demirseren D.D.
      • Akoglu G.
      • Oztekin A.
      • Neselioglu S.
      • et al.
      The association of oxidative stress and disease activity in seborrheic dermatitis.
      ,
      • Ozturk P.
      • Arican O.
      • Belge Kurutas E.
      • Karakas T.
      • Kabakci B.
      Oxidative stress in patients with scalp seborrheic dermatitis.
      ). In contrast to these studies and the hypothesis, participants with a higher overall dietary antioxidant capacity did not have a decreased odds of having seborrheic dermatitis. This observation suggests that the effect of oral antioxidants in the treatment of seborrheic dermatitis might be limited. However, it has been questioned if methods assessing the overall antioxidant capacity should be used to make claims concerning the antioxidant defense system (
      • Turck D.
      • Bresson J.L.
      • Burlingame B.
      • Dean T.
      • Fairweather-Tait S.
      • Heinonen M.
      • et al.
      Guidance for the scientific requirements for health claims related to antioxidants, oxidative damage and cardiovascular health.
      ). In addition, we also did not find a consistent association between dietary antioxidant capacity and inflammatory markers in The RS (
      • Stringa N.
      • Brahimaj A.
      • Zaciragic A.
      • Dehghan A.
      • Ikram M.A.
      • Hofman A.
      • et al.
      Relation of antioxidant capacity of diet and markers of oxidative status with C-reactive protein and adipocytokines: a prospective study.
      ). Therefore, other biomarkers of antioxidant capacity or oxidative stress may provide additional insights in the role of antioxidants in seborrheic dermatitis.
      In this study we investigated the role of diet in seborrheic dermatitis. The strengths of our study are the large sample size, the population-based setting, the physician-based diagnosis, and the availability of different epidemiological factors that allowed us to control for potential confounders. There are several limitations to the study. The cross-sectional design does not allow us to make causal inferences, and because this study covers a middle-aged and elderly population, the generalizability to younger patients might be limited. Unfortunately, the disease severity and distribution of seborrheic dermatitis was not specifically documented during the FBSE. Therefore, we cannot elaborate further on the relationship between location or severity of the disease and the associations with the diet components. Also, because we did not investigate individual dietary components, this study cannot be used to select possible supplements that might reduce seborrheic dermatitis risk. Furthermore, the use of a food frequency questionnaire (FFQ) to assess dietary intake is prone to measurement error. To account for systematic measurement errors and to limit the influence of outliers, we adjusted our analyses for total energy intake and categorized the dietary data into quartiles. Although self-reported dietary intake is subject to measurements error when it concerns absolute intake, it has been shown that the FFQ is able to adequately rank the intake of individuals according to their food group (
      • Goldbohm R.A.
      • van den Brandt P.A.
      • Brants H.A.
      • van ’t Veer P.
      • Al M.
      • Sturmans F.
      • et al.
      Validation of a dietary questionnaire used in a large-scale prospective cohort study on diet and cancer.
      ). To account for potential confounding by food supplementation, we adjusted the analyses for any dietary supplement use. For RS-I and RS-II, the FFQ was conducted in the same period as the FBSE. For RS-III, the FFQ data were registered 5 years earlier than the FBSE. However, we previously showed that dietary patterns are relatively stable in this population (in the same quartile of intake), in particular for components such as vegetables (73%), fruits (93%), dietary fiber (91%), saturated fat (92%) and alcohol (79%) (
      • Schoufour J.D.
      • de Jonge E.A.L.
      • Kiefte-de Jong J.C.
      • van Lenthe F.J.
      • Hofman A.
      • Nunn S.P.T.
      • Franco O.H.
      Socio-economic indicators and dietary quality in an elderly population.
      ). There might always be residual confounding of variables that were not documented in this cohort. Self-reported stress, for example, could be such a variable that might have influenced both dietary choices and disease risk (
      • Misery L.
      • Touboul S.
      • Vincot C.
      • Dutray S.
      • Rolland-Jacob G.
      • Consoli S.G.
      • et al.
      Stress et dermatite séborrhéique [Stress and seborrheic dermatitis].
      ). However, we previously showed that measurements of anxiety and depression (proxies of stress) were not associated with seborrheic dermatitis in this cohort (
      • Sanders M.G.H.
      • Pardo L.M.
      • Franco O.H.
      • Ginger R.S.
      • Nijsten T.
      Prevalence and determinants of seborrhoeic dermatitis in a middle-aged and elderly population: the Rotterdam Study.
      ). Also, including these variables in the multivariable logistic regression of this study did not influence the associations between the dietary patterns and seborrheic dermatitis (data not shown).
      In conclusion, a high intake of fruit was associated with lower odds of seborrheic dermatitis, and a high adherence to the Western dietary pattern seems to be associated with a higher risk of seborrheic dermatitis in females. These findings were not driven by the overall dietary antioxidant capacity. Although the results of this study cannot be used to provide exact recommendations, it seems advisable for seborrheic dermatitis patients to follow national diet guidelines regarding fruit, which recommends at least 200 g/day of fruit in The Netherlands (

      Brink L, Smeets AP-, Stafleu A, Wolvers D. The Netherlands Nutrition Centre. Richtlijnen Schijf van Vijf. http://www.voedingscentrum.nl/professionals/schijf-van-vijf/naslag-richtlijnen-schijf-van-vijf.aspx. 2016. Accessed 4 October 2018.

      ). Furthermore, although a high adherence to the Western dietary pattern was associated with an increased disease risk only for females, it might be beneficial for both sexes to reduce meat consumption, considering recent literature linking meat intake with an increased risk of mortality (
      • Etemadi A.
      • Sinha R.
      • Ward M.H.
      • Graubard B.I.
      • Inoue-Choi M.
      • Dawsey S.M.
      • et al.
      Mortality from different causes associated with meat, heme iron, nitrates, and nitrites in the NIH-AARP Diet and Health Study: population based cohort study.
      ). Dietary pattern studies seem appropriate, because initial dietary association study and might be meaningful in other skin diseases as well. Replication of our findings in an independent cohort, or conducting an interventional study, would be necessary to substantiate these claims. A prospective study measuring relapse rate in seborrheic dermatitis patients with a high or low fruit intake would be of high value.

      Methods

      Study design

      The RS is an ongoing prospective population-based cohort study of chronic diseases in a middle aged and elderly population in the Ommoord district of Rotterdam, The Netherlands (
      • Ikram M.A.
      • Brusselle G.G.O.
      • Murad S.D.
      • van Duijn C.M.
      • Franco O.H.
      • Goedegebure A.
      • et al.
      The Rotterdam Study: 2018 update on objectives, design and main results.
      ). The Rotterdam Study has been approved by the Medical Ethics Committee of the Erasmus MC (registration number MEC 02.1015) and by the Dutch Ministry of Health, Welfare and Sport (Population Screening Act WBO, license number 1071272-159521-PG). The RS has been entered into the Netherlands National Trial Register (NTR; www.trialregister.nl) and into the World Health Organization International Clinical Trials Registry Platform (www.who.int/ictrp/network/primary/en/) under shared catalog number NTR6831. All participants provided written informed consent to participate in the study and to have their information obtained from treating physicians. The study started in 1990 and now comprises three cohorts (RS-1, RS-II, and RS-III) with a total of 14,926 participants aged 45 years or older. Dermatological examinations were introduced in 2010, and since then, 5,498 participants have had a skin examination. The current study is a cross-sectional study containing all participants with a skin examination and available nutrition data.

      Case definition

      Seborrheic dermatitis was diagnosed by a dermatology-trained physician during a scheduled FBSE. The diagnoses were based on greasy scaling, erythema, and a characteristic distribution in areas rich in sebaceous glands. Participants without seborrheic dermatitis were considered as controls.

      Nutritional data

      Dietary intake was assessed by means of an FFQ, which included 389 questions regarding the consumption of food over the last month. The Dutch Food Composition Table of 2006 and 2011 was then used to transform the data into daily macronutrient intake and total energy intake (kcal/day) (

      Dutch Food Composition Table. Nevo tabel 2006/2011. Voorlichtingsbureau voor de Voeding. The Hague, The Netherlands, 2011. https://www.rivm.nl/en/Topics/D/Dutch_Food_Composition_Database/Publications. Accessed 4 October 2018.

      ). This FFQ was based on a validated FFQ for Dutch adults (
      • Feunekes G.I.
      • Van Staveren W.A.
      • De Vries J.H.
      • Burema J.
      • Hautvast J.G.
      Relative and biomarker-based validity of a food-frequency questionnaire estimating intake of fats and cholesterol.
      ,
      • Goldbohm R.A.
      • van den Brandt P.A.
      • Brants H.A.
      • van ’t Veer P.
      • Al M.
      • Sturmans F.
      • et al.
      Validation of a dietary questionnaire used in a large-scale prospective cohort study on diet and cancer.
      ). This FFQ was validated against 3-day food records, 4–5 months apart, and showed an energy- and sex-adjusted correlation for macronutrients between 0.47 (fat) to 0.79 (polysaccharides).

      Total antioxidant capacity

      The total antioxidant capacity was calculated as described earlier (
      • Pantavos A.
      • Ruiter R.
      • Feskens E.F.
      • de Keyser C.E.
      • Hofman A.
      • Stricker B.H.
      • et al.
      Total dietary antioxidant capacity, individual antioxidant intake and breast cancer risk: the Rotterdam Study.
      ). In short, an Antioxidant Food Table (
      • Carlsen M.H.
      • Halvorsen B.L.
      • Holte K.
      • Bohn S.K.
      • Dragland S.
      • Sampson L.
      • et al.
      The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
      ) was used to assess the antioxidant capacity of each dietary item. This Antioxidant Food Table contains the antioxidant capacity of food items determined assessed on the basis of an existing table that evaluated the FRAP for more than 3,000 food items (
      • Carlsen M.H.
      • Halvorsen B.L.
      • Holte K.
      • Bohn S.K.
      • Dragland S.
      • Sampson L.
      • et al.
      The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
      ). For every participant, the consumption frequency of each dietary item was multiplied by the FRAP value in this table. The total antioxidant capacity was then adjusted for the total energy intake using the residual method and categorized into quartiles (
      • Willett W.C.
      • Howe G.R.
      • Kushi L.H.
      Adjustment for total energy intake in epidemiologic studies.
      ).

      Dietary patterns

      To define dietary patterns, we used the a posteriori dietary pattern analysis described by
      • Hu F.B.
      • Rimm E.
      • Smith-Warner S.A.
      • Feskanich D.
      • Stampfer M.J.
      • Ascherio A.
      • et al.
      Reproducibility and validity of dietary patterns assessed with a food-frequency questionnaire.
      . The list of 389 food items from the FFQ was categorized into 34 main food groups. The groups were categorized on the basis of the Nevo table (

      Dutch Food Composition Table. Nevo tabel 2006/2011. Voorlichtingsbureau voor de Voeding. The Hague, The Netherlands, 2011. https://www.rivm.nl/en/Topics/D/Dutch_Food_Composition_Database/Publications. Accessed 4 October 2018.

      ), while accounting for effects of specific subgroups (e.g., vegetables were split up because of the high vitamin A content in green leafy vegetables, and fruits were split up because of the high vitamin C content in citrus fruits). These food groups were then analyzed with a PCA to identify dietary patterns that explained the maximum variation of food intake. To minimize correlation between the dietary patterns, a Varimax rotation was used. In determining the number of dietary patterns (factors) that should remain; we interpreted the scree test for factors with an Eigenvalue of 1.0 or greater. The factors were then used to rank participants in low to high adherence for each of the dietary patterns. These ranks were used to create quartiles, which explain how good a participant fits in a specific pattern.

      Covariates

      Age, sex, skin color (
      • Jacobs L.C.
      • Hamer M.A.
      • Verkouteren J.A.
      • Pardo L.M.
      • Liu F.
      • Nijsten T.
      Perceived skin colour seems a swift, valid and reliable measurement.
      ), height, weight, and season were documented during the visit to the research center. During the home interview, participants were asked about their education level (low = primary education, medium = lower–intermediate vocational education, high = general secondary education and higher) and tobacco use (never and former vs. current). Physical activity was assessed using the Longitudinal Aging Study Amsterdam Physical Activity Questionnaire and expressed in metabolic equivalent hours/week (
      • Stel V.S.
      • Smit J.H.
      • Pluijm S.M.
      • Visser M.
      • Deeg D.J.
      • Lips P.
      Comparison of the LASA Physical Activity Questionnaire with a 7-day diary and pedometer.
      ). Participants were categorized as supplement users if they used supplements at least once a week.

      Statistical analyses

      Missing data on the covariates were imputed using the fully conditional specification imputation method with 20 imputations (
      • Sterne J.A.
      • White I.R.
      • Carlin J.B.
      • Spratt M.
      • Royston P.
      • Kenward M.G.
      • et al.
      Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls.
      ). Multivariable logistic regression was used to calculate the strength of the association between total antioxidant capacity and seborrheic dermatitis and the PCA-extracted dietary patterns and seborrheic dermatitis, both adjusted for age, sex, skin color, smoking, total energy intake, body mass index, season, physical activity, education, and supplement use. In addition, we tested for interaction between the dietary outcomes and all other variables.
      The threshold for significance was set at a P-value of 0.05. All analyses were conducted in IBM SPSS Statistics for Windows, version 21.0 (IBM, Armonk, NY).

      Conflict of Interest

      RG is an employee of Unilever with a shareholding in this company. TN received research grants from Unilever. The other authors state no conflict of interest.

      Acknowledgments

      The Rotterdam Study is funded by Erasmus Medical Center and Erasmus University Rotterdam; Netherlands Organization for the Health Research and Development (ZonMw); the Research Institute for Diseases in the Elderly (RIDE); the Ministry of Education, Culture and Science ; the Ministry for Health, Welfare and Sports ; the European Commission (DG XII); and the Municipality of Rotterdam . MGHS is supported by Unilever , and RG is a Unilever employee.

      References

        • Beckenbach L.
        • Baron J.M.
        • Merk H.F.
        • Loffler H.
        • Amann P.M.
        Retinoid treatment of skin diseases.
        Eur J Dermatol. 2015; 25: 384-391
        • Bedard A.
        • Riverin M.
        • Dodin S.
        • Corneau L.
        • Lemieux S.
        Sex differences in the impact of the Mediterranean diet on cardiovascular risk profile.
        Br J Nutr. 2012; 108: 1428-1434
        • Bickers D.R.
        • Athar M.
        Oxidative stress in the pathogenesis of skin disease.
        J Invest Dermatol. 2006; 126: 2565-2575
      1. Brink L, Smeets AP-, Stafleu A, Wolvers D. The Netherlands Nutrition Centre. Richtlijnen Schijf van Vijf. http://www.voedingscentrum.nl/professionals/schijf-van-vijf/naslag-richtlijnen-schijf-van-vijf.aspx. 2016. Accessed 4 October 2018.

        • Carlsen M.H.
        • Halvorsen B.L.
        • Holte K.
        • Bohn S.K.
        • Dragland S.
        • Sampson L.
        • et al.
        The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
        Nutr J. 2010; 9: 3
        • Cepeda A.M.
        • Del Giacco S.R.
        • Villalba S.
        • Tapias E.
        • Jaller R.
        • Segura A.M.
        • et al.
        A traditional diet is associated with a reduced risk of eczema and wheeze in Colombian children.
        Nutrients. 2015; 7: 5098-5110
        • Dugrand A.
        • Olry A.
        • Duval T.
        • Hehn A.
        • Froelicher Y.
        • Bourgaud F.
        Coumarin and furanocoumarin quantitation in citrus peel via ultraperformance liquid chromatography coupled with mass spectrometry (UPLC-MS).
        J Agric Food Chem. 2013; 61: 10677-10684
      2. Dutch Food Composition Table. Nevo tabel 2006/2011. Voorlichtingsbureau voor de Voeding. The Hague, The Netherlands, 2011. https://www.rivm.nl/en/Topics/D/Dutch_Food_Composition_Database/Publications. Accessed 4 October 2018.

        • Emre S.
        • Metin A.
        • Demirseren D.D.
        • Akoglu G.
        • Oztekin A.
        • Neselioglu S.
        • et al.
        The association of oxidative stress and disease activity in seborrheic dermatitis.
        Arch Dermatol Res. 2012; 304: 683-687
        • Etemadi A.
        • Sinha R.
        • Ward M.H.
        • Graubard B.I.
        • Inoue-Choi M.
        • Dawsey S.M.
        • et al.
        Mortality from different causes associated with meat, heme iron, nitrates, and nitrites in the NIH-AARP Diet and Health Study: population based cohort study.
        BMJ. 2017; 357: j1957
        • Feunekes G.I.
        • Van Staveren W.A.
        • De Vries J.H.
        • Burema J.
        • Hautvast J.G.
        Relative and biomarker-based validity of a food-frequency questionnaire estimating intake of fats and cholesterol.
        Am J Clin Nutr. 1993; 58: 489-496
        • Giugliano D.
        • Ceriello A.
        • Esposito K.
        The effects of diet on inflammation: emphasis on the metabolic syndrome.
        J Am Coll Cardiol. 2006; 48: 677-685
        • Goldbohm R.A.
        • van den Brandt P.A.
        • Brants H.A.
        • van ’t Veer P.
        • Al M.
        • Sturmans F.
        • et al.
        Validation of a dietary questionnaire used in a large-scale prospective cohort study on diet and cancer.
        Eur J Clin Nutr. 1994; 48: 253-265
        • Gupta A.K.
        • Bluhm R.
        Seborrheic dermatitis.
        J Eur Acad Dermatol. 2004; 18: 13-26
        • He F.J.
        • Nowson C.A.
        • MacGregor G.A.
        Fruit and vegetable consumption and stroke: meta-analysis of cohort studies.
        Lancet. 2006; 367: 320-326
        • Hu F.B.
        Dietary pattern analysis: a new direction in nutritional epidemiology.
        Curr Opin Lipidol. 2002; 13: 3-9
        • Hu F.B.
        • Rimm E.
        • Smith-Warner S.A.
        • Feskanich D.
        • Stampfer M.J.
        • Ascherio A.
        • et al.
        Reproducibility and validity of dietary patterns assessed with a food-frequency questionnaire.
        Am J Clin Nutr. 1999; 69: 243-249
        • Ikram M.A.
        • Brusselle G.G.O.
        • Murad S.D.
        • van Duijn C.M.
        • Franco O.H.
        • Goedegebure A.
        • et al.
        The Rotterdam Study: 2018 update on objectives, design and main results.
        Eur J Epidemiol. 2017; 32: 807-850
        • Innes J.K.
        • Calder P.C.
        Omega-6 fatty acids and inflammation.
        Prostaglandins Leukot Essent Fatty Acids. 2018; 132: 41-48
        • Jacobs L.C.
        • Hamer M.A.
        • Verkouteren J.A.
        • Pardo L.M.
        • Liu F.
        • Nijsten T.
        Perceived skin colour seems a swift, valid and reliable measurement.
        Br J Dermatol. 2015; 173: 1084-1086
        • Klein S.L.
        • Flanagan K.L.
        Sex differences in immune responses.
        Nat Rev Immunol. 2016; 16: 626-638
        • Kwon H.H.
        • Yoon J.Y.
        • Hong J.S.
        • Jung J.
        • Park M.S.
        • Suh D.H.
        Clinical and histological effect of a low glycaemic load diet in treatment of acne vulgaris in Korean patients: a randomized, controlled trial.
        Acta Derm Venereol. 2012; 92: 241-246
        • Millsop J.W.
        • Bhatia B.K.
        • Debbaneh M.
        • Koo J.
        • Liao W.
        Diet and psoriasis, part III: role of nutritional supplements.
        J Am Acad Dermatol. 2014; 71: 561-569
        • Misery L.
        • Touboul S.
        • Vincot C.
        • Dutray S.
        • Rolland-Jacob G.
        • Consoli S.G.
        • et al.
        Stress et dermatite séborrhéique [Stress and seborrheic dermatitis].
        Ann Dermatol Venereol. 2007; 134: 833-837
        • Ozawa M.
        • Shipley M.
        • Kivimaki M.
        • Singh-Manoux A.
        • Brunner E.J.
        Dietary pattern, inflammation and cognitive decline: the Whitehall II prospective cohort study.
        Clin Nutr. 2017; 36: 506-512
        • Ozturk P.
        • Arican O.
        • Belge Kurutas E.
        • Karakas T.
        • Kabakci B.
        Oxidative stress in patients with scalp seborrheic dermatitis.
        Acta Dermatovenerol Croat. 2013; 21: 80-85
        • Pantavos A.
        • Ruiter R.
        • Feskens E.F.
        • de Keyser C.E.
        • Hofman A.
        • Stricker B.H.
        • et al.
        Total dietary antioxidant capacity, individual antioxidant intake and breast cancer risk: the Rotterdam Study.
        Int J Cancer. 2015; 136: 2178-2186
        • Perdigoto C.N.
        • Valdes V.J.
        • Bardot E.S.
        • Ezhkova E.
        Epigenetic regulation of epidermal differentiation.
        Cold Spring Harb Perspect Med. 2014; 4: a015263
        • Pirkhammer D.
        • Seeber A.
        • Honigsmann H.
        • Tanew A.
        Narrow-band ultraviolet B (ATL-01) phototherapy is an effective and safe treatment option for patients with severe seborrhoeic dermatitis.
        Br J Dermatol. 2000; 143: 964-968
        • Sanders M.G.H.
        • Pardo L.M.
        • Franco O.H.
        • Ginger R.S.
        • Nijsten T.
        Prevalence and determinants of seborrhoeic dermatitis in a middle-aged and elderly population: the Rotterdam Study.
        Br J Dermatol. 2018; 178: 148-153
        • Sanders M.G.H.
        • Pardo L.M.
        • Uitterlinden A.G.
        • Smith A.M.
        • Ginger R.S.
        • Nijsten T.
        The genetics of seborrheic dermatitis: a candidate gene approach and pilot genome-wide association study.
        J Invest Dermatol. 2018; 138: 991-993
        • Schoufour J.D.
        • de Jonge E.A.L.
        • Kiefte-de Jong J.C.
        • van Lenthe F.J.
        • Hofman A.
        • Nunn S.P.T.
        • Franco O.H.
        Socio-economic indicators and dietary quality in an elderly population.
        Maturitas. 2018; 107: 71-77
        • Stel V.S.
        • Smit J.H.
        • Pluijm S.M.
        • Visser M.
        • Deeg D.J.
        • Lips P.
        Comparison of the LASA Physical Activity Questionnaire with a 7-day diary and pedometer.
        J Clin Epidemiol. 2004; 57: 252-258
        • Sterne J.A.
        • White I.R.
        • Carlin J.B.
        • Spratt M.
        • Royston P.
        • Kenward M.G.
        • et al.
        Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls.
        BMJ. 2009; 338: b2393
        • Stringa N.
        • Brahimaj A.
        • Zaciragic A.
        • Dehghan A.
        • Ikram M.A.
        • Hofman A.
        • et al.
        Relation of antioxidant capacity of diet and markers of oxidative status with C-reactive protein and adipocytokines: a prospective study.
        Metabolism. 2017; 71: 171-181
        • Trouba K.J.
        • Hamadeh H.K.
        • Amin R.P.
        • Germolec D.R.
        Oxidative stress and its role in skin disease.
        Antioxid Redox Signal. 2002; 4: 665-673
        • Turck D.
        • Bresson J.L.
        • Burlingame B.
        • Dean T.
        • Fairweather-Tait S.
        • Heinonen M.
        • et al.
        Guidance for the scientific requirements for health claims related to antioxidants, oxidative damage and cardiovascular health.
        EFSA J. 2018; 16: 5136
        • Willett W.C.
        • Howe G.R.
        • Kushi L.H.
        Adjustment for total energy intake in epidemiologic studies.
        Am J Clin Nutr. 1997; 65 (1220S–8S)
        • Wu S.
        • Han J.
        • Feskanich D.
        • Cho E.
        • Stampfer M.J.
        • Willett W.C.
        • et al.
        Citrus consumption and risk of cutaneous malignant melanoma.
        J Clin Oncol. 2015; 33: 2500-2508