| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From Corporate Research (MCC, OHF, and AEM) and Statistics Group (PGM), Unilever Colworth Park, Bedford, United Kingdom, and Unilever R&D Port Sunlight, Wirral, United Kingdom (SPG)
2 Presented in part in poster form at the 4th International Conference on the Health Benefits of Fruit and Vegetables within a Mediterranean style diet (EGEA IV), Brussels, Belgium, 17-19 April 2007. 3 Supported by Unilever Corporate Research United Kingdom. 4 Reprints not available. Address correspondence to MC Cosgrove, Unilever Corporate Research, Colworth Park, Sharnbrook, Bedford, MK44 1LQ. United Kingdom. E-mail: maeve.cosgrove{at}unilever.com.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We evaluated the associations between nutrient intakes and skin-aging appearance.
Design: Using data from the first National Health and Nutrition Examination Survey, we examined associations between nutrient intakes and skin aging in 4025 women (40–74 y). Nutrients were estimated from a 24-h recall. Clinical examinations of the skin were conducted by dermatologists. Skin-aging appearance was defined as having a wrinkled appearance, senile dryness, and skin atrophy.
Results: Higher vitamin C intakes were associated with a lower likelihood of a wrinkled appearance [odds ratio (OR) 0.89; 95% CI: 0.82, 0.96] and senile dryness (OR: 0.93; 95% CI: 0.87, 0.99). Higher linoleic acid intakes were associated with a lower likelihood of senile dryness (OR: 0.75; 95% CI: 0.64, 0.88) and skin atrophy (OR: 0.78; 95% CI 0.65, 0.95). A 17-g increase in fat and a 50-g increase in carbohydrate intakes increased the likelihood of a wrinkled appearance (OR: 1.28 and 1.36, respectively) and skin atrophy (OR: 1.37 and 1.33, respectively). These associations were independent of age, race, education, sunlight exposure, income, menopausal status, body mass index, supplement use, physical activity, and energy intake.
Conclusions: Higher intakes of vitamin C and linoleic acid and lower intakes of fats and carbohydrates are associated with better skin-aging appearance. Promoting healthy dietary behaviors may have additional benefit for skin appearance in addition to other health outcomes in the population.
Key Words: National Health and Nutrition Examination Surveys • NHANES • vitamin C • wrinkles • skin aging • linoleic acid • nutritional epidemiology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Balanced nutrition is essential not only to prevent chronic disease such as cardiovascular disease, certain cancers, and diabetes (13) but also to maintain health and ensure normal functioning. Certain nutrients were identified to play a critical role in the normal functioning of the skin, particularly when nutrient deficiencies are apparent, eg, vitamin C in collagen synthesis (14, 15). Several studies have observed improved protection of the skin against sun damage (photoprotection) by dietary supplementation with vitamins E and C, carotenoids (ß-carotene and lycopene), and polyunsaturated fatty acids (PUFAs) (16, 17). However, those studies are limited by the use of supplements, some with several active ingredients, making it difficult to determine which nutrient is having an effect. One study observed a photoprotective effect of a diet higher in vegetables, fruit, and olive oil on the skin (18). Whether habitual dietary intakes have a significant effect on skin-aging appearance has not been shown. This is critical to adequately design potential interventions targeted to improve or delay the skin-aging process. Therefore, in the present cross-sectional analysis, using the first National Health and Nutrition Examination Survey (NHANES I) (19), we examined the relation between nutrient intakes and the prevalence of the appearance of wrinkles, senile dryness (dryness as a result of aging), and skin atrophy (thinning) in middle-aged women. This examination allowed us to examine the relation between nutrient intakes, rather than supplements, and skin-aging appearance, rather than photoprotection, for the first time to our knowledge.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
32 000 noninstitutionalized persons aged 1–74 y. Details of methods, including the development, plan, and operation of the survey and data collection forms have been published and are available elsewhere (20-22).
Study sample
All women aged 
40 y were eligible for inclusion. From the original available sample of 23 808 subjects, 10 152 men, 8477 women aged < 40 y, 1062 women who did not have a dermatologic examination, and 92 women with unsatisfactory 24-h dietary recalls were excluded. The final sample consisted of 4025 women aged 40–74 y.
Assessment of skin aging
A complete clinical dermatologic examination of the skin was undertaken to evaluate variations in texture and color, certain manifestations of aging, and all pathologic changes. The dermatologic examinations were performed by 101 dermatologists that followed a studywide protocol after uniform training and standardized definitions (23). To ensure consistency with the examination protocol, a random sample of the 20 637 examinations was checked by a senior dermatologist (24). The dermatologist classified the subjects into 1 of 3 categories of cumulative lifetime sunlight exposure: 1) low (unimpressive), 2) moderate, and 3) high (considerable), based on the subjects’ occupation and amount of leisure time spent outdoors. Skin aging was defined by 3 independent determinants of the process: wrinkled appearance, senile dryness (dryness as a result of aging), and skin atrophy (thinning).
Dietary assessment
A 24-h dietary recall was administered to each respondent by a trained dietary interviewer, with 3-dimensional food models, including household measures, to estimate food portions (23). The interviewer probed for clarity of foods and beverages consumed and for commonly forgotten items (eg, sugar in tea, beverages with meals). Estimates of nutrient intake for each food and beverage reported were obtained from the US Department of Agriculture food composition data (25).
Assessment of other variables
Height and weight were measured using standard methods (26), and BMI was calculated. The physical activity question about nonrecreational activity was: "In your usual day, aside from recreation, how active are you?" The possible responses were 1) very active, 2) moderately active, or 3) quite inactive. Additional covariate information about age, smoking habits, menopausal status, race, education, and family income was obtained with the use of questionnaires.
Statistical analyses
We first conducted univariate analyses to describe the distribution of nutrient intakes and demographic and lifestyle attributes by having a wrinkled appearance, senile dryness, and skin atrophy. Significant differences were assessed using Pearson's chi-square and independent t tests for categorical and continuous variables, respectively. Differences in mean nutrient intakes were assessed with the use of analysis of variance, adjusting for age (continuous), race (white, black, other), energy intake (continuous), education [<12 y, high school graduation (12 y), >12 y], sunlight exposure (unimpressive, moderate, considerable), total family income (<$5000, $5000–9999; $10 000), menopausal status (responded yes to the question "Have your menstrual periods stopped entirely?" compared with no), BMI (continuous), supplement use (responded yes to the question "Are you taking vitamins or minerals?" compared with no), and self-reported daily physical activity (very active, moderately active, quite inactive). Intakes of linoleic acid, total dietary cholesterol, calcium, and vitamins A and C were log transformed, because they were not normally distributed.
To determine the associations of nutrient intakes with skin-aging appearance, nutrients that substantially changed the effect on the outcomes (P < 0.2) were examined with the use of binary logistic regression analysis (27). We used logistic regression models in 3808 women with complete data, controlling for age, race, energy intake, education, sunlight exposure, family income, menopausal status, BMI, supplement use, and physical activity to determine the associations of fat, carbohydrate, thiamine, and vitamin C with a wrinkled appearance; of saturated fatty acids, oleic acid, linoleic acid, and vitamin C with senile dryness; and of fat, carbohydrate, and linoleic acid with skin atrophy. Variables that substantially changed the effect on the outcomes examined (P < 0.2) (27) or were associated with skin aging in prior studies (3-5, 22) were considered as potential confounders. Family income may be related to skin-aging appearance thorough pathways other than its influence on diet; therefore, it was included in all models. The odds ratio (OR) represents the odds for a 1-unit increment (eg, 1 g protein). To illustrate nutritionally relevant estimates, new ORs were calculated on the basis of 33% of the median intake for each nutrient (except for log-transformed nutrients and thiamine, whose median intake was < 1 mg) (28).
Subgroup analysis
Cigarette smoking is a well-established independent risk factor for facial wrinkling and skin aging (6-9). In this analysis, smoking data were only collected for 1401 women (n = 539 current and previous smokers; n = 862 never smoked). Therefore, stratified analysis was conducted to examine skin-aging appearance without ignoring the smoking data or losing large sample numbers because of the missing data.
We used a P value < 0.05 for significance, and all tests were 2-sided. All analyses were performed with SPSS, version 14 (SPSS Inc, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Of the 3 signs of skin-aging appearance, wrinkles were significantly correlated with senile dryness (r = 0.36) and skin atrophy (r = 0.52). Senile dryness was significantly correlated with skin atrophy (r = 0.30). Characteristics of all women and of each outcome separately are shown in Table 1. Skin-aging outcomes were found significantly in older women; women with a wrinkled appearance, senile dryness, and skin atrophy were 10 y, 10 y, and 9 y older, respectively. Women with the appearance of wrinkles, with senile dryness, and with skin atrophy were more likely to be white, have <12 y education, have a low family income, be postmenopausal, and have higher sunlight exposure. Women with skin atrophy were significantly associated with a lower BMI, and women with a wrinkled appearance and with senile dryness were less likely to be physically active. No significant differences in supplement users and skin-aging outcomes were observed.
|
|
. Women with a wrinkled appearance had significantly lower intakes of protein, total dietary cholesterol, phosphorus, potassium, vitamin A, and vitamin C than did women without a wrinkled appearance. Women with senile dry skin had significantly lower intakes of linoleic acid and vitamin C than did women without senile dry skin. Women with skin atrophy had a significantly lower linoleic acid intake than did women without skin atrophy.
|
. A 1-unit increase on the log scale in intakes of vitamin C were associated with an 11% reduction in the odds of a wrinkled appearance and a 7% reduction in the odds of senile dryness. Similarly, a 1-unit increase on the log scale in intakes of linoleic acid were associated with 25% and 22% reductions in the odds of senile dryness and skin atrophy, respectively. A 17-g increase in fat and a 50-g increase in carbohydrate intakes increased the ORs of a wrinkled appearance and skin atrophy. The multivariate-adjusted logistic regression models yielded an R2 of 0.17, 0.16, and 0.15 for a wrinkled appearance, senile dryness, and skin atrophy, respectively. Variables that substantially affected the regression coefficients in the models were age, race, and sunlight exposure for all outcomes, plus menopausal status for wrinkles and senile dryness and BMI for skin atrophy. Supplement use was not significantly associated with any of the 3 signs of skin-aging appearance; no interactions were observed between mean nutrient intakes and supplement use (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Lower intakes of vitamin C were significantly associated with the prevalence of a wrinkled appearance and senile dryness, independent of age, sun exposure, race, menopausal status, energy intake, education, family income, BMI, supplement use, and physical activity. To our knowledge, this is the first study to directly relate dietary intakes of vitamin C with skin aging. In skin, vitamin C exerts different biologic roles, including participation in collagen synthesis, the regeneration process, and wound repair (29). Vitamin C is an important antioxidant found in the skin and may lower the prevalence of wrinkles and senile dryness by its actions as an antioxidant (29). Moreover, several studies have shown that vitamin C has photoprotective properties through oral or topical applications (17). Topical application of vitamin C shows photoprotective properties and suggests improvements of wrinkles (30, 31). Reviewing dietary supplementation, Boelsma et al (16) identified 4 studies that showed a photoprotective effect of vitamin C on skin. Those studies were short-term supplementation trials with high doses of vitamin C and in combination with vitamin E, whereas our findings are from habitual intakes of vitamin C from food sources.
Our results also suggest that a higher dietary intake of linoleic acid has a beneficial role in reducing the chances of developing senile dryness and skin atrophy in middle-aged women. Linoleic acid (n–6 PUFA) is an essential fatty acid that cannot be produced endogenously and is converted after ingestion to other PUFAs such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The association between linoleic acid intake and skin aging has not been investigated; however, several studies have examined the photoprotective effects of EPA and DHA intakes from fish-oil supplements (16, 17). Topical application of EPA has also showed photoprotective properties (32). Linoleic acid may therefore lower the prevalence of senile dryness and skin atrophy by acting as a source of EPA and DHA. Furthermore, low intake of linoleic acid can lead to dermatitis with marked abnormalities (16, 33), and this could also be important in its role in skin-aging appearance.
Our results also suggest that a higher dietary intake of fats and carbohydrates has a negative role in skin-aging appearance. To our knowledge this is the first study to show an association between dietary intakes of fats and carbohydrates with features of skin aging. Furthermore, higher thiamine intakes were associated with an increased likelihood of a wrinkled appearance. Thiamine is an essential B vitamin found in enriched cereals and whole grains. More research is needed to understand these associations.
Our findings add evidence to a predominately supplement and topical application-based hypothesis that what we eat affects our skin-aging appearance. One other study by Purba et al (18) has reported a significant effect of vegetables, fruit, and olive oil on wrinkles; however, they examined the effect on skin surface microstructure photodamage, whereas in our study we looked at the effect of nutrient intakes on visual features of skin aging that were determined by trained dermatologists. Because most existing evidence comes from supplemental trials or topical application of nutrients (16-18, 30-32, 34), our study adds to the existing evidence in this field, because it examined nutrient intakes from foods rather than from nutritional supplements.
Although vitamin A has long been noted to have antiwrinkle properties, and retinol (vitamin A) is commonly used in the cosmetics industry as a topical antiwrinkle agent (34, 35), clinical trials have failed to show this effect when taken orally as supplements. Our findings that women with a wrinkled appearance had lower vitamin A intakes support the evidence that vitamin A benefits skin-aging appearance. We also found that women with a wrinkled appearance had lower protein intakes. Lower protein intakes in older adults were shown to increase skin fragility (36). Vitamin A and protein however did not affect the prevalence of a wrinkled appearance in the multivariate-adjusted logistic regression models.
The favorable association of vitamin C and linoleic acid intakes with skin aging may be attributed to the dietary sources of these nutrients. At the time of this survey, the main sources of vitamin C in the US diet were orange juice (more than a quarter of total vitamin C intake), citrus fruit, fruit juices, and tomatoes (37). Linoleic acid is found in oils such as rapeseed and soybean oils and in foods such as green leafy vegetables and nuts (38). Furthermore, as mentioned earlier, the body can convert linoleic acid into EPA and DHA (n–3 PUFAs). These nutrients were not measured in NHANES I, but their main dietary sources are fish and fish oils (39). Current dietary recommendations promote higher intakes of fruit and vegetables and fish and PUFAs (13, 40, 41). Despite several campaigns to promote the consumption of fruit and vegetables and the evidence that diets poor in fruit and vegetables are associated with poor health outcomes, most adults are still not eating enough fruit and vegetables (42, 43). Therefore, a benefit for skin-aging appearance from eating aspects of a healthy diet, such as fruit, vegetables, nuts, and fish, and reducing fat intake may motivate people and improve current promotions for healthy eating.
Our findings need to be interpreted while considering some limitations. This study is a cross-sectional analysis, and no conclusions about the direction of associations between nutrient intakes and skin-aging appearance can be determined. NHANES I data were collected in 1971–1975 and a cohort effect cannot be ruled out. Furthermore, changes in nutrient intakes have occurred in the past 30 y (44); hence, extrapolation to current populations might be affected. Although we used NHANES I data, which has a well-characterized population with good measures of a number of variables such as age, sun exposure, race, BMI, and menopausal status, allowing potential confounding to be controlled for, residual confounding cannot be discarded. One factor that was not measured was the use of facial cosmetics. Nevertheless, on the basis of current knowledge we cannot determine whether use of such products is modified by nutrient intake. The dermatologic component of this survey was not designed to specifically detect associations between nutrients and skin appearance; therefore, post hoc analyses have to be interpreted with caution. However, few studies exist in which comprehensive measures of both dietary intake data and skin-aging appearance are present and thus can be used to test the hypothesis of this study. The dietary data were collected using a 24-h dietary recall. Estimates from a single day of intake may not represent usual, long-term intakes, because of day-to-day variation in the subjects’ food intakes. However, this study used trained interviewers to improve data collection.
This is the first study to examine the effect of nutrient intakes rather than supplements on skin-aging appearance. Our findings suggest that higher intakes of vitamin C and linoleic acid are associated with a lower prevalence of a wrinkled appearance, senile dryness, and skin atrophy, whereas higher intakes of fats and carbohydrates are associated with a higher likelihood of features of skin aging. The favorable associations may be attributable to the dietary sources of these nutrients (fruit, vegetables, and nuts) and are independent of factors known to affect skin aging. Perhaps appealing benefits such as reducing skin-aging appearance may motivate healthy eating, and new campaigns to promote healthy dietary behaviors could consider this issue. Our findings support current recommendations that promote aspects of a healthy diet such as higher intakes of fruit, vegetables, and nuts and indicate a new direction for nutrition research in relation to public health.
|
ACKNOWLEDGMENTS |
|---|
The author's responsibilities were as follows—MCC: was responsible for the design and concept, analysis of the study, data interpretation, and preparation of the manuscript; OHF: commented on this work and helped prepare the manuscript; SPG: was responsible for study conception and design; PGM: was responsible for acquisition of the data and participated in the data analysis and data interpretation; AEM (study supervisor): was responsible for skin content and helped revise the manuscript. None of the authors had a conflict of interest in relation to this study.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
