Depression is a common and very serious medical condition accompanied by a high degree of emotional distress and functional impairment(Reference Ayuso-Mateos, Vazquez-Barquero and Dowrick1). The two main symptoms of depression include depressed mood and anhedonia (i.e. loss of interest in daily activities), which are often accompanied by a variety of additional symptoms(Reference Uher, Payne and Pavlova2). The lifetime prevalence of depression varies from 11·1 % in developing countries to 14·6 % in high-income countries(Reference Bromet, Andrade and Hwang3). Moreover, depression is estimated to count for 9·6 % of years living with disability and about 4·0 % of global disability-adjusted life years(Reference Ferrari, Charlson and Norman4). The prevalence of depression substantially increases during adolescence, particularly in girls, due to physical and emotional development as well as exposure to novel stressors(Reference Hankin, Abramson and Moffitt5). A systematic review and meta-analysis of twelve studies revealed that the prevalence of depression among Iranian children and adolescents was 43·5 % using the Beck Depression Inventory, with a higher prevalence among girls than boys(Reference Sajjadi, Kamal and Rafiey6). Moreover, using the Center for Epidemiologic Studies Depression Scale, the prevalence of severe depression among Iranian female adolescents in secondary schools has been reported to be 52·6 %(Reference Moeini, Bashirian and Soltanian7). Since depression is probably the most important risk factor for teenage suicide(Reference Windfuhr, While and Hunt8), the management of adolescents with depressive disorders is of particular importance.
Inflammation and oxidative stress are linked with a number of chronic diseases including CVD, hypertension, diabetes, chronic kidney disease and cancer(Reference Ridker9–Reference Madeddu, Gramignano and Floris12). Although the aetiology of depression is not completely understood, it has been shown that depression is significantly associated with systemic inflammation up-regulation indicated by an increased production of pro-inflammatory cytokines and inflammatory mediators(Reference Shafiee, Tayefi and Hassanian13,Reference Tayefi, Shafiee and Kazemi-Bajestani14) . Moreover, several studies have reported that disturbed oxidative stress/antioxidant equilibrium increased the generation of reactive oxygen species and decreased antioxidant defences may be related to pathogenesis of depression(Reference Shafiee, Ahmadnezhad and Tayefi15–Reference Sarandol, Sarandol and Eker17). In line with this, Rawdin et al. observed that the homoeostatic buffering mechanisms regulating inflammation and oxidation in healthy subjects become dysregulated in untreated depression(Reference Rawdin, Mellon and Dhabhar18). Thus, considering the link between depressive disorders and oxidative stress, it was hypothesised that higher dietary antioxidant intake would be associated with less severe depressive symptoms.
The most important dietary antioxidants include vitamin C (ascorbic acid), vitamin E (tocopherols and tocotrienols), carotenoids (e.g. β-carotene), flavonoids and several trace minerals. Moreover, it should be noted that dietary fibre often serves as a carrier for dietary antioxidants(Reference Saura-Calixto19). We have previously found that dietary Zn intake was inversely associated with depressive symptoms, which was suggested to be related to its role in antioxidant enzymes(Reference Gonoodi, Moslem and Ahmadnezhad20). Several previous studies have investigated the association between dietary antioxidants and depressive symptoms in different population groups(Reference Bonaccio, Di Castelnuovo and Bonanni21–Reference Xu, Li and Song24). Bonaccio et al. observed that adherence to a Mediterranean dietary pattern was associated with better mental health, which was independently explained by dietary total antioxidant and fibre content(Reference Bonaccio, Di Castelnuovo and Bonanni21). In another study, Payne et al. observed that vitamin C, lutein and β-cryptoxanthin intakes were significantly lower among participants with depression than in healthy individuals(Reference Payne, Steck and George22). Moreover, Ye et al. observed an inverse association between intake of total carotenoids and depression score after adjustment for a range of dietary and non-dietary potential confounders(Reference Ye, Scott and Falcon23). Xu and colleagues conducted a study in 16 807 adults aged 20 years or older and observed an inverse association between dietary intakes of total, fruit and vegetable fibre, and depressive symptoms(Reference Xu, Li and Song24).
Despite these observations, to our knowledge, no studies to date have investigated the association between dietary antioxidants and fibre intake and depressive symptoms in Iranian adolescent girls, a population known to have a high prevalence of depression. Therefore, the primary objective of the present study was to investigate (1) the association between dietary intake of antioxidants and severity of depression symptoms and (2) the association between dietary intake of fibre and severity of depression symptoms among Iranian adolescent girls.
Methods
Study population
A total of 1026 adolescent girls aged 12–18 years were randomly recruited from several schools in two cities located in northeastern Iran (Mashhad and Sabzevar), using a random cluster sampling method. The participation rate was 96 %, and a total of 988 adolescent girls were studied. Subjects with a history of autoimmune diseases, cancer, metabolic bone disease, hepatic or renal failure, cardiovascular disorders, malabsorption or thyroid, parathyroid or adrenal diseases were not included in our study. All the subjects and their parents were asked to complete written informed consent before participating in the study. This study was approved by the ethic committee of Mashhad University of Medical Sciences (MUMS), Mashhad, Iran (ID: 931188).
Demographic, anthropometric and biochemical measurements
Demographic information of the study participants including age, supplement use, chronic diseases, smoking status, menstruation status, psychological treatment and medical history were collected by trained interviewers. Anthropometric variables including weight, height and waist circumference were obtained using standard protocols. Body weight was measured in an overnight fasting status without shoes in a minimal clothing state by the use of a digital scale to the nearest 0·1 kg. Height (without shoes) was assessed to the nearest 0·1 cm using a stadiometer. Thereafter, BMI was calculated by dividing weight (kg) by the square of height (m2). Waist circumference was measured at the midpoint between the lowest rib margin and the iliac crest during minimal respiration(Reference Bahrami, Mazloum and Maghsoudi25,26) . Systolic blood pressure and diastolic blood pressure were measured by sphygmomanometer twice in exactly the same manner. It was measured on the left arm when the individuals remained seated at rest for 15 min. The third measurement was taken if the first two readings differ by more than 15 mmHg in diastolic blood pressure or more than 25 mmHg in systolic blood pressure. Physical activity information was gathered using a validated questionnaire allowing conversion into metabolic equivalent hours(Reference Delshad, Ghanbarian and Ghaleh27). Blood samples were collected after 12–14 h overnight fast, and serum was separated after centrifugation and stored at −20 °C until analysed. Biochemical parameters including TAG, total cholesterol, LDL-cholesterol, HDL-cholesterol and fasting blood glucose were measured for all participants, as described previously(Reference Khayyatzadeh, Bagherniya and Fazeli28). Friedewald formula was used to calculate LDL-cholesterol if serum TAG concentration was lower than 400 mg/dl(Reference Castelli, Garrison and Wilson29). Serum high-sensitivity C-reactive protein concentration was estimated using an immunoturbidimetric method, with a detection limit of 0·06 mg/l (Pars Azmun, Karaj, Iran).
Assessment of dietary intake
A validated FFQ was used to measure dietary intakes(Reference Asghari, Rezazadeh and Hosseini-Esfahani30,Reference Hosseini Esfahani, Asghari and Mirmiran31) . To estimate energy and nutrient intakes, the reported portion size in FFQ and dietary records was converted to grams using household measures and then entered to the Nutritionist IV software (version 7.0; N-Squared Computing) which was modified for Iranian food items.
Assessment of depression
In this study, the twenty-one-item Beck Depression Inventory was used for assessing the symptoms of depression(Reference Dozois, Dobson and Ahnberg32). Each item represents a single symptom associated with depression including feelings of guilt, feelings of hopelessness, sadness, crying, sleep disturbance, fear and loss of appetite over the past 2 weeks(Reference Scogin, Beutler and Corbishley33). Scores are classified as the following: 0–13 minimal or no depression, 14–19 mild depression, 20–28 moderate depression and 29–63 severe depression(Reference Scogin, Beutler and Corbishley33). Ghassemzadeh et al. (2005) have validated this questionnaire in its Persian (Farsi) translation, with an acceptable internal consistency (Cronbach’s alpha = 0·87) and test–retest reliability (r = 0·74)(Reference Ghassemzadeh, Mojtabai and Karamghadiri34).
Statistical analysis
Data analysis was carried out using SPSS-18 software (SPSS Inc.). The normality of data was evaluated using Kolmogorov–Smirnov test. Descriptive statistics including mean, frequency and sd were determined for all variables and expressed as mean ± sd for normally distributed variables. χ 2 tests were used to compare the qualitative variables. For normally distributed variables, independent sample t test was performed. All the analyses were two-sided, and P-value < 0·05 was considered as significant. Moreover, crude and adjusted logistic regression analyses were conducted to investigate the relationship between depression symptoms and quartiles of nutrient intakes. Age and energy intake were controlled in the first model. Further adjustments were made for menstruation, family members, parental death, parental divorce, physical activity and BMI in the second model.
Results
Over 25 % (n 255) of the subjects were diagnosed with mild-to-severe depression symptoms, and about 74 % had no or minimal depression symptoms. Demographic and biochemical characteristics of study participants in no or minimal and mild-to-severe groups are presented in Table 1. There were no significant differences in age, weight, BMI, waist circumference, physical activity, menstruation, high-sensitivity C-reactive protein, fasting blood glucose, HDL-cholesterol, LDL-cholesterol, total cholesterol and TAG between subjects with no or minimal or mild-to-severe depression symptoms. Patients with mild-to-severe depression symptoms had significantly lower diastolic blood pressure compared with healthy subjects (P = 0·03), but there was no significant difference in the case of systolic blood pressure (P = 0·1). The percentage of passive smokers was significantly higher in mild to severe group as compared with no or minimal group (P = 0·006).
Table 1 Demographic and biochemical characteristics of study population with no or minimal and mild-to-severe depression symptoms
WC, waist circumference; MET, metabolic equivalent; SBP, systolic blood pressure; DBP, diastolic blood pressure; hs-CRP, high-sensitivity C-reactive protein; FBG, fasting blood glucose; TC, total cholesterol.
* Values are expressed as mean ± sd. Independent sample t test and χ 2 test were used to analyse differences in demographic and biochemical characteristics.
Dietary fibre and antioxidant intakes of participants in no or minimal and mild-to-severe groups are reported in Table 2. No significant difference was observed between two groups regarding the amount of energy intake (P = 0·9). There were no significant differences in dietary intakes of vitamin A (P = 0·7), vitamin E (P = 0·7) and lycopene (P = 0·8) between groups. Subjects with no or minimal depression symptoms had significantly higher dietary intakes of α-carotene (P = 0·01), β-carotene (P = 0·006), lutein (P = 0·03) and vitamin C (P = 0·04) when compared with subjects with mild-to-severe depression symptoms. Soluble dietary fibre and insoluble dietary fibre intakes were also significantly higher in healthy adolescents compared with those with depression symptoms (P < 0·001, Table 2).
Table 2 Dietary fibre and antioxidant intakes of participants with no or minimal and mild-to-severe depression symptoms
* Values are expressed as mean ± sd. Independent sample t test was used to analyse differences in dietary fibre and antioxidant intakes.
Crude and adjusted OR for depression symptoms across quartiles of nutrients intakes are presented in Table 3. In all our multivariate analyses, the group who was in the first quartile of dietary fibre and antioxidant intake served as a reference group (Q2, Q3, Q4 v. Q1). Although we observed no significant association between depression severity and vitamin C intake in crude model (P trend = 0·18), significant associations were found after adjusting for age and energy intake (P trend = 0·04) and also after further adjusting for menstruation, family members, parental death, parental divorce, physical activity and BMI (P trend = 0·04). The multivariable-adjusted (model II) OR (95 % CI) for the lowest through the highest quartiles of β-carotene intake were 1·00 (reference), 0·91 (95 % CI 0·58, 1·42), 0·77 (95 % CI 0·50, 1·20) and 0·42 (95 % CI 0·26, 0·69) (P trend = 0·003). The multivariable-adjusted (model II) OR (95 % CI) for the lowest through the highest quartiles of α-carotene intake were 1·00 (reference), 0·69 (95 % CI 0·45, 1·07), 0·63 (95 % CI 0·41, 0·98) and 0·50 (95 % CI 0·31, 0·79) (P trend = 0·004). In multivariate-adjusted model 2, the OR (95 % CI) of depressive symptoms were 0·71 (95 % CI 0·44, 1·15), 0·51 (95 % CI 0·32, 0·82) and 0·42 (95 % CI 0·25, 0·68) for the highest v. lowest quartile of lutein, soluble dietary fibre and insoluble dietary fibre cereal intakes, respectively (Table 3).
Table 3 Crude and adjusted OR (with 95% CI) of dietary intake of antioxidants (expressed as quartiles, Q2, Q3, Q4 v. Q1) and elevated depressive symptoms among Iranian adolescent girls
* OR with 95% CI obtained from multinomial logistic regression analysis both crude and adjusted for potential confounders. Model I: Adjusted for age and energy intake. Model II: Additionally, adjusted for menstruation, family members, parental death, parental divorce, physical activity and BMI.
Discussion
Our results suggest that higher dietary intakes of vitamin C, but not vitamin E, were associated with lower depression symptoms among Iranian adolescent girls. Moreover, dietary intakes of some carotenoids (β-carotene, α-carotene and lutein) and dietary fibre intake (both soluble and insoluble) were inversely associated with the severity of depression symptoms. Overall, subjects with mild-to-severe depression symptoms had significantly lower intake of dietary antioxidants.
Several studies have investigated the association between dietary antioxidant intakes and depressive symptoms in different population groups(Reference Payne, Steck and George22,Reference Prohan, Amani and Nematpour35,Reference Tsai, Chang and Chi36) . In agreement with our findings, Prohan et al. found that young male university students with major depressive disorder consumed less dietary antioxidants such as vitamin C, β-carotene, lutein and zeaxanthin than healthy controls(Reference Prohan, Amani and Nematpour35). In another study on 278 elderly participants, Payne et al. observed that vitamin C, lutein and β-cryptoxanthin intakes were significantly lower among individuals with depression than in healthy participants. Additionally, fruit and vegetable consumption was lower in individuals with depression(Reference Payne, Steck and George22). Amr et al. conducted a 6-month, double-blind, placebo-controlled pilot trial and found vitamin C as an effective adjuvant agent in the treatment of paediatric major depressive disorder(Reference Amr, El-Mogy and Shams37). Moreover, adherence to a Mediterranean dietary pattern was found to be associated with better mental health, which was independently explained by dietary total antioxidant and fibre content(Reference Bonaccio, Di Castelnuovo and Bonanni21). Furthermore, Tsai and colleagues conducted a prospective cohort study among a sample of free-living elderly and found that more frequent consumption of vegetables was protective against depressive symptoms, which was suggested to be due to high antioxidant content(Reference Tsai, Chang and Chi36).
We found an inverse association between dietary intakes of β-carotene, α-carotene and lutein, but not lycopene, and depression symptoms. In this regard, several previous studies have investigated the relationship between dietary intake and plasma status of carotenoids and depressive symptomatology(Reference Ye, Scott and Falcon23,Reference Milaneschi, Bandinelli and Penninx38,Reference Beydoun, Beydoun and Boueiz39) . In line with our observations, Ye et al. observed an inverse association between intake of total carotenoids and depression score after adjustment for age, sex, education, smoking, BMI, energy intake, plasma cholesterol, vitamin B6, vitamin C and homocysteine. Interestingly, similar association was observed for plasma carotenoids(Reference Ye, Scott and Falcon23). Milaneschi et al. also found that low plasma concentrations of carotenoids were associated with depressive symptoms among a sample of women and men aged 65 years and older(Reference Milaneschi, Bandinelli and Penninx38). Similarly, total carotenoids (mainly β-carotene and lutein + zeaxanthins) in serum were found to be associated with less severe depression symptoms among community-dwelling US adults(Reference Beydoun, Beydoun and Boueiz39). Furthermore, Li and colleagues showed that, in a group of 6680 adults aged 17–39 years, a history of attempted suicide was associated with low levels of carotenoids and antioxidant vitamins(Reference Li and Zhang40).
Similar to our findings, several studies have found an inverse association between dietary intake of fibre and severity of depressive symptoms(Reference Xu, Li and Song24,Reference Miki, Eguchi and Kurotani41,Reference Woo, Lynn and Lau42) . In a study on 1977 Japanese workers between ages 19 and 69 years, Miki et al. found a significant inverse association between dietary fibre intake from vegetables and fruits and depressive symptoms. However, no association was found between depressive symptoms and dietary intake of total, soluble, insoluble and cereal fibre(Reference Miki, Eguchi and Kurotani41). The authors suggested that the stronger association for fibre from vegetable and fruit could be attributable to its more readily fermentable properties compared with fibre from cereal. However, since the participants were Japanese manufacturing workers, care must be taken before generalising the findings(Reference Miki, Eguchi and Kurotani41). In another study on 16 807 adults aged 20 years or older, it was revealed that dietary intakes of total fibre, fruit fibre and vegetable fibre were inversely associated with depressive symptoms(Reference Xu, Li and Song24). Moreover, a Chinese study, which was conducted in 3394 community-dwelling older adults showed that total fibre intake was inversely associated with depression score(Reference Woo, Lynn and Lau42). The exact mechanism linking dietary intake of fibre to depression is unclear, but several possibilities are proposed. First, this can be partly explained by the potential involvement of dietary fibre in the transportation of dietary antioxidants through the gastrointestinal tract(Reference Saura-Calixto19). Second, SCFA, which are the products of colonic bacterial fermentation of dietary fibre, may improve inflammation(Reference Vinolo, Rodrigues and Nachbar43), an underlying mechanism of depression(Reference Berk, Williams and Jacka44). Further, another possible mechanism may be related to the effects of fibre on gut microbiota and its subsequent impact on gut–brain axis for which there is emerging evidence of the importance in depression pathogenesis(Reference Foster and McVey Neufeld45,Reference Dinan and Cryan46) .
Our results revealed that adolescents with mild-to-severe depressive symptoms had significantly lower diastolic blood pressure compared with those with no or minimal depressive symptoms. A cross-sectional population-based study conducted in 60 799 men and women aged 20–89 years also showed that there is an association between low blood pressure and depression across all age and sex groups(Reference Hildrum, Mykletun and Stordal47). Further, in the 22-year follow-up of the Nord-Trøndelag Health Study in Norway, Hildrum et al. found that symptoms of depression and anxiety are associated with decrease in blood pressure(Reference Hildrum, Romild and Holmen48). Our results also showed that adolescents with mild-to-severe depressive symptoms were more likely to be passive smokers. Similarly, using data from the 2005–2006 National Health and Nutrition Examination Survey, it was found that second-hand smoke exposure was positively associated with depressive symptoms in never-smokers aged ≥ 20 years(Reference Bandiera, Arheart and Caban-Martinez49). Using cross-sectional data from a series of ten population surveys, Patten et al. reported that the pooled annual prevalence of major depressive disorder was 6·1 % in non-smokers with second-hand smoke exposure compared with 4·0 % in non-smokers without second-hand smoke exposure(Reference Patten, Williams and Lavorato50). These results provide support for public health measures aiming to reduce second-hand smoke exposure to protect the health of non-smokers.
The strengths of our study include a relatively large sample of adolescent girls, a group with a high prevalence of depressive disorders, and using a standardised tool for assessment of depression symptoms. We acknowledge the limitations in our study including: (a) inclusion of only female sex, (b) the use of self-administered tool instead of more accurate face-to-face interviews and (c) the fact that we have assessed both depression symptoms and dietary intakes at baseline and no longitudinal assessment was performed. Therefore, we cannot say whether having depressive symptoms leads to lower consumption of foods rich in antioxidants or low dietary intakes of antioxidants and fibre contributes to the emergence of depressive symptoms.
In conclusion, this study showed a significant inverse association between dietary intake of some antioxidants and depression symptoms among Iranian adolescent girls. Moreover, higher dietary intakes of both soluble and insoluble fibre were associated with less severe depression symptoms.
Acknowledgements
Acknowledgements: The authors acknowledge with grateful appreciation the kind assistance and financial support provided by Mashhad University of Medical Sciences (MUMS). Financial support: This work was supported by a grant provided by Mashhad University of Medical Sciences (MUMS). Conflict of interest: The authors have no conflicts of interest to declare Authorship: S.S.K. and M.G.M. made substantial contributions to the conception or design of the work; F.N.S., L.K. and A.Z. contributed to analysis and interpretation of data for the work; A.O., M.M.M.M. and S.A. contributed to drafting the article and revising it critically; A.N. and G.A.F. contributed to final approval of the version to be published. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving study participants were approved by the ethics committee of Mashhad University of Medical Sciences (MUMS). Written informed consent was obtained from all subjects.
