Determinants of oxidative stress in orthodox Christian women: A comparative analysis of monastic and non-monastic lifestyles

Abstract

Background:

Oxidative stress has been implicated in aging and metabolic dysfunction, modulated by diet, lifestyle, and body composition. While structured dietary practices such as intermittent fasting and plant-based regimens have demonstrated antioxidative benefits, the biological impact of long-term religious fasting remains underexplored. Orthodox Christian fasting, characterized by periodic abstention from animal products and alignment with circadian rhythms, offers a unique naturalistic model for examining redox adaptation in humans.

Objectives:

To identify the physiological and biological factors of oxidative stress markers in Orthodox Christian monastic women compared to age-matched controls, focusing on the roles of adiposity, insulin, vitamin D status, and age.

Design:

In this cross-sectional study, 52 Orthodox nuns and 56 women from the general population were assessed.

Methods:

Serum levels of total antioxidant capacity (TAC), reduced glutathione (GSH), and thiobarbituric acid reactive substances were measured. Anthropometric indices (body mass index, body fat percentage, visceral fat), fasting insulin, 25-hydroxyvitamin D (25(OH)D), and age were recorded. Statistical analyses included group comparisons, Spearman correlations, and multivariable linear regression models.

Results:

In the monastic group, body fat percentage (beta coefficient = 0.387, p = 0.003) and age (beta coefficient = 0.301, p = 0.014) were associated with TAC levels. Among controls, insulin positively correlated with GSH (ρ = 0.480, p < 0.001) and marginally inversely with TAC (ρ = −0.321, p = 0.060). No significant associations were found between 25(OH)D and oxidative markers in either group.

Conclusions:

Vitamin D and insulin levels were not significantly associated with oxidative stress markers in this cohort. These findings highlight the potential of long-term, culturally structured fasting to modulate redox homeostasis and suggest a complex interplay between age, adiposity, and antioxidant defenses. These findings should be interpreted within the context of a highly specific religious and cultural lifestyle and may not be generalizable to other populations. Further research is needed to elucidate underlying mechanisms and long-term clinical implications.

Keywords

oxidative stress religious fasting monastic lifestyle total antioxidant capacity glutathione TBARS Vitamin D insulin body fat visceral adiposity

Introduction

Oxidative stress, resulting from an imbalance between the generation of reactive oxygen species (ROS) and the body’s antioxidant defenses, is a well-established factor in the pathogenesis of cardiovascular disease, insulin resistance, and age-related disorders.^1–3 Dietary practices and lifestyle interventions that modulate oxidative stress are increasingly being studied as non-pharmacological strategies for enhancing metabolic health and delaying aging-related decline.⁴

Religious fasting, particularly within the Orthodox Christian tradition, represents a unique long-term dietary model characterized by periodic abstention from animal products, limited caloric intake, and simplified meal structure for over 180 days annually.⁵ In monastic populations, fasting is practiced consistently throughout adulthood, often accompanied by reduced exposure to environmental toxins, lower psychosocial stress, and enhanced physical routine, making them an ideal model to examine metabolic adaptation to sustained dietary restriction.^6,7

Prior evidence suggests that intermittent fasting and plant-based diets may improve antioxidant capacity and reduce oxidative biomarkers.^8,9 However, the mechanistic relationship between religious fasting and oxidative status in real-life human cohorts remains underexplored. Most studies to date have focused on short-term interventions or surrogate metabolic endpoints, with limited data on validated oxidative biomarkers such as total antioxidant capacity (TAC), reduced glutathione (GSH), and lipid peroxidation by-products (thiobarbituric acid reactive substance (TBARS)).^10–12

Body composition is an important determinant of oxidative balance. Adipose tissue, particularly visceral fat, is metabolically active and contributes to low-grade inflammation and excess ROS production.^13,14 Beyond adiposity-related oxidative mechanisms, additional metabolic and endocrine factors may also influence redox homeostasis. Physiological factors such as age and insulin resistance may independently modulate antioxidant defenses, either through mitochondrial dysfunction or impaired enzymatic detoxification pathways.^15,16 Vitamin D, via its nuclear receptor, has also been implicated in the transcriptional regulation of antioxidant enzymes such as glutathione peroxidase and superoxide dismutase. However, data regarding the impact of vitamin D status on oxidative biomarkers remain inconsistent, especially in populations with modified dietary patterns.^17,18 The interplay between vitamin D, insulin, and adiposity in influencing oxidative stress is still poorly understood in the context of long-term religious fasting. Given the inconsistent evidence regarding vitamin D and oxidative stress in observational studies, the present analysis focuses primarily on lifestyle-related and metabolic factors associated with redox balance.

We have recently described that both Orthodox religious fasting and intermittent fasting practiced by the general population, exert beneficial effects on oxidative equilibrium¹⁰; however, evidence on the potential bio-regulators in monastic and lay populations, remain scarce. Although previous studies, including our own,^10,11 have demonstrated differences in oxidative stress markers between fasting and non-fasting populations, the physiological factors associated with inter-individual variability in oxidative stress within long-term fasting populations remain poorly characterized.

To provide a comprehensive assessment of redox status, the present study focused on a panel of complementary oxidative stress biomarkers rather than a single indicator. GSH represents a central component of intracellular antioxidant defense and plays a key role in maintaining cellular redox homeostasis. TAC reflects the cumulative activity of circulating antioxidants in plasma and serves as an integrative measure of systemic antioxidant potential. In contrast, TBARS are widely used as an index of lipid peroxidation and oxidative damage to cell membranes. The combined evaluation of GSH, TAC, and TBARS enables simultaneous assessment of antioxidant defenses and oxidative injury, offering a broader characterization of redox balance than any individual biomarker alone. In this context, the present analysis focuses primarily on lifestyle-related and metabolic factors associated with redox balance and physiological predictors of oxidative status in a real-world model of prolonged Orthodox fasting. We hypothesized that women following the monastic lifestyle would exhibit improved oxidative profiles compared to controls, and that factors such as body composition, insulin, and 25-hydroxyvitamin D (25(OH)D) levels would differentially predict oxidative stress across groups.

Materials and methods

Study design and participants

This was a cross-sectional, observational study conducted between March and September 2024 in Northern Greece. A total of 108 adult women were enrolled, consisting of 52 Orthodox Christian nuns and 56 age-matched women from the general population, serving as controls. Inclusion criteria for both groups were: female sex, age 30–75 years, and absence of chronic systemic illness or current use of antioxidant supplements. Participants with known endocrine disorders (e.g., thyroid dysfunction, diabetes mellitus), autoimmune disease, or recent acute infection were excluded. Monastic participants were recruited from two female Orthodox convents in the region, all of whom had practiced the traditional fasting regimen continuously for more than 15 years. Control women were recruited through community outreach, matched for age but not for body mass index (BMI) or body composition, to reflect real-world differences.

Anthropometric and biochemical assessments

Both study groups underwent systematic evaluation of anthropometric parameters and laboratory indices, using uniform and validated protocols. Details regarding instruments, calibration, and protocols have been extensively documented in prior publications.^8–11 Briefly, body weight was measured to the nearest 10 g with a certified electronic scale (K-Tron P1-SR; Onrion LLC, Bergenfield, NJ, USA), with participants in light clothing and without footwear. BMI was derived as weight divided by squared height (kg/m²). Body composition metrics—including fat mass and percentage, visceral adiposity, skeletal muscle, lean mass, and total body water—were quantified using bioelectrical impedance analysis (BIA) via the SC-330 S model (Tanita Corp., Tokyo, Japan). Although body composition was assessed in both groups using BIA, complete and technically reliable estimates of visceral fat were not available for participants in the control group, due to technical reasons. Therefore, these parameters were not included in correlation analyses for controls.

Venous blood was collected after a 12-h overnight fast, and samples were stored at −20°C until analysis. All samples were centrifuged and frozen immediately, except whole blood samples, which were processed fresh. Biochemical assays included serum calcium, which was quantified with the COBAS8000 autoanalyzer (Roche Diagnostics, Munich, Germany), while parathyroid hormone (PTH) and 25(OH)D were measured using electrochemiluminescence immunoassays on the COBAS e 602 platform (Roche Diagnostics). The analytical performance [inter- and intra-assay Coefficient variations (CVs)] and clinical reference intervals were: Ca: 8.4–10.2 mg/dL (CVs: 0.8%–1.3% and 0.5%–1.3%), PTH: 15–65 pg/mL (1.6–6.9 pmol/L; CVs: 1.1%–2.0% and 2.5%–3.4%), and 25(OH)D: ⩾30 ng/mL (CVs: 2.2%–6.8% and 3.4%–13.1%). Insulin resistance was assessed via the HOMA-IR formula: fasting insulin (µU/mL) × fasting glucose (mmol/L) ÷ 22.5.¹⁹

Oxidative stress biomarkers

Measurement of GSH in erythrocytes

GSH levels were determined following the method previously described.¹¹ A volume of 400 µL of erythrocyte lysate was mixed with an equal volume of 5% trichloroacetic acid (TCA) and centrifuged (1500 g, 5 min, 5°C). The supernatant was further processed with TCA and incubated in the dark for 45 min. Absorbance was read at 412 nm, and concentrations were derived using the molar extinction coefficient for TNB (13.6 L/mmol/cm).

Total antioxidant capacity

Plasma TAC was assessed using the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay, as described.¹¹ Specifically, 20 µL of plasma was diluted in phosphate buffer (10 mM, pH 7.4), followed by the addition of 0.1 mM DPPH solution. After incubation for 1 h at room temperature in the dark, absorbance was measured at 520 nm. TAC values were expressed in terms of DPPH reduction equivalents.

Thiobarbituric acid reactive substances

Lipid peroxidation was estimated via TBARS using a modified protocol.¹¹ Each 100 µL plasma sample was mixed with Tris-HCl buffer and TCA, followed by the addition of sodium sulfate and thiobarbituric acid. After heating at 95°C for 45 min in a water bath, the samples were centrifuged and read at 530 nm. The MDA concentration was calculated using its known molar extinction coefficient (156,000 L/mol/cm).

Dietary patterns and physical activity

Orthodox nuns followed the Athonian type of fasting, as previously described.^5–9 In brief, this pattern includes an 8 h feeding window (approximately between 07:00 and 15:00), aligning with traditional monastic practice and long fasting intervals between meals. Dietary intake data from the 3-day food diaries were analyzed using standardized Greek food composition tables. Portion sizes were estimated using household measures and converted to grams. Energy and macronutrient intake were calculated as mean daily intake over the 3-day recording period. Quantitative analyses of energy, macro- and micronutrient intake were also included in the present study.

General population fasters adopted a 16:8 dietary pattern, with an 8 h feeding time frame from 11.00 to 19.00, without avoidance of animal products and particular sartorial or religious habits. Time-restricted eating in the control group involved no restrictions on food type or dietary composition, in contrast to the monastic dietary pattern, which combines long-term vegetarian fasting with structured meal timing. Finally, levels, frequency, and duration of physical activity, divided into light, moderate, and intense physical activity, were recorded for all participants as previously described.¹⁰

Ethical compliance

The study protocol adhered to the principles set forth in the Declaration of Helsinki. All participants provided written informed consent. Approval for the inclusion of the monastic cohort was obtained in writing from the central ecclesiastical authority following submission of the full study documentation 1 year prior to study initiation. The study protocol was approved by the Institutional Review Board of Aristotle University of Thessaloniki, Greece (Approval No. 25224/2019).

Statistical analysis

Data distribution was assessed using the Shapiro–Wilk test. Continuous variables are presented as mean ± standard deviation (SD) or median (interquartile range (IQR)), depending on normality. Categorical variables are shown as counts and percentages. Group comparisons were performed using Student’s t-test or Mann–Whitney U test for continuous variables, and chi-square test for categorical data. Age differences between the groups with light, moderate and intense physical activity were tested using one-way analysis of variance with Tukey post hoc test. The effect of level of physical activity on overall health markers was tested with analysis of covariance to control for age. Normality of distribution was tested with one sample Kolmogorov–Smirnov test. Multivariable linear regression models were constructed for each oxidative stress marker, adjusted for age, body fat percentage, visceral fat, and either insulin or 25(OH)D, depending on the model. Adjusted R² and standardized β-coefficients were reported. Given the sample size of the study, multivariable regression analyses were considered exploratory and aimed at identifying potential associations rather than building predictive models. The number of covariates included in each model was selected based on biological plausibility and prior literature.

Spearman’s rank correlation coefficients were calculated to examine bivariate associations between oxidative stress markers (TAC, GSH, TBARS) and predictors including 25(OH)D, insulin, age, and adiposity indices. Variables with normal distribution are presented as mean ± SD, whereas non-normally distributed variables are presented as median (IQR), based on Shapiro–Wilk normality testing.

This study used a convenience sample comprising all eligible participants who met the inclusion criteria during the recruitment period (N = 108; approximately n = 54 per group), reflecting the feasibility constraints of studying a unique monastic population. A post hoc power consideration (two-sided α = 0.05) indicates that the available sample provides 80% power to detect a correlation of approximately r = 0.27 in the total sample and r = 0.37 within each group. For multivariable linear regression models with three predictors, the study has 80% power to detect an overall R² of approximately 0.10 in the total sample and 0.18 within each group. Accordingly, multivariable analyses were interpreted as exploratory, and smaller effects may not have been detectable. A p-value <0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics v.25 (IBM Corp., Armonk, NY, USA). The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.²⁰

Results

Baseline characteristics

Table 1 illustrates anthropometric and demographic characteristics of the two groups. Comparison of nuns and lay women did not show differences in body fat (%), lean body mass (%) and waist circumference, as well as degrees of physical activity, with the exception of intense activity, in which lay women reported higher rates. Continuous variables are presented as mean ± SD or median (IQR), depending on their distribution. Correlations involving body fat percentage and visceral fat could not be evaluated in the control group due to incomplete availability of these measurements.

Table 1.

Demographics of orthodox nuns compared to lay women.

Demographics	Orthodox nuns (n = 52)	Lay women (n = 56)	p
Age (years)	42 (36–50)	38 (34–42)	0.45
Weight (kg)	71.5 (64–82)	66 (60–87)	0.31
BMI (kg/m²)	27.0 (24.2–29.0)	26.8 (22.0–32.0)	0.19
Body fat (%)	24.5 ± 9.4	22.1 ± 8.1	0.23
Lean body mass (%)	39.9 ± 6.3	41.2 ± 7.1	0.15
Waist circumference (cm)	92.4	89.1	0.11
Physical activity
Light (n)	N = 11	N = 13	0.31
Moderate (n)	N = 27	N = 25	0.48
Intense (n)	N = 14	N = 18	0.03
Years of monasticism	10.5 ± 9.8	—	—
Deaconship of Orthodox nuns	Baker (3); Botanist (2); Cook (5); Cooking assistant (5); Dining assistant (5); Ecclesiastical chanter (6); Gardener (3); Housekeeper (3); Iconographer (6); Laundry assistant (4); Pharmacist (2);

BMI: body mass index.

Predictors of stress markers

A total of 108 female participants were included in the analysis, comprising 52 women living in monastic communities and 56 from the general population. Multiple linear regression models were used to evaluate the influence of metabolic and demographic variables on oxidative stress biomarkers (TAC, GSH, TBARS) in each group. Figures 1 and 2 present dot-whisker plots of the standardized regression coefficients along with their 95% confidence intervals, separately for the monastic (Figure 1) and control (Figure 2) groups. Significant predictors of TAC included body fat percentage and age for Orthodox nuns. No statistically significant predictors were identified for the general population group.

Figure 1.

Dot-whisker plot showing standardized regression coefficients and 95% confidence intervals for predictors of oxidative stress markers in the monastic group.

Figure 2.

Dot-whisker plot showing standardized regression coefficients and 95% confidence intervals for predictors of oxidative stress markers in the control group.

Nutritional habits and oxidative stress markers

Comparison of the nutritional habits of Orthodox nuns to those of lay women is shown in Table 2. Lay women consumed higher amounts of carbohydrates (g; 194.3 ± 23.4 versus 159.6 ± 21.8) and total and saturated fat (24.4 ± 0.6 versus 21.0 ± 0.1 and 16.4 ± 0.0 and 12.7 ± 0.0, respectively), whereas Orthodox nuns reported higher amounts of protein and fiber intake (36.1 ± 0.8 versus 24.2 ± 0.8).

Table 2.

Nutritional habits and oxidative stress markers comparison between Orthodox nuns and lay women.

Variable	Orthodox nuns	Lay women	p
Energy (kcal)	1565.9 ± 64.5	1890.0 ± 71.0	<0.01
Carbohydrates (g)	159.6 ± 21.8	194.3 ± 23.4	0.03
Protein (g)	89.2 ± 1.3	72.3 ± 1.3	0.04
Daily fat intake (g)	21.0 ± 0.1	24.4 ± 0.6	0.02
Daily saturated fat intake (g)	12.7 ± 0.0	16.4 ± 0.0	0.01
Total fiber intake (g)	36.1 ± 0.8	24.2 ± 0.8	0.02
25-hydroxy-vitamin D3 (ng/mL)	15.7 (11.4–19.8)	26.1 (18.2–31.9)	0.02
PTH (pg/mL)	45.6 (39.6–54.7)	19.4 (13.1–28.5)	<0.001
Calcium (mg/dL)	9.4 (9.1–9.7)	9.1 (8.8–9.3)	0.15
Insulin (IU/L)	5.3 (3.4–6.7)	7.1 (4.7–11)	0.02
Fasting glucose (mg/dL)	84.4 ± 10.1	89.2 ± 9.7	0.43
HOMA-IR	1.02 ± 0.4	1.26 ± 0.7	0.21
Oxidative status
TAC	0.93 (0.87–0.99)	0.77 (0.65–0.90)	<0.001
GSH	6.0 (4.4–6.8)	7.2 (5.5–8.8)	0.04
TBARS	7.3 (5.8–8.3)	7.6 (6.9–8.4)	0.28

TAC: total antioxidant capacity; GSH: reduced glutathione; TBARS: thiobarbituric acid reactive substance; PTH: parathyroid hormone.

Correlations between oxidative stress markers and metabolic parameters

Correlation analyses are presented in Tables 3 and 4. In the monastic group, TAC showed statistically significant positive correlations with body fat percentage (ρ = 0.387, p = 0.041), visceral fat (ρ = 0.317, p = 0.031), and age (ρ = 0.301, p = 0.014), but no association with 25(OH)D levels or insulin. Similarly, GSH levels were positively correlated with body fat percentage (ρ = 0.387, p = 0.041), visceral fat (ρ = 0.317, p = 0.031), and age (ρ = 0.301, p = 0.014) in the monastic group, mirroring the pattern observed for TAC. No significant association was observed between GSH and either 25(OH)D (ρ = 0.026, p = 0.729) or insulin (ρ = −0.112, p = 0.321).

Table 3.

Spearman’s rho (ρ) coefficients and corresponding p-values illustrating the strength and direction of monotonic correlations between key oxidative stress markers (TAC, GSH, TBARS) and metabolic parameters (25(OH)D levels, body fat percentage, visceral adiposity, and age) in the monastic cohort.

Marker	Variable	Spearman’s rho (ρ)	p-value
TAC	25(OH)D	0.074	0.331
TAC	Body fat %	0.387	0.041
TAC	Visceral fat	0.317	0.031
TAC	Age	0.301	0.014
GSH	25(OH)D	0.026	0.729
GSH	Body fat %	0.387	0.041
GSH	Visceral fat	0.317	0.031
GSH	Age	0.301	0.014
TBARS	25(OH)D	−0.038	0.617
TBARS	Body fat %	N/A	0.061
TBARS	Visceral fat	N/A	N/A
TBARS	Age	N/A	N/A

TAC: total antioxidant capacity; GSH: reduced glutathione; TBARS: thiobarbituric acid reactive substance; 25(OH)D: 25-hydroxyvitamin D.

Table 4.

Spearman’s rho (ρ) and p-values for correlations between oxidative stress markers and metabolic variables in the control group.

Marker	Variable	Spearman’s rho (ρ)	p-value
TAC	25(OH)D	−0.293	0.0875
TAC	Insulin	−0.321	0.0598
TAC	Body fat %	N/A	N/A
TAC	Visceral fat	N/A	N/A
TAC	Age	−0.629	0.0001
GSH	25(OH)D	0.212	0.2225
GSH	Insulin	0.480	0.0035
GSH	Body fat %	N/A	N/A
GSH	Visceral fat	N/A	N/A
GSH	Age	N/A	N/A
TBARS	25(OH)D	−0.120	0.110
TBARS	Insulin	0.209	0.229
TBARS	Body fat %	N/A	N/A
TBARS	Visceral fat	N/A	N/A
TBARS	Age	N/A	N/A

TAC: total antioxidant capacity; GSH: reduced glutathione; TBARS: thiobarbituric acid reactive substance; 25(OH)D: 25-hydroxyvitamin D.

In contrast, TBARS did not demonstrate any significant correlations with any of the variables assessed (e.g., 25(OH)D: ρ = −0.038, p = 0.617; insulin: ρ = 0.041, p = 0.676). In the control group, the correlation pattern was somewhat different. A significant negative association was observed between TAC and age (ρ = −0.629, p < 0.001), suggesting that antioxidant defenses decline with advancing age in non-fasting women. GSH showed a moderate but significant positive correlation with insulin levels (ρ = 0.480, p < 0.001). No other statistically significant associations were observed for GSH or TBARS, and correlations with body fat percentage and visceral fat could not be assessed due to the unavailability of these measurements in the control dataset.

Table 4 shows Spearman’s rho (ρ) and p-values for correlations between oxidative stress markers and metabolic variables in the control group. The table is structured identically to Table 1 for comparability. N/A indicates variables for which complete data were not available in the control group. Significant correlations included a negative association between TAC and age, and a positive correlation between GSH and insulin levels. The absence of complete body composition data for all control participants limits direct comparability of multivariable analyses between groups and should be considered when interpreting group-specific associations.

Regression analysis of oxidative stress markers

Tables 5 and 6 present the fit indices (R² and adjusted R²) from multiple linear regression models assessing whether age, BMI, and serum 25(OH)D levels predict variability in oxidative stress markers (TAC, GSH, TBARS) in the monastic and control groups. For each outcome, two models were evaluated: a basic model including age and BMI, and an extended model with the addition of serum 25(OH)D.

Table 5.

Multiple linear regression results evaluating the contribution of age, BMI, and serum 25(OH)D levels to oxidative stress markers (TAC, GSH, TBARS) in the monastic group.

Outcome	Model	R ²	Adj R²
TAC (monastics)	Age + BMI	0.009	−0.046
TAC + 25(OH)D	Extended	0.019	−0.067
GSH (monastics)	Age + BMI	0.02	−0.035
GSH + 25(OH)D	Extended	0.023	−0.063
TBARS (monastics)	Age + BMI	0.005	−0.051
TBARS + 25(OH)D	Extended	0.024	−0.062

TAC: total antioxidant capacity; GSH: reduced glutathione; TBARS: thiobarbituric acid reactive substance; BMI: body mass index; 25(OH)D: 25-hydroxyvitamin D.

Table 6.

Multiple linear regression results evaluating the contribution of age, BMI, and serum 25(OH)D levels to oxidative stress markers (TAC, GSH, TBARS) in the control group.

Outcome	Model	R ²	Adj R²
TAC (controls)	Age + BMI + fat + VF	0.019	−0.099
TAC + 25(OH)D	Extended	0.026	−0.126
GSH (controls)	Age + BMI + fat + VF	0.157	0.055
GSH + 25(OH)D	Extended	0.157	0.026
TBARS (controls)	Age + BMI + fat + VF	0.101	−0.007

TAC: total antioxidant capacity; GSH: reduced glutathione; TBARS: thiobarbituric acid reactive substance; BMI: body mass index; 25(OH)D: 25-hydroxyvitamin D.

In the monastic group (Table 4), none of the models demonstrated meaningful predictive power. All adjusted R² values were negative or close to zero, indicating that neither age, BMI, nor vitamin D levels could adequately explain the variability in oxidative status among fasting individuals. For example, the extended model for TAC yielded R² = 0.019 and Adj R² = −0.067, while the GSH model showed similarly low values (R² = 0.023, Adj R² = −0.063). In contrast, the control group (Table 5) showed a more nuanced pattern. The GSH model that included age, BMI, body fat percentage, and visceral fat explained a modest proportion of variance (R² = 0.157, adj R² = 0.055), but this was not improved by the inclusion of vitamin D concentrations. Models for TAC and TBARS in the control group, remained weak, with negative adjusted R² values (e.g., TAC: R² = 0.019, adj R² = −0.099), indicating poor fit (Table 6). Overall, the low and occasionally negative adjusted R² values indicate limited explanatory power of the models, suggesting that a substantial proportion of variability in oxidative stress markers remains unexplained.

Table 5 shows multiple linear regression results evaluating the contribution of age, BMI, and serum 25(OH)D levels to oxidative stress markers (TAC, GSH, TBARS) in the monastic group. Two models were constructed for each outcome: a basic model including age and BMI, and an extended model that additionally incorporated 25(OH)D concentrations. In all cases, the explained variance was low, and adjusted R² values were negative or near-zero, indicating poor model fit.

Table 6 shows multiple linear regression results evaluating the contribution of age, BMI, and serum 25(OH)D levels to oxidative stress markers (TAC, GSH, TBARS) in the control group. Two models were tested for each oxidative outcome: a base model including age and BMI, and an extended model that additionally incorporated 25(OH)D levels. While model fit was poor for TAC and TBARS (with negative adjusted R² values), the model predicting GSH showed modest explanatory power (R² = 0.157, adj R² = 0.055), which was not improved with the addition of vitamin D.

Bivariate associations between oxidative stress and metabolic variables

Figures 3 –5 illustrate selected bivariate relationships between oxidative stress markers and key metabolic indicators, based on Spearman correlation analysis. In the monastic group (Figure 3), a significant inverse correlation was observed between serum 25(OH)D concentrations and body fat percentage (ρ = −0.178, p = 0.018). In the control group, Figure 4 demonstrated a statistically significant positive association between insulin and GSH levels (ρ = 0.480, p = 0.0035). Figure 5 showed a borderline inverse correlation between insulin and TAC (ρ = −0.321, p = 0.060).

Figure 3.

Scatterplot illustrating a negative correlation between serum 25(OH)D levels and body fat percentage in the monastic group (ρ = −0.178, p = 0.018). The solid line represents the line of best fit based on Spearman correlation analysis.

Figure 4.

Scatterplot showing a significant positive correlation between fasting insulin levels and GSH in the control group (ρ = 0.480, p = 0.0035). The solid line represents the line of best fit based on Spearman correlation analysis.

Figure 5.

Scatterplot depicting a borderline inverse correlation between fasting insulin and TAC in the control group (ρ = −0.321, p = 0.060). The solid line represents the line of best fit based on Spearman correlation analysis.

Discussion

This study examined oxidative stress status and associated factors in Orthodox Christian women following long-term religious fasting, in comparison to age-matched women from the general population. The present comparison distinguishes the effects of a comprehensive monastic dietary–lifestyle pattern, including vegetarian fasting and chrononutrition, from chrononutrition alone as practiced by the control group.

We assessed three validated markers of redox balance—TAC, GSH, and lipid peroxidation (TBARS)—and explored their associations with age, insulin, 25(OH)D, and adiposity indices. Among the monastic participants, body fat percentage and age were associated with TAC, suggesting an adaptive antioxidant response potentially shaped by lifestyle and dietary patterns. In contrast, insulin and 25(OH)D were not associated with oxidative status in either group. No statistically significant associations were observed between 25(OH)D concentrations and oxidative stress markers in either group, indicating a lack of detectable relationship in the present study. Given the cross-sectional design, the observed associations should be interpreted as descriptive rather than mechanistic. Οf major interest, the study population consists of Orthodox Christian monastic women, representing a highly specific cultural and lifestyle context; therefore, the generalizability of these findings to other populations is limited.

Our previous findings demonstrated that the monastic group exhibited significantly higher TAC and lower TBARS concentrations than controls, indicating a more favorable oxidative profile.¹¹ These differences are consistent with prior research suggesting that prolonged adherence to plant-based or calorically restricted diets may enhance endogenous antioxidant defenses and attenuate lipid peroxidation.^17,18 The fasting regimen observed by Orthodox monastics includes extended abstention from animal-derived foods, reduced caloric density, and natural circadian dietary rhythm—all factors implicated in oxidative modulation.²¹

Interestingly, no significant differences were observed between groups in GSH levels or 25(OH)D status. The generally low explanatory power of the multivariable models suggests that the examined variables account for only a small proportion of the variability in oxidative stress markers. This finding indicates that additional unmeasured factors, including dietary composition, inflammatory status, genetic variability, or other lifestyle components, may play a more prominent role in determining oxidative status. This association should be interpreted cautiously, as the present study does not include mechanistic biomarkers that would allow inference regarding underlying biological pathways. The positive association between body fat percentage and TAC observed in the monastic group indicates a relationship between adiposity and systemic antioxidant capacity, without implying an underlying compensatory or adaptive mechanism.

This may reflect the combined influence of sunlight exposure, vitamin D intake variability, and antioxidant homeostasis mechanisms beyond glutathione recycling.²² Despite prior hypotheses that 25(OH)D enhances redox control via nuclear factor erythroid 2–related factor 2-dependent pathways,²² our analysis did not identify vitamin D as an independent predictor of any oxidative stress marker in either group. This aligns with a growing body of literature showing conflicting associations between serum 25(OH)D and redox biomarkers in humans.^23,24 In multivariate models, we identified body fat percentage and age as significant positive predictors of TAC among monastics, with an adjusted R² of 0.28. This observation appears paradoxical, as increased adiposity is generally associated with greater oxidative burden.^25,26 However, the specific composition and functionality of adipose tissue in long-term fasters may differ metabolically.

Prior work has shown that non-obese individuals adhering to Mediterranean or ascetic dietary patterns manifest an improvement of antioxidant systems, although such mechanisms have been proposed in experimental and interventional studies, the present observational data do not allow mechanistic inference.¹⁰ Furthermore, mild lipid peroxidation may serve as a hormetic stimulus for antioxidant enzyme production,²⁷ potentially explaining the enhanced TAC in older, slightly more adipose monastics. In contrast, in the control group, insulin levels were significantly positively correlated with GSH and marginally inversely correlated with TAC. These findings reflect the known dual role of insulin in oxidative regulation. On one hand, insulin resistance promotes ROS generation through NADPH oxidase activation and mitochondrial overload.²⁸ On the other hand, insulin also upregulates the expression of glutathione synthetase and supports cysteine availability,²⁹ which may explain the positive correlation between insulin and GSH observed in our study. Importantly, these associations were not retained in adjusted models, suggesting that insulin’s redox effects are confounded by underlying adiposity or age.

The absence of strong associations between oxidative stress markers and vitamin D levels in both groups further supports the notion that vitamin D may not act as a consistent predictor of redox status across populations. While in vitro studies suggest that vitamin D upregulates glutathione synthesis and inhibits pro-oxidant pathways,²⁹ clinical trials have yielded inconsistent results.³⁰ A recent meta-analysis found only marginal effects of vitamin D supplementation on TAC and GSH, and no effect on TBARS.³¹ Taken together, our results suggest that long-term adherence to Orthodox fasting may promote a more balanced oxidative status, independent of vitamin D or insulin levels. The role of body composition appears to be more complex than previously assumed, potentially modulated by dietary quality, inflammation, and mitochondrial function.

This highlights the need to go beyond simple BMI metrics in oxidative stress research and consider metabolic phenotyping of adipose tissue. Another consideration is the physiological role of TAC as an integrative measure of both enzymatic and non-enzymatic antioxidant defense. While elevated TAC is typically interpreted as beneficial, it may also reflect compensatory responses to subtle redox imbalances or subclinical inflammation.³² This nuance is particularly relevant in monastic populations, where unique environmental exposures and long-term dietary patterns may upregulate antioxidant enzymes or increase the availability of dietary polyphenols and endogenous scavengers. Indeed, studies have demonstrated that plant-based fasting regimens are associated with increased plasma levels of polyphenols, vitamins C and E, and selenium—micronutrients contributing significantly to TAC.³³ Future studies should assess dietary antioxidant intake and relate it quantitatively to serum antioxidant potential. The absence of group differences in GSH concentrations contrasts with findings in studies of Ramadan fasting or intermittent energy restriction, where GSH levels often increase post-intervention.^34,35 One explanation may be that GSH homeostasis is tightly regulated under stable metabolic conditions, as seen in monastics, and may not fluctuate in response to chronic but balanced dietary restrictions. Moreover, GSH levels are sensitive to intracellular redox status, hepatic function, and sulfur amino acid availability—all of which may differ subtly between individuals despite similar body composition or insulin levels.³⁶

The borderline inverse correlation between insulin and TAC observed in the control group also warrants further discussion. This relationship may indicate early oxidative dysregulation in the context of mild insulin resistance. While not statistically significant after adjustment, such trends support existing literature linking hyperinsulinemia to impaired redox signaling, via inhibition of AMP-activated protein kinase and induction of ROS-producing enzymes.^37,38 It is plausible that in controls with higher visceral adiposity and lower metabolic flexibility, elevated insulin serves as a marker of systemic oxidative load. Furthermore, although visceral fat did not emerge as an independent predictor of oxidative markers in our models, it remains a well-established source of pro-inflammatory cytokines and oxidative mediators.³⁹ The lack of association in our cohort may reflect limited variance in visceral fat across groups or insufficient statistical power. Alternatively, the influence of visceral adiposity may be masked by overriding factors such as age and overall dietary pattern.

The observation that different parameters act as predictors in monastic versus non-monastic women supports the notion that lifestyle and environmental context play a critical role in shaping metabolic and biochemical outcomes. Monastic women adhere to a consistent pattern of religious fasting, with reduced caloric intake, limited sun exposure, and a structured daily routine, all of which may amplify the influence of factors such as body fat percentage or other factors related to oxidative status.

In contrast, non-monastic women exhibit greater variability in diet, activity, and environmental exposures, making parameters like age, seasonality, or dietary intake more relevant predictors. This divergence suggests the presence of effect modification, whereby the same outcome is driven by distinct mechanisms depending on the lived context. Therefore, interpreting predictors within each group requires careful consideration of the broader lifestyle framework.

From a public health perspective, the findings of this study lend support to the hypothesis that sustained religious fasting, when practiced in a structured and culturally integrated way, may confer antioxidant benefits. Unlike short-term diets or commercial detox regimens, Orthodox fasting is embedded in the rhythm of daily life and practiced across decades. The biological implications of such long-term behavioral regularity—especially in relation to redox adaptation, metabolic resilience, and healthy aging—deserve further longitudinal investigation.

Finally, it is noteworthy that our findings parallel results from studies on calorie restriction and Mediterranean-style diets, both of which are characterized by low animal protein intake, high fiber, and an emphasis on plant-derived antioxidants.⁴⁰ Whether the benefits observed in monastic women stem primarily from caloric content, macronutrient balance, timing of meals, or psychosocial harmony remains to be clarified. Nonetheless, these data suggest that the monastic model may serve as a naturalistic framework for exploring sustainable dietary strategies to improve redox balance and possibly reduce long-term cardiometabolic risk.

Our study has several strengths, including the use of validated oxidative biomarkers, detailed body composition analysis, and comparison of a relatively homogeneous monastic population with real-world controls. To our knowledge, this is the first study to evaluate predictors of oxidative stress in Orthodox nuns using both bivariate and multivariate approaches.

Limitations

Several limitations should be acknowledged. First, the cross-sectional design precludes causal inference. Second, dietary intake data were not quantified, preventing nutrient-level correlations. In the absence of quantitative dietary intake data, the observed differences in oxidative stress markers should be interpreted as reflecting overall lifestyle patterns rather than the isolated effect of fasting itself. Therefore, it is not possible to disentangle the specific contribution of fasting from other dietary components, such as fiber or polyphenol intake. Although dietary patterns were documented, the lack of detailed nutrient and antioxidant intake data limits the ability to disentangle the potential contribution of specific dietary components to oxidative stress markers.

Third, the sample size, while adequate for primary comparisons, limits the detection of smaller interaction effects. Finally, we did not assess gene expression or enzymatic activity of antioxidant systems, which could offer mechanistic insights. Although the sample size was sufficient for primary comparisons, it limited the ability to support complex multivariable modeling with multiple predictors, and therefore, regression results should be interpreted with caution. The absence of complete body composition data for all control participants limits direct comparability of multivariable analyses between groups and should be considered when interpreting group-specific associations. The modest sample size, particularly for group-stratified multivariable models, limits the ability to detect small associations and contributes to limited model fit in some analyses. Therefore, regression findings should be interpreted as exploratory, and unmeasured factors may account for a substantial proportion of variability in oxidative stress markers. The sample size was determined by feasibility and represented all available eligible participants from this hard-to-recruit population.

Finally, this study did not include mechanistic biomarkers such as antioxidant enzyme activity or inflammatory markers, which limits insight into the biological pathways underlying the observed associations. The absence of direct measures of antioxidant enzyme activity or inflammatory markers limits the mechanistic interpretation of the observed associations. Future studies incorporating inflammatory and enzymatic antioxidant markers are required to elucidate the biological mechanisms underlying these associations.

Conclusions

These findings underscore the complexity of redox regulation in humans and support the potential health-promoting role of structured, culturally embedded fasting practices. No significant associations were identified between vitamin D status or insulin levels and oxidative stress markers. Future longitudinal and mechanistic studies are warranted to explore the causal links between fasting, adiposity, and oxidative resilience. The present findings apply specifically to women adhering to long-term Orthodox monastic lifestyle practices and should not be extrapolated to other populations without further study. While both groups shared elements of meal timing, they differed substantially in dietary composition and lifestyle context, which should be considered when interpreting the observed associations.

Future studies should aim to evaluate longitudinal changes in oxidative status in fasting populations, incorporate dietary assessment tools, and include functional assays such as glutathione peroxidase or SOD activity. The role of microbiota, circadian eating patterns, and physical activity in modulating redox balance during fasting also warrants exploration.

Footnotes

Acknowledgements

The authors thank all participants for their contribution to this study.

ORCID iD

Fatme Al Anouti

Ethical Considerations

The study was approved by the appropriate institutional ethics committee the study was conducted in accordance with the Declaration of Helsinki, and approved by the Aristotle University of Thessaloniki (approval number 25224/2019, approval date 14 August 2019).

Consent to participate

All participants provided written informed consent.

Consent for publication

Informed consent was obtained from all subjects involved in the study.

Author contributions

Spyridon N. Karras: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Software; Formal analysis; Project administration; Data curation; Supervision; Resources.

Konstantinos Michalakis: Writing – original draft; Writing – review & editing.

Fatme Al Anouti: Writing – original draft; Writing – review & editing.

Maria Kypraiou: Writing – original draft; Writing – review & editing.

Antonios Vlastos: Writing – review & editing; Writing – original draft.

Marios Anemoulis: Writing – original draft; Writing – review & editing.

Fotios Tekos: Writing – review & editing; Writing – original draft.

Zoi Skaperda: Writing – original draft; Writing – review & editing.

Periklis Vardakas: Writing –review & editing; Writing – original draft.

Costas Haitoglou: Writing – original draft; Writing – review & editing.

Evangelos G. Papanikolaou: Writing – original draft; Writing – review & editing.

Neoklis Georgopoulos: Writing – original draft; Writing – review & editing.

Georgios Tzimagiorgis: Writing – original draft; Writing – review & editing.

Demetrios Kouretas: Writing – original draft; Writing – review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

Trepanowski

Bloomer

. The impact of religious fasting on human health. Nutr J 2010; 9: 57. https://doi.org/10.1186/1475-2891-9-57

Sarri

Linardakis

Bervanaki

, et al. Greek Orthodox fasting rituals: a hidden characteristic of the Mediterranean diet of Crete. Br J Nutr 2004; 92: 277–284. https://doi.org/10.1079/BJN20041197

Sarri

Tzanakis

Linardakis

, et al. Effects of Greek Orthodox Christian Church fasting on serum lipids and obesity. BMC Public Health 2003; 3: 16. https://doi.org/10.1186/1471-2458-3-16

Sarri

Linardakis

Codrington

, et al. Does the periodic vegetarianism of Greek Orthodox Christians benefit blood pressure? Prev Med 2007; 44: 341–348. https://doi.org/10.1016/j.ypmed.2006.11.009

Karras

Koufakis

Adamidou

, et al. Effects of orthodox religious fasting versus combined energy and time restricted eating on body weight, lipid concentrations and glycaemic profile. Int J Food Sci Nutr 2021; 72: 82–92. https://doi.org/10.1080/09637486.2020.1760218

Mehta

Roth

. Caloric restriction and longevity: the science and the ascetic experience. Ann N Y Acad Sci 2009; 1172: 28–33.

Papadaki

Vardavas

Hatzis

, et al. Calcium, nutrient and food intake of Greek Orthodox Christian monks during a fasting and non-fasting week. Public Health Nutr 2008; 11(10): 1022–1029.

Wilhelmi

Toledo

Grundler

Goutzourelas

, et al. Influence of long-term fasting on blood redox status in humans. Antioxidants (Basel) 2020; 9(6): 496.

Romeu

Aranda

Giralt

, et al. Diet, iron biomarkers and oxidative stress in a representative sample of Mediterranean population. Nutr J 2013; 12: 102.

10.

Karras

Michalakis

Tekos

, et al. Effects of religious fasting on markers of oxidative status in vitamin D-deficient and overweight Orthodox nuns versus implementation of time-restricted eating in lay women from Central and Northern Greece. Nutrients 2024; 16(19): 3300.

11.

Deng

Wei

, et al. Efficacy of antioxidant supplementation in improving endocrine, hormonal, inflammatory, and metabolic statuses of PCOS: a meta-analysis and systematic review. Food Funct 2024; 15(4): 1779–1802.

12.

Colamatteo

Fusco

Matarese

, et al. Obesity and autoimmunity epidemic: the role of immunometabolism. Annu Rev Nutr 2025; 45(1): 115–140.

13.

Ježek

. Physiological fatty acid-stimulated insulin secretion and redox signaling versus lipotoxicity. Antioxid Redox Signal 2025; 42(10–12): 566–622.

14.

Iheagwam

Joseph

Adedoyin

, et al. Mitochondrial dysfunction in diabetes: shedding light on a widespread oversight. Pathophysiology 2025; 32(1): 9.

15.

Karras

Michalakis

Katsiki

, et al. Interrelations of leptin and interleukin-6 in vitamin D deficient and overweight Orthodox nuns from Northern Greece: a pilot study. Nutrients 2025; 17(7): 1144.

16.

Hertiš Petek

Homšak

Svetej

, et al. Metabolic syndrome, inflammation, oxidative stress, and vitamin D levels in children and adolescents with obesity. Int J Mol Sci 2024; 25(19): 10599.

17.

Mattson

. Roles of the lipid peroxidation product 4-hydroxynonenal in obesity, the metabolic syndrome, and associated vascular and neurodegenerative disorders. Exp Gerontol 2009; 44(10): 625–633.

18.

. Membrane alteration as a basis of aging and the protective effects of calorie restriction. Mech Ageing Dev 2005; 126(9): 1003–1010.

19.

Matthews

Hosker

Rudenski

, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28(7): 412–419.

20.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ 2007; 335(7624): 806–808.

21.

Licker

Ellenberger

. Impact of the circadian rhythm and seasonal changes on the outcome of cardiovascular interventions. J Clin Med 2025; 14(8): 2570.

22.

, et al. Protective effects of 1,25 dihydroxyvitamin D3 against high-glucose-induced damage in human umbilical vein endothelial cells involve activation of Nrf2 antioxidant signaling. J Vasc Res 2021; 58(4): 267–276.

23.

Asghari

Hamedi-Shahraki

Amirkhizi

. Vitamin D status and systemic redox biomarkers in adults with obesity. Clin Nutr ESPEN 2021; 45: 292–298.

24.

Alvarez

Chowdhury

Jones

, et al. Vitamin D status is independently associated with plasma glutathione and cysteine thiol/disulphide redox status in adults. Clin Endocrinol (Oxf) 2014; 81(3): 458–466.

25.

McLeod

Datta

Fuller

. Adipokines as cardioprotective factors: BAT steps up to the plate. Biomedicines 2025; 13(3): 710.

26.

Blasco-Roset

Quesada-López

Mestres-Arenas

, et al. Acyl CoA-binding protein in brown adipose tissue acts as a negative regulator of adaptive thermogenesis. Mol Metab 2025; 96: 102153.

27.

Zhang

Humes

Almond

, et al. A mitohormetic response to pro-oxidant exposure in the house mouse. Am J Physiol Regul Integr Comp Physiol 2018; 314(1): R122–R134.

28.

Zavarzadeh

Panchal

Bishop

, et al. Exploring proinsulin proteostasis: insights into beta cell health and diabetes. Front Mol Biosci 2025; 12: 1554717.

29.

Pawlukianiec

Lauko

Michalak

, et al. A comparative study on the antioxidant and antiglycation properties of different vitamin D forms. Eur J Med Chem 2025; 285: 17263.

30.

Della Nera

Sabatino

Gaggini

, et al. Vitamin D determinants, status, and antioxidant/anti-inflammatory-related effects in cardiovascular risk and disease: not the last word in the controversy. Antioxidants (Basel) 2023; 12(4): 948.

31.

Moslemi

Musazadeh

Kavyani

, et al. Efficacy of vitamin D supplementation as an adjunct therapy for improving inflammatory and oxidative stress biomarkers: an umbrella meta-analysis. Pharmacol Res 2022; 186: 106484.

32.

Byrne

Soni

Takahara

, et al. Chronically elevating circulating ketones can reduce cardiac inflammation and blunt the development of heart failure. Circ Heart Fail 2020; 13(6): e006573.

33.

Karras

Koufakis

Adamidou

, et al. Different patterns of changes in free 25-hydroxyvitamin D concentrations during intermittent fasting among meat eaters and non-meat eaters and correlations with amino acid intake. Int J Food Sci Nutr 2023; 74(2): 257–267.

34.

Al-Shafei

. Ramadan fasting ameliorates oxidative stress and improves glycemic control and lipid profile in diabetic patients. Eur J Nutr 2014; 53(7): 1475–1481.

35.

Al-Shafei

. Ramadan fasting ameliorates arterial pulse pressure and lipid profile, and alleviates oxidative stress in hypertensive patients. Blood Press 2014; 23(3): 160–167.

36.

Lapenna

. Glutathione and glutathione-dependent enzymes: from biochemistry to gerontology and successful aging. Ageing Res Rev 2023; 92: 102066.

37.

Peifer-Weiß

Al-Hasani

Chadt

. AMPK and beyond: the signaling network controlling RabGAPs and contraction-mediated glucose uptake in skeletal muscle. Int J Mol Sci 2024; 25(3): 1910.

38.

Weinberg Sibony

Segev

Dor

, et al. Overview of oxidative stress and inflammation in diabetes. J Diabetes 2024; 16(10): e70014.

39.

Jain

Srivastava

Sharma

, et al. Impaired fibroblast growth factor 21 (FGF21) associated with visceral adiposity leads to insulin resistance: the core defect in diabetes mellitus. Curr Diabetes Rev 2025; 21(5): e260424229342.

40.

Gonçalves

Vale

Silva

. Neuroprotective effects of olive oil: a comprehensive review of antioxidant properties. Antioxidants (Basel) 2024; 13: 762.