Study setting

This cross-sectional study was performed at Birjand University of Medical Sciences on a sample of 350 mothers who were recruited between January–February 2021 from 4 health centers in different regions of Birjand, South Khorasan, Iran. Inclusion criteria included; healthy women between the ages of 20–35 years, with an infant 1–6 months old. We did not include mothers with any acute or chronic illnesses. Mothers in this survey participated voluntarily and all of them gave written informed consent. Demographic, anthropometric, psychological, and dietary pattern information was determined by the means of a questionnaire. The sample size for the study was calculated according to 80% power and the following formula: n = [(Z1-α/2)2 -S2]/d2 which SD = 5.2, d = 0.504, and α = 0.1 [10]. According to this calculation, 294 participants were required; nevertheless, 350 participants were added due to data availability and to account for every possible exclusion. The study protocol was approved by the Ethics Committee of Birjand University of Medical Sciences.

Demographic and anthropometric variables evaluation

A trained nurse evaluated information including mother age, systolic blood pressure (SBP), type of delivery, body mass index (BMI), economic status, parent death, parent divorce, father s’ and mother s’ educational attainment, infant age, infant sex and infant head circumference. Participants’ height and weight were measured and BMI was calculated from these values as weight in kilograms divided by height in meters squared. Height and around the infant’s head circumference (cm) was recorded using an inelastic measuring tape to the nearest mm. Electronic scales were used to measure weight to the nearest 0.1 kg. SBP values in individuals in resting conditions were measured and repeated during the same visit [11].

Social-economic status and family situation

A standard questionnaire was used to collect data on family factors and socio-economic status. In terms of socio-economic variables, the highest mother’s and father’s education attainments were evaluated separately, with 3 response options: 'elementary (9 years),' 'intermediate (10–12 years),' and 'university (13 years).' Moreover, economic status was assessed by asking mothers to report their economic condition by choosing one of three answer choices: less than enough, enough, or more than enough. In related to family structure, mothers were asked to indicate parent death (Yes or No) and parent divorce (Yes or No).

Milk sampling

Human milk samples (1 to 6 months postpartum) were collected manually in the morning between 7 and 10 am. Samples were stored in a sterile tube. Before infant feeding, samples were taken of one breast in the morning. Each mother provided two 20 ml human milk samples (for a total of 700 samples). Samples were stored in sterile tubes and were transferred to our laboratory on dry ice. Samples were freeze-dried and then stored at − 80 °C until they were analyzed.

Oxidant-antioxidant status measurementFerric reducing, antioxidant power (FRAP) assay

The FRAP assay method was as previously explained by Benzie and Strain; based on the reduction of a Fe3+, tripyridyltriazine complex to the blue-colored ferrous form in the presence of reductants (antioxidants) [12]. Briefly, 10 μL of the test sample, standard (FeSO4) or blank (for each milk sample, a blank sample was used to remove milk turbidity) was mixed with 250 μL of FRAP reagent, freshly prepared (10 volumes of 300 mM acetate buffer, plus1 volume of 10 mM TPTZ solution in 40 mM HCl, plus 1 volume of 20 mM FeCl3· 6H2O) incubated for 10 minutes at 37 °C. The absorbance was measured at 593 nm. All tests were performed in duplicate and the results were described in μmol/L.

A-diphenyl- B-Picrylhydrazyl (DPPH) assay

The antioxidant activity of human milk samples was assessed in terms of free radical scavenging activity, according to the Brand-Williams method with a slight modification [13]. Briefly, 50 μl of each human milk and control sample were mixed with 950 μl of the DPPH solution in a test tube. After a 10-minute incubation at 37 °C, it was centrifuged at 3000 g for 3minutes. The absorbance of the reaction mixtures was measured at 517 nm by a spectrophotometric method, and a methanol solution of DPPH was utilized as a control sample. The percent of antiradical efficiency was assessed by the equation that follows: DPPH radical-scavenging activity (%) = [(absorbance of the control – mean absorbance of the sample)/absorbance of the control] * 100. Each test was carried out in duplicate and the results are expressed in μmol Trolox equivalent /L.

Thiobarbituric acid reactive substances (TBARs) assay

The final product of lipid peroxidation, malonyldialdehyde (MDA), was measured using the TBARs assay in human milk samples [14]. Samples (100 μl) were mixed with 1 ml of TBARs reagent (7.5 g trichloroacetic acid, 187 mg TBA, and 6.25 ml of chloridric acid). The mixture was heated for 20 minutes in a boiling water bath. Then 1 ml of N-butanol was used to extract TBARs adducts and the solution was centrifuged for 10 minutes at 1500 g at 4 °C. The fluorescence spectrum of these samples was measured at the excitation (515 nm) and emission (553 nm) wavelengths. The results (μmol TBARs/L) were compared to a standard curve.

Ellman assay

Total milk thiol concentration or sulfhydryl groups (T-SH) were assessed using the techniques originally described by Ellman and modified by Hu [15]. TS-H interacts with 5-thio-2-nitrobenzoic acid (DTNB), forming a highly colored anion at a maximum peak of 412 nm. In this assay, an aliquot of 50 μL of fresh milk was homogenized with 1 mL of Tris/EDTA buffer (0.25 mmol/L Tris base, 20 mmol/L EDTA, pH 8), 50 μL aliquot of DTNB stock solution (10 mmol/L in absolute methanol) and 650 μL N-butanol. Each duplicate sample was centrifuged at 3000 g for 5 minutes, then absorbance at 412 nm was measured [16]. The absorbance was measured again after 15 min at room temperature with a DTNB blank (with 50 μL methanol). The concentration of T-SH groups was estimated with reduced glutathione as the T-SH group standard and the data was demonstrated in μmol/L.

Milk calcium, protein, and triglyceride

All photometric analysis was evaluated using a plate reader (Epoch™, BioTek, Winooski, VT, USA), at 37 °C. Monochromatic readings were conducted to assess all absorbance data [17]. Calcium was estimated with the Arsenazo III kit (Pars Azmoon, Tehran, Iran) according to the manufacturer’s instructions for use. Calcium makes a compound with Arsenazo III at neutral-pH, and the color severity is proportional to the calcium content in the sample [18]. The absorbance of all samples and blanks was assessed at 660 nm.

The Bradford protein assay was undertaken using 10 μl of human milk sample and 1 ml of color reagent in duplicate. For 30 seconds, the contents were mixed and, after 5 minutes of incubation at room temperature, the absorbance values of all samples and blank (10 μl of each human milk sample and 1 ml of distilled water duplicate) were determined using a microplate reader at 595 nm [19].

Milk triglyceride values were assessed using a Pars Azmoon® kit (Tehran, Iran). It is an assay that employs enzymatic hydrolysis and quantification by measuring absorbance at 546 nm and the data is evaluated in mg/dL. An aliquot of 10 μL of human milk sample (1:10 diluted) duplicate was added with 1 mL of triglyceride reagent, vortex mixed, and incubated at 37 °C for 30 mins. The absorbance of all samples and blanks (1 ml of distilled water and 10 μl samples) was assessed at 546 nm.

Nutritional assessment

The dietary intake of all mothers was estimated using a validated food frequency questionnaire. This tool contained 65-items with the frequency of intake (per day, week, month, rarely, and never) and portion size for every food item [20]. Expert nutritionists performed face-to-face interviews with all individuals to complete the questionnaires. The analysis of the nutrient consumption of each person was conducted based on the US Department of Agriculture’s National Nutrient Databank [21]. The dietary nutrient intakes of participants were calculated [21] and then dietary patterns were determined based on 25 predefined food groups (Table 1) conforming to the similarity of food frequency questionnaire food items.

Table 1 Food groupings used in factor analysis of dietary patternsEvaluation of dietary patterns

The process for determining dietary patterns (exposure) from food intake data involved first grouping the 65 food items into the 25 predefined food groups (g day − 1) as predictor variables based on their similarity. We used factor analysis (Principal Component Analysis) to create dietary habits based on 25 foods or food groups [22]. To preserve the factors uncorrelated and facilitate interpretation, they were rotated orthogonally (varimax rotations). Factors with an eigenvalue > 1, the screen test, and the interpretability of the factors were taken into account while choosing the number of factors to keep [23]. As a result, the current study yielded a three-component dietary pattern (tertile). The first component (first tertile) shows how one dietary pattern is consumed less frequently, while the last component (third tertile) shows how it is consumed more frequently. Because it was the most interpretable component in the sample population. The first component of each dietary pattern was selected as the reference group. The factor scores for every dietary pattern were determined by summing the intake of the food groups adjusted by their factor loadings which were obtained by factor loadings. Finally, each mother was assigned a score in each of the two defined dietary patterns. Because the percentage of variance described by each dietary pattern is highly dependent on the number of variables used in the analysis, this criterion was not reported [24]. The reference category for every dietary pattern was selected as the first tertile.

Statistical analysis

All analyses were conducted using the Statistical Package for the Social Sciences (SPSS) version 16. Statistical data was measured for normality by using the Kolmogorov-Smirnov test. Continuous and categorical indices are displayed by Mean ± SD and number (percent), respectively. The ANOVA test was applied to measure the significant difference in normal variables among groups. Participants in the type of delivery and baby sex across tertiles of dietary patterns were compared by using the chi-square test. Multivariate logistic regression was used to evaluate the associations between a component of the score built to evaluate the adherence to the healthy, unhealthy pattern. All analyses were considered using two-tailed tests and a P-value < 0.05 was set as significant.