2.1. Questionnaire
To investigate the consumption and cooking patterns of Tah-Dig among the community, a questionnaire containing 34 questions was designed taking into account the cooking methods, consumables, and also the types of interests of the people. For content validity, CVR = Content validity ratio and CVI = Content validity Index were calculated. Five nutritionists, three food industry specialists, a statistician, and an epidemiologist familiar with Iranian cuisines were consulted. Questions that did not meet the quorum of CVR = 0.062 and CVI = 0.79, and had less than 5% response were eliminated. To evaluate the reliability, Test-Retest was used in a pilot population equivalent to 10% of the prototype volume at 15-day intervals. Subjects with a special diet were excluded from this study. The final questionnaire consisting of 25 questions was prepared. KUPPUSWAMY’s SES scale was used for socio-economic ranking with three criteria of the job, education, and monthly income. Individuals' monthly income was measured indirectly based on the level of the urban area. Three levels of SES were considered high, medium, and low.
2.2. Determine the sample and size of the population
The cross-sectional, descriptive-analytical study was conducted on the population of Kermanshah (Western Iran) from 2019 to 2020. To identify possible differences in cooking and consumption methods among Iranian population, questionnaires were randomly completed in four other cities besides Kermanshah. The sample size considering the confidence level of P = 0.095 which is proportional to the consumption of Tah-Dig in the community and due to the lack of similar studies, 600 people for the city of Kermanshah (with eight urban areas) as the study center and the cities of Sari (north), Isfahan (center), Bandar Abbas (south) and Hamedan (west) a total of 150 people were identified. The sample size was proportional to the total population of that area. 48 and 270 Tah-Dig in pieces were required to measure the levels of the acrylamide and malondialdehyde content, respectively.
2.3. Sample collection and categorization
Participants were asked to provide us Tah-Dig in one piece cooked in the same way as usual in the time allotted for collecting the sample, and because of the stability of acrylamide and oxidized fatty acids we stored them in a -20 ° C freezer until delivery. A total of 343 samples were collected. Due to the non-uniform cooking conditions such as temperature, time, and type of oil in households, unlike laboratory conditions, the color was considered as a criterion, and samples with similar colors were placed in a group due to having the same amount of acrylamide. The samples were compared using Adobe Photoshop CC20.0.1 (Figure 1), and the samples with the same RAL and mean RBG±15 were grouped (Table 1). The classification was based on the spectrum of colors in gold, light brown, and dark brown for each type of Tah-Dig. Four control samples of high-consumption flat breads in Iran (Lavash, Taftoon) used in baking Tah-Digs were also prepared. Oils were classified based on the three main groups (solid, liquid, and combined) and nine subgroups (Kermanshahi oil, vegetable solid, canola cooking oil, sunflower cooking oil, oil Sunflower frying and combination of Kermanshahi oil with other oils). Kermanshahi oil is prepared by animal butter especially sheep butter, with a pleasant taste and smell.
Table 1
Classification of Tah-Dig samples based on color analysis (RBG).
Type of Tah-Dig
|
Color grade⃰
|
Red
|
Green
|
Blue
|
Rice
|
R1
|
194.2 ± 12.93
|
175.4 ± 12.9
|
97.36 ± 16.36
|
|
R2
|
188.36 ± 14.5
|
166.45 ±8.9
|
60. 36 ± 15.22
|
|
R3
|
147.20 ± 35.7
|
109.00 ±42.6
|
34.00 ± 12.36
|
Bread
|
B1
|
215.40 ± 15.23
|
187.70 ±27.166
|
82.90 ± 24.97
|
|
B2
|
182.10 ± 12.81
|
134.00 ±16.06
|
14.00 ± 7.3
|
|
B3
|
163.00 ±28.26
|
109.00 ± 32.42
|
15.20 ± 7.16
|
Potato
|
P1
|
232.70 ± 3.16
|
196.00 ± 17.98
|
80.80 ± 26.39
|
|
P2
|
204.90 ±5.01
|
176.10 ± 13.81
|
48.5 ± 15.50
|
|
P3
|
208.50 ± 8.700
|
110.90 ± 15.98
|
6.900 ± 3.695
|
* The numbers represent the pixels counted by color and are reported as Mean ± SD.
- The samples had a significant difference (P <0.0001 ) within the group at the color level, which represents three separate color spectrums.
- R = rice, B = bread, P = potato. Grades 1, 2, and 3 indicate an increase in color.
2.4. Statistical analyses
In the present study, central tendency indicators such as mean and standard deviation or percentage for continuous and classified variables were used to describe the data. Due to the method and nature of the research, analysis of variance (ANOVA) was used to compare continuous demographic variables and the Chi-square test was used for variables classified into different amounts of consumption. Data were analyzed using SPSS 24.0 software package. The data were collected precisely according to hypothesis and the methods were planned in-line with major purpose of the study.
2.5. Standard preparation for acrylamide
Stock solutions of acrylamide and acrylamide-d3 (Sigma-Aldrich, USA) were diluted to 1 mg/ml in deionized water. A standard diluted solution of 10 μg/ml in deionized water (18 Ω) was then prepared. Tah-Dig samples were evaluated within the dynamic standard range of acrylamide 20-3000 ng/ml. To prevent light exposure, all standard solutions were placed in a volumetric flask and kept at 4 °C for 4 hours. Carrez I solution was prepared by dissolving 1.5 g of potassium hexacyanoferrate in 10 ml of water, and Carrez II solution was prepared by dissolving 3 g of zinc sulfate in 10 ml of water to make diluted standard solutions.
2.6. Extraction of acrylamide
Tah-Dig samples were homogenized at 100 mg/ml in deionized water (18Ω). Then, 50 ng/ml of acrylamide-d3 as internal standard (Sigma-Aldrich, Germany) was added to the samples and incubated for 20 min at 4 ° C in the dark to allow for equilibration. It was then centrifuged at 6000 × g for 20 minutes. One milliliter from the aqueous phase of the supernatant was removed and 650 µl of supernatant was purified by Oasis HLB 6 cc Vac Cartridge, 200 mg Sorbent (Waters Corp, USA). The samples were then stored at -20 °C until analysis.
2.7. LC-MS/MS analysis
Samples were analyzed by LC-Tandem Mass (Quattro Micro API micro mass Waters 2695, USA) [1]. EC HPLC column (analytical), NUCLEODUR C18 Gravity, 5 µm, 250x2 mm (MACHERY-NAGEL, GERMANY) was used for chromatography with isocratic conditions: Methanol: Formic acid (94: 5: 1) was used at a flow rate of 1 ml/min. The LC-MS/MS parameters were as follows: Source temperature 500 ° C; Separation Gas Collision (CAD) 5; Nebulizer gas (GS1) and auxiliary gas (GS2) below 40 psi and 50 psi, respectively, and curtain gas (CUR) below 10 psi. Declining potential (DP) and collision energy (CE). Finally, a general scan was performed to determine the standard m/z ratio of acrylamide and acrylamide-d3 to select the most appropriate mass spectrum [20]. Dynamic range was considered 15-1000 ng/ml and LOQ was considered 15 ng/gram of Tah-dig that the device could detect.
2.8. MDA Extraction and Evaluation
Tah-Digs samples were homogenized at 100 mg/ml in 50% acetic acid. It was then centrifuged at 6000 × g for 20 minutes and the supernatant was stored at -80 °C until the time of the experiment.
The commercial lipid peroxidation kit (Kiazist Life Sciences, IRAN) was used to measure MDA. 200 µl of samples, standards, and blanks were poured into the microtubes, then 600 µl of TBA solution in the kit was added and the lids of the microtubes were tightly sealed with parafilm to prevent evaporation. It was then incubated for one hour at 95 °C. After incubation, the samples were cooled to room temperature and 200 µl of samples were transferred into a 96-well plate of the kit and their absorption was read at 560 nm by plate reader.
2.9. Measurement of triglycerides
Due to the difference in the amount of oil consumed for making Tah-Digs, MDA samples were evaluated based on grams of oil consumed. The triglyceride content of the samples was measured by the commercial kit (Pars Azmoon, Iran) to normalize the MDA values. Triglyceride in the samples was evaluated by the glycerol phosphate dehydrogenase (GPO) method. Homogenized samples were exposed to Working Solution for 10 minutes at room temperature to measure MDA. Afterwards, the absorption of TG at 500 nm was measured by a plate reader (BioTek Instruments' USA) and the TG value of the samples was calculated based on the standard absorption diagram.
2.10. Risk assessment
The incremental lifetime cancer risk equation (ILCR) was used to assess the risk of Tah-Dig consumption:
ILCR = CDI × CSF
The ILCR is an estimate of chroninc daily intake (CDI) of the chronic daily intake of acrylamide (mg / kg/day), the values of which are based on the average intake based on questionnaire data and measurement of LC-MS/MS acrylamide. The equation is shown below. CSF equivalent to cancer slope factor -1 (mg / kg-day), which was considered equal to 0.5 based on 2009 Oehha [20].
The CDI is derived from the above-mentioned equation in which C: the amount of acrylamide per gram Tah-Dig, IRi: the amount of Tah-Dig received based on the defined serving (the amount of Tah-Dig consumed per person according to the usual cooking utensils on the market (sizes 18 cm, 24 cm, 26 cm and above, with an estimated amount of 300, 200 and 100 g of Tah-Dig). Then, the amount of Tah-Dig (gram) was divided by the number of family members to get the consumption (per person (g) / 365), BW: Individual weights (45, 65, and 70 kg), which were defined for people aged 18, 20 years and those over 70 years, respectively. EDi: years of consumption (due to not consuming Tah-Dig in childhood, the age minus four was considered), and EFi: exposure time per year (repeated per week × 56).
2.11. Data mining analysis of Tah-Dig cooking patterns related to cancer risk
Data mining analysis is a process to extract implicit patterns based on the factors leading to the prediction of an event. This study aimed at developing a predictive model to categorize the participants into two groups: those at high risk of cancer and those with no known risks based on the answers to the questions in the questionnaire and the effect they had on ILCR; therefore, the decision tree algorithm was used [21]. According to this classification, items such as high flame temperature, MUFA, and PUFA lipids, and fried Tah-Dig texture were found to be associated with high levels of acrylamide, while low flame temperature, SFA oils, and soft tissue of Tah-Dig were found to be associated with low levels of acrylamide. Other questionnaire responses were also included in the program. The ILCR index was defined as a dependent variable and other questionnaire questions were defined as an independent variable. Finally, the accuracy and performance of the model for use in similar conditions were evaluated.