Analysis of Fibrous Compounds Using a Pressurized and Non-pressurized Conditions

Neutral detergent fiber (NDF) and acid detergent fiber (ADF) contents were evaluated in pressurized and unpressurized conditions using samples of roughage and concentrates. In summary, the samples were dried, processed in a knife mill, weighed in nonwoven bags (100g/m²), placed in a container and treated with neutral or acid detergents. Extractions of NDF and ADF content were carried out in a non-pressurized condition at temperature of 100°C for 60min and in pressurized condition using different temperatures of 100 and 110°C for 60min. Results of the different temperatures using the pressurized procedure were compared to those obtained with the pressurized through the linear regression analysis. The method with the temperature 110°C for 60 min had a high level of agreement. Was not observed a bias potential of proportion (P>0.05). There was not a systematic inclination of the methods to overestimate or underestimate errors. This methodology can be carried out with roughage and concentrate feedstuffs simultaneously.


Introduction
The concentration of structural carbohydrates has been used to determine the nutritional and digestible quality of food, as well as being a predictor in mathematical models to estimate energy. Since 1960s the analytical methods to determine fiber content have been used and improved by researchers. The fiber detergent analysis system was initially proposed by Van Soest in 1963 to determine neutral detergent fiber (NDF) and acid detergent fiber (ADF). Initially, the proposed w to apply this methodology in forage, however it can be extended to analyze concentrated foods. Is well know that starch can contaminate samples, causing an overestimation of fiber values (Van Soest et al., 1991). Thus, the method to determine NDF and ADF has been improved to reduce the amounts of starch by using α-amylase, sodium sulfite, or 8M urea solution (Van Soest et al., 1991). Other adaptations have been proposed as an alternative to the original test, including an application of the filter bag procedure and a replacement of the conventional digester such as the use of pressurized equipment, e.g., an autoclave. (Pell and Schofield, 1993;Deschamps, 1999;Ferreira and Mertens, 2007;Senger et al., 2008).
The technique to determine NDF and ADF requires: (a) specialized equipment, which is very expensive and usually is not available in all laboratories, there may still be the formation of air bubbles inside the bags, at the time of the detergent boiling, which compromises its contact with the sample reducing the extraction efficiency of the non-fibrous components of the food (Gomes et al., 2011), or (b) requires pressurized equipment or an autoclave, which is more common in laboratories. According to Senger et al. (2008), the use of filter bags and autoclave treatment for the analysis of NDF or ADF results in a more practical and rapid test when compared with the conventional method using Gooch crucibles. In addition, it can be cheaper than using the ANKOM® fiber analyzer. However, there is no agreement about treatment duration and temperature described in literature.
The objective of this study was to test the accuracy when using pressurized equipment for the Journal of Agricultural Studies ISSN 2166-0379 2020 analysis NDF and ADF by means of evaluating two temperatures in a pressurized environment and comparing them to the non-pressurized equipment (fiber analyzer) operated at a temperature recommended by the manufacturer.

Food Sampling and Location
The experiment was performed at the Laboratory of Forage and Animal Nutrition at the Federal University of Pampa -Uruguaiana Campus/ RS. For this study, were used different samples from roughage and concentrate food to determine NDF (n=19 samples) and ADF  (Table 1). Additional information about samples and fiber content (NDF and ADF) can be found in Table 1. Averages in g/NDF and ADF

Fiber Analysis in Pressurized and Non-pressurized Conditions
The samples used in this study were pre-dried in an oven with forced ventilation at 60°C for 72 hours and processed in a knife mill using a sieve with a porosity of 1 mm. To carry out the analyses, nonwoven bags (100 g/m² ) were made, with approximately 25cm² , heat sealed, and dried in the oven for 12 hours at temperature of 105°C. After this procedure, the bags were weighed on an analytical balance and properly labeled. The samples were placed in the nonwoven bags, respecting the ratio of 20 mg of dry matter per cm² (Nocek, 1997).
The extraction of the detergent fiber was carried out in triplicate for each sample and both were arranged in a device with a pressurized and non-pressurized condition, following the detergent to sample ratio of approximately 100 ml/g. A repetition of the all analytical execution was performed. For the concentrated samples, 8M urea and heat-stable α-amylase were used (Termamyl 120L, Novozymes Latin Amé rica ltda.) To substantially reduce the amounts of starch, samples were submerged in a 1L beaker for 4 hours ( VAN SOEST et al., 1991). To assess the fiber content in neutral detergent (NDF) and acid detergent (ADF), the detergent was prepared according to the recommendations of AOAC 2002.04; MERTENS, 2002.
To determine NDF and ADF in a non-pressurized condition, an equipment model TE-149, manufactured by Tecnal ® , with a capacity of 30 tests separated into 10 perforated discs, was used. The foods were separated into concentrates foods and roughage and, subjected to a temperature of 100 º C for 60 minutes. After the procedure, the bags were washed sequentially, at least three times, with hot distilled water and soaked with acetone to remove the remaining detergent. Bags were dried in an oven with forced ventilation at 60º C for 24h.Subsequently, bags were dried again in an unventilated oven at 105º C for 2h. Then, the bags were placed in a desiccator until they reached room temperature and afterwards, weighed in a precision analytical balance ISSN 2166-0379 2020 To determine the NDF and ADF in a pressurized condition, model AV 18L equipment, manufactured by Phoeniz luferco, was used. The samples were placed in a Becker containing a solution of neutral or acid detergent. The Becker was properly sealed to prevent the entry of steam, and placed in a vertical autoclave for 60 minutes. This procedure was tested using two diferent temperatures: 100º C with pressure at 0 Kgf / cm² and 110 ° C 0.33 Kgf / cm² . As in the previous method, the samples were also separated into roughage foods and concentrated. After treatment, the bottles were removed from the autoclave and the bags followed the same procedure describe previously for non-pressurized condition.

Statistical Analysis
The data were analyzed using the IBM SPSS Statistics software version 20.0. Shapiro -Wilk test was used to verify the normality of the data and Levene's test was used to check the homogeneity of variances. To assess whether there are differences between the variables, where the hypothesis of bias being zero or not was tested by the two-tailed t test, where there is agreement by P> 0.05. A simple linear regression equation was performed for the values obtained in a pressurized environment (Y) over the values obtained in a non-pressurized environment (X). the statistical evaluation being conducted under the following assumptions: H0: β0 = 0, and β1 = 1; vs. Ha: not H0, where the regression slope deviation of 1 was assessed using a two-tailed t-test. For the case of non-acceptance of the null hypothesis, it was concluded that the extraction environments are different.

Results
The difference in NDF and ADF values was evaluated between the non-pressurized environment and the two temperatures in a pressurized environment (Table 2). In the NDF analysis, we observed a significant difference between the pressurized at 100 °C and nonpressurized environments (P<0.05), and there was an agreement (p-value = 0.59) between the pressurized at 110 °C and non-pressurized environments. In the ADF analysis, there was an agreement between the non-pressurized environment and both pressurized environments at 100 °C (p-value = 0.31) and 110 °C (p-value = 0.87).  ISSN 2166-0379 2020 linear regression model y= β0+ β1x. 3 P-value of the predictor variable. NDF: neutral detergent fiber. ADF: acid detergent fiber.
The Table 2 also shows that a potential proportion bias was not observed (P<0.05) in either of the methods, i.e., there was no tendency for differences to be concentrated above or below the mean. Therefore, there was no systematic inclination of the methods to overestimate or underestimate errors. In the evaluation of potential proportion bias, it was observed that the methods had evenly distributed values (P<0.05). The linear regression analysis used the difference between methods as the dependent variable and the mean between the methods as the independent variable ( Table 2).
The slope of the regression for both NDF and ADF did not differ from 1 (P> 0.05). The treatment with autoclave for NDF using the temperature of 110 º C had the lowest coefficient of determination R² = 0.76. For the ADF analysis, the highest coefficient of determination was for the temperature of 100 º C (R² 85%), however, for the temperature of 100 the R² was 78% (Table 3).

Discussion
The use of pressurized equipment, such as autoclave allows for a greater number of samples to be processed simultaneously. Deschamps (1999), tested a similar approach that we used in this study, using filter bags and a pressurized equipment at 120 °C for 40 minutes. He described that autoclave had a greater productivity since it allows to use 120 samples per operation and generated considerable reagent savings. Cordeiro et al. (2007) compared the contents of NDF and ADF using the conventional method described by Pell and Schofield (1993), which uses digesting blocks/filter crucibles and a temperature of 105 °C for 60 minutes.
There was no difference (p-value = 0.12) between the methods analyzed in this study, which proved the effectiveness to use pressurized equipment's to analyze fiber content, such as autoclave. To find an alternative procedure that was not different from the conventional method, Senger et al. (2008) evaluated different autoclave durations and temperatures in the NDF and ADF analyses. These authors concluded that the autoclave stated at 110 º C for 40 minutes did not differ (P<0.05) from the conventional method, that uses a temperature at 105 °C for 60 minutes. Additionally, the analysis of forages and concentrates could be performed simultaneously. Gomes et al. (2011) utilized filter bags and compared a pressurized conditiosn with a non-pressurized conditions with effective extraction time of one hour at a temperature of 100º C, observing differences between two conditions.
The results obtained in this experiment for NDF, showed to be sensitive to a temperature of 100 ° C. However, at a temperature of 110 °C, the NDF proved to be accurate for treatment using non-pressurized equipment. In the FDA tests, all temperatures tested were using a non-pressurized method and were consistent with the non-pressurized physical condition. Although the coefficient of determination, it is lower, both for NDF and FDA, using a temperature of 110º C, its use is recommended, as it is closer to the differences between the environments. According our findings we can recommend the use of pressurized equipment, such as the autoclave to determine fiber content. Other observations include: (a) the pressurization to avoid accumulation of gas in the bags, which can compromise the action of the detergent, and (b) the nitrogen content of the waste can be analyzed in a subsequent step by the Kjeldhal method (Senger et al., 2008;Gomes et al., 2011).

Conclusions
The analysis of fiber in neutral and acid detergent in pressurized condition at a temperature of 110 º C for 60 min was consistent with the non-pressurized method, just as roughage and concentrated foods can be performed simultaneously.