TOTOX: Measuring oxidation in EPA and DHA oils

PART 1 – Type of EPA/DHA product class: Unflavored Oils
Oxidation generates a sequence of breakdown products, starting with primary oxidation products (peroxide value, dienes, free fatty acids), then secondary products (carbonyls, aldehydes, trienes), and then tertiary products. These oxidation by-products are what gives off the unpleasant smell and taste of spoiled fish. Measuring oxidation in Omega-3 oils is complicated due to the differences in chemical and physical characteristics of many commercially available products, which means not all methods to determine quality are appropriate for all types of oils [1].
There are primarily 2 analytical methods used to measure oxidation in Omega-3 oils; the peroxide value (PV) and the para-anisidine value (AV/pAV). Tests that measure PV and pAV are widely used to determine oxidative quality of EPA/DHA-containing oils. PV measures primary oxidation products (peroxides) that occur early in an oxidation process, whereas AV measures the secondary oxidation products (aldehydes) that occur a little later in an oxidation process [2].
General info about TOTOX 
TOTOX is a useful calculation often utilized to assess unflavored fish oil quality by estimating ‘total oxidation’ in the oil. The TOTOX value is an international target value that describes the total oxidation value by using both PV and pAV in the following calculation: PVx2 + pAV [3].
According to the strict limits of GOED, a PV value should not exceed 5 mEq / kg, an AV value should not exceed 20, and a TOTOX value should not exceed 26.
Peroxide Value
5 mEq / kg
p-Anisidine Value
Table 1. Maximum limits for Peroxide Value, para-Anisidine Value and TOTOX specified by the GOED Voluntary Monograph
“The p-anisidine value is NOT a valid test for many flavored oils, or for oils with colors like krill or virgin salmon oils” p.55 Ismail et al. (2016)
The equation for TOTOX calculation is only suited for unflavored fish oils [3-5]. This is because the pAV test is very sensitive and can react strongly to added components such as flavors and pigmentation, and can react to these despite not being the result of oxidation. Thus, there is significant variance in pAV results between various types of oils, and GOED recommends limiting the above calculation method to EPA and DHA oils with no added ingredients other than antioxidants.

PART 2 – Type of EPA/DHA product class: Flavored Oils
Flavored oils are a class of Omega-3 oils with growing importance in the Omega-3 industry. However, the p-AV test is not appropriate for measuring secondary oxidation in Omega-3 oils that contain flavorings [1, 3, 5]
Simplified version: When the fat in fish oil becomes rancid, peroxide and aldehyde molecules are formed as a result of lipid oxidation. However, natural aldehydes are also present in flavor extract but are not the result of lipid oxidation. Thus, when the AV measurement cannot differentiate between the aldehydes derived from flavor components and aldehydes derived from lipid oxidation, this leads to inaccurately high values.
Detailed version: Anisidine value (AV) is a measure of the aldehyde levels in an oil or fat, in particular those that are unsaturated (and principally the 2-alkenals). To determine AV, a solution of the oil is reacted with p-anisidine solvent with anisidine reagent to form yellowish reaction products. The AV is then determined from the absorbance measured at 350 nm, both before and after reaction. Authors describing the pAV test commonly add the statement “The method is not suitable for flavored or colored oils since they may contain compounds that have a high absorbance at this wavelength” [6].
In many fruit-derived flavors, the desirable odors, taste and colors are carried by compounds containing aldehydes (i.e. natural aldehydes, NOT caused by PUFA oxidation). Since pAV measures the presence of aldehydes, these flavorings can interfere with pAV results when added to oils, consequently leading to inaccurate results.
This has been shown in several studies, e.g Norveel Semb (2012) demonstrated that a 2% inclusion of lemon flavoring increased pAV more than 12-fold (Figure 1).

Figure 1. Effect of antioxidants and additives on AV measurements. The AV measurements given are mean values based on four parallels and uncertainty is given as a 5% RSD [5].
Similarly, Ye et al. (2020) evaluated 14 flavors to identify those with the greatest contribution to the measured pAV of fish oil. All fourteen flavors increased the pAV when added to fresh fish oil but chocolate-vanilla and lemon flavors generated the largest increase (Figure 2).

Figure 2. The changes in pAV caused by the addition of 2% (w/w) flavor. The values reported are the means of duplicate measurements with error bars indicating the propagated standard deviation [4].
This is why pAV measurements of oil with added flavor give highly unreliable results and
an alternative protocol is needed to evaluate the aldehydes that result from lipid oxidation [3].
Additionally, since one component of the TOTOX calculation is pAV, it also is not valid for any oils containing other ingredients (such as olive oil) or that have strong colors, including flavored oils, krill oils and virgin salmon oils.
PART 3 – Calculation of TOTOX of flavored oils
Despite the above explanation, GOED has recognized that a TOTOX value is necessary for flavored oils, too. Until a more reliable measurement has been developed, GOED presents a protocol on how to better account for these interferences by the following calculation:
TOTOX for flavored fish oils: PVx2 + (pAVt - (pAV* - pAV)) [3].
GOED implements the following methodological approach for PV and pAV:
As specified in GOED Voluntary Monograph

Measure the baseline pAV without flavor added

Determine the difference between the measured pAV and the maximum pAV (i.e. max 20). The difference is called delta pAV and is the allowable oxidation over time.
delta pAV

Mix the flavor into the oil and measure the batch-specific pAV of the flavored oil = pAV* value.

Calculate the maximum pAV that the flavored oil batch may achieve to remain compliant with GOED limits, by summing (delta pAV + pAV* = maximum pAV*). The maximum pAV* value constitutes a useful reference value for shelf-life testing of flavored oils.
maximum pAV*

If the batch is tested at a later time to test if the oil still stays within calculated maximum pAV*, the new value is called pAVt.
GOED implements the following methodological approach for TOTOX (maximum 26) [3]:

TOTOX at initial time point (or time of manufacture) = 2PV + pAV, where pAV is the anisidine value of the unflavored oil.
PVx2 + pAV
2x2 + 4.9 =
TOTOX: 8.9

TOTOX during shelf-life testing = 2PV + (pAVt - (pAV* - pAV)), where pAVt is the pAV at given time, and (pAV* - pAV) are the initial measured values for flavored and unflavored oils.
PVx2 + (pAVt - (pAV* - pAV))
2x2 + (56 - (48 - 4.9))
TOTOX: 12.9

So, we know that flavors interfere with the AV measurement, but what about olive oil?
  • Olive oil contains natural components that react to both PV and AV measurements.
  • These are NOT products of PUFA-derived oxidation, nonetheless raise the values, especially for PV.
  • Olive oil has a higher threshold value for PV (15 for olive oil vs 5 for fish oil) because it naturally contains these components interfering with the PV value1.
  • Unfortunately, there are no available calculation methods that take this into account.
  • Zinzino’s BalanceOil stays within all limits specified by GOED, even though olive oil contributes to an increased PV value.
1The limit of PV is 15 mEq/kg for extra virgin olive oils defined in the Codex Alimentarius Commission’s standard for olive oils (Agriculture and Consumer Protection Department, 1999).
Reminder: The whole purpose of PV, pAV and TOTOX is to measure PUFA-derived oxidation.
1. Ismail, A., et al., Oxidation in EPA- and DHA-rich oils: an overview. Lipid Technology, 2016. 28: p. n/a-n/a.
2. Dubois, J., Determination of peroxide value and anisidine value using Fourier transform infrared spectroscopy. 1996.
3. GOED, Technical Guidance Documents. 2020:
4. Ye, L., et al., Flavors' Decreasing Contribution to p-Anisidine Value over Shelf Life May Invalidate the Current Recommended Protocol for Flavored Fish Oils. Journal of the American Oil Chemists' Society, 2020. 97(12): p. 1335-1341.
5. Norveel Semb, T., Analytical Methods for Determination of the Oxidative Status in Oils, in Department of Biotechnology. 2012, Norwegian University of Science and Technology NTNU - Trondheim.
6. Steele, R., Understanding and Measuring the Shelf-Life of Food. 2004: Woodhead Publishing. 396-407.

Zinzino (Updated )
The post is closed to further comments.