What follows is the scientific analysis conducted by Dr. Jacob Friest, PhD Organic Chemistry, Chair of the Nebraska Section of the American Chemical Society. He had access to any test results he wanted. We tested anything he wanted and provided samples from our vats for further independent evaluation. Below is just a sample of the many tests he reviewed:
Midwest Laboratories, Inc. Report of Analysis FCLO 20946ABS September 23, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO 07252ABS September 23, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO 05751ABS September 23, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO 31641ABS September 23, 2015
Midwest Laboratories, Inc. Report 15-265-4118 FCLO September 22, 2015
Eurofins Nutritional Analysis Center Certificate of Analysis AR-15-QD-115048-02
September 23, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO C14061815 September 15, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO C15061815 September 15, 2015
Midwest Laboratories, Inc. Report of Analysis FCLO C16061815 September 15, 2015
Midwest Laboratories, Inc. Report of Analysis Fermented Liquid Fish Brine September 28, 2015
Fermented Cod Liver Oil (FCLO): Investigation of Green Pastures Fermentation Process and Food Safety Implications.
Author: Jacob A. Friest, PhD Organic Chemistry, 2015 Chair of the Nebraska Section of the American Chemical Society.
Release Date: 27 October 2015
A recent article published by Dr. Kaayla T. Daniel, “Hook, Line and Stinker! The Truth About Fermented Cod Liver Oil” has recently raised concerns regarding the safety of fermented cod liver oil (FCLO) products manufactured by Green Pastures. David Wetzel, owner of Green Pastures, requested that I conduct a review of Dr. Daniel’s assertions to determine their validity and to further evaluate Green Pastures product testing to ensure that their product is safe.
The key assertion of Dr. Daniel, that fermented cod liver oil (FCLO) as manufactured by Green Pastures cannot be from the fermentation of cod liver oil (CLO) or fats, is confirmed. Fermentation experts would all agree with Dr. Daniel that oils/fats cannot be fermented. However, Green Pastures does not claim, neither do they advertise, that their product is produced by the fermentation of CLO but by that of the carbohydrates found in the whole cod livers themselves. As Dr. Daniel pointed out in her article, cod livers contain between 1 and 2 grams of carbohydrates for every 100 grams of liver. Typical fermentation processes only require 0.62 grams of carbohydrate per kilogram to lower the pH by 0.1 pH units1. Based on the 1 to 2 grams of carbohydrate/100 grams of liver, this would give rise to a total drop in pH between 1.6 to 3.2 pH units. If we assume that the process of fermenting a cod liver begins at neutral pH of approximately 7.0, the natural levels of carbohydrate found in cod livers would be sufficient to lower the pH of the final fermented product batch to between 3.8 and 5.4. This brings us to the question of what is an acceptable pH range to achieve a safe-for-human consumption product. Dr. Daniel asserts that the pH needed to prevent food spoilage for true lacto-fermented product is less than 4.6. However, as published by the Food and Agriculture Organization of the United Nations (FAO)), raw fermented sausages are only moderately acidic with a pH range of 5.0-5.5 and are safe for human consumption2.
In her article, Dr. Daniel indicates that the pH of the bottled FLCO product is between 5.17 and 6.0. After questioning David Wetzel about Green Pastures fermentation process these numbers are as expected and in no way disprove that effective levels of fermentation were achieved to ensure product safety. To make sense of this I will first describe Green Pastures fermentation process: (1) the frozen cod livers are added to the fermentation vats along with Green Pastures starter culture and salt. The vats are then sealed and allowed to ferment. This process produces three distinct layers within the fermentation vats. At the bottom of the vat is the solid liver material and sediment. On top of this is a water layer which is formed as water is released from the cod livers during fermentation. On top of the water, an oil layer is formed. (2) After the fermentation is complete, this top layer containing the coveted oil is pulled from the fermentation vat and centrifuged to remove all sediment and liver material as well as separate out any water that was pulled from the vats with the desired oil. It is important to understand the layers formed during the fermentation process and how the oil is separated/purified by centrifugation to understand why the FCLO pH falls outside the normal pH range for fermented animal meat products. This is because the major acid by-product of a lacto fermentative process is lactic acid. Lactic acid is a highly water soluble acid and is extracted from the resulting oil during the pulling and centrifugation of the oil. To be certain of this, Green Pastures has measured the pH of the resulting water layer (brine) at the end of the fermentation process and found it to be between 4.8 and 5.04 (see attached testing from Midwest Laboratories). These pH values fall well within the normal pH levels accepted for the fermentation of raw meat products as described by the FAO.
Determination of FCLO Rancidity
There are two pathways by which fatty esters or triglycerides become rancid. Hydrolytic rancidity, which occurs when the fatty esters are hydrolyzed to free fatty acids, and oxidative rancidity leading to the formation of hydroperoxides and aldehyde by-products. According to Dr. Daniel’s article, fatty acid levels based on acid value are the most reliable markers to determine the hydrolytic rancidity in FCLO. However, acid value is determined by an acid/base titration with potassium hydroxide or sodium hydroxide and is not selective for fatty acids. Acid value is a measure of total acids in a sample. Since acids are a desired by-product of lacto fermentation it is not surprising that the acid values for FCLO are high since it would undoubtedly be expected to contain other free acids, not related to fatty acids, but rather by-products of carbohydrate fermentation. Therefore, acid value determination is not a good measure of hydrolytic rancidity for fermented cod liver oil since it is not selective for measuring only free fatty acids.
Dr. Daniel also states that fatty acids, the product of hydrolytic rancidity, are toxic to cell membranes. Unfortunately, this is a tactic to add scare value to her argument, and in the case of potential fatty acids being present in FCLO, is completely untrue. What Dr. Daniel has failed to state is how fatty esters or triglycerides are metabolized by the body3. When fatty esters and triglycerides enter the intestinal tract they cannot be absorbed by the duodenum. Instead, pancreatic lipase and bile hydrolyze the ester bond and release the fatty acids which are then absorbed by the duodenum. In this regard, if there were any free fatty acids present in FCLO, they would truly be in a “pre-digested form” and more readily taken up by the absorptive enterocyte cells lining the intestines where they are reassembled into triglycerides and packaged together with cholesterol and proteins to form chlyomicrons. The resulting chlyomicrons are excreted from the cells and collected by the lymph system and mixed into the blood. Various tissues can then recapture the chlyomicrons and release the re-built triglycerides to be used as a source of energy. In short, our bodies would take the ester or triglyceride form of cod liver oil and hydrolyze it to the free fatty acid form so our bodies can adsorb it. Free fatty acids in FCLO would not require this digestive step and would be readily adsorbed by the intestinal enterocytes directly and reassembled into triglycerides for transport.
Oxidative rancidity in fatty oils occurs as oxygen reacts with the oils to form hydroperoxides. Secondary to this is the decomposition of the hydroperoxides over time, to aldehydes and other carbonyl compounds. The key indicators for oxidative oil rancidity are to measure the oils peroxide value, p-anisidine value, or a TBA/TBARS levels. As Dr. Daniel explains in her article, peroxide value and the TBARS test are poor indicators of rancidity for FCLO due to the breakdown of the peroxides to aldehydes, and in the case of the TBARS test, interfering compounds in the product matrix. However, since a primary concern of rancidity is the formation of hydroperoxides and other reactive oxygen species, I would contend that peroxide value is indeed a valuable snapshot of the product in its current state, showing that reactive oxygen species are at safe levels for human consumption. In this regard, the peroxide values for several FCLO lots tested at levels between 2.9-6.7 meq/Kg, well below the levels established by the FAO, indicating that FLCO is not rancid.
The TBA or TBARS tests are also measures of aldehydes present in oils and are very commonly used to determine oxidative rancidity of older oils. Specifically, these methods are concerned with the detection of malondialdehyde formed during oxidative degradation of oils, which is reactive and potentially mutagenic and has been found in edible oils that have been heated. It is interesting that 2 out of the 3 labs that tested FCLO for Dr. Daniel reported safe levels for the TBA and TBARS values yet she is quick to disregard this testing as not being a valuable indicator of whether FCLO is safe since the formed aldehydes must have reacted with the proteins/DNA in the cod liver itself. Since the key question is whether FCLO is safe, it is a very important test! The process by which Green Pastures separates FCLO from the fermentation mixture likely removes all protein from the oil. First, the proteins/DNA would be more water soluble and therefore remain in the water layer from the fermentation process and secondly, if any solid liver material was removed with FCLO during the “pulling” of the oil it would be completely removed during centrifugation. Therefore, if any oxidative rancidity of the pure FCLO were to occur, the aldehydes, especially malondialdehyde, would not be able to react with any proteins/DNA since it has been removed from the oil. Therefore, these tests are in fact demonstrating again that FCLO is not rancid.
The third testing method discussed by Dr. Daniel for determining oxidative rancidity is p-Anisidine Value (AV). This is a test which measures aldehyde levels that arise as secondary oxidation products formed during the decomposition of hydroperoxides. The determination of aldehyde levels is a valuable indicator in determining oxidative rancidity in older oils. Not surprisingly, all testing results by Dr. Daniel are well below FAO guidelines, yet she concludes that the testing is not valid since many of these aldehyde by-products are volatile and escape to the environment before testing can be carried out. However, whether Dr. Daniel’s testing was performed on freshly pulled FCLO right out of fermentation or on the finished product, the testing PV should provide an accurate picture of late-stage oxidative rancidity. Why is this? Because the fermentation process occurs in a sealed vat, meaning that any aldehydes formed during fermentation cannot be lost, and since the finished product is purified by centrifugation and not distillation it would not be expected that these aldehydes would be lost at this point in the process as well. In fact, unless the labs who performed the testing of FCLO left the FCLO sample bottles uncapped and exposed to the air for extended periods of time, it is unlikely that this is a logical argument as to why FCLO tested within safe limits for all three testing methods (PV, TBA, and AV). What seems more likely is that Dr. Daniel is coming to these conclusions based on the acid value testing, even when all other results are to the contrary, and rather than consider why the acid value testing might be flawed, she simply discredits all other testing to support a hypothesis. Dr. Daniel’s conclusion about her own testing suggests that she began her investigation into FCLO already convinced that FCLO was rancid and this left her blinded to results that have clearly demonstrated otherwise.
Finally, let’s consider whether either of these forms of rancidity would even be expected to occur. As far as hydrolytic rancidity is concerned the conditions that would favor this process would be either very acidic or highly basic conditions. The fatty esters would be expected to be very stable in the pH range for which FCLO is produced. Additionally, since water is required for the hydrolysis of the fatty esters during hydrolytic rancidity, once the FCLO is pulled and centrifuged in the Green Pasture process, most of the water is removed thereby reducing the likelihood of hydrolytic rancidity. Lastly, oxidative rancidity is also greatly minimized in the Green Pastures fermentation process since it is carried out in the absence of oxygen which is required for hydroperoxide formation to cause oxidative rancidity.
Levels of Biogenic Amines Found in FCLO
Biogenic amines (BA) are organic, basic, nitrogenous compounds of low molecular weight that are formed by the decarboxylation of amino acids and occur due to biological activity. BAs occur naturally in animals and humans and are important as neurotransmitters, blood pressure regulation, and cellular growth regulators. However, BAs can become hazardous if their levels reach a critical threshold. The most important BAs found in food products are histamine, tyramine, putrescine, cadaverine, and phenylethylamine. Biogenic amines, especially histamine, are associated with the pathogenesis of food poisoning and Scombroid fish poisoning (SFP)4. SFP occurs in healthy individuals when a dose of 50 mg of Histamine is consumed from a single serving of a fish product (250g serving size).4 This correlates to fish with histamine levels exceeding 200 mg/Kg. For fish to contain such levels of histamine, which would cause SFP, the following conditions must be met4:
the fish are of a species that contain sufficient free histidine to be converted to histamine,
the presence of histamine producing bacteria,
conditions that support the growth of histamine producing bacteria and their production of the enzyme histidine decarboxylase enzymes.
In the case of FCLO, cod is not a species of fish that is associated with having high or sufficient levels of histidine to lead to histamine levels of 200 mg/Kg4. The conditions for fermentation do support the growth of histamine-producing bacteria, however, this is easily controlled by the use of starter cultures that do not contain amino acid decarboxylase enzymes. Moreover, the risk of SFP is greatly mitigated even before Green Pastures receives the frozen livers by the implementation of a HACCP plan. An HACCP (hazard analysis critical control point) is a management system in which food safety is addressed through the analysis and control of biological, chemical, and physical hazards from raw material production, procurement and handling, to manufacturing, distribution and consumption of the finished product. Controls for histamine, as well as other biogenic amines, in susceptible fish have been identified5. The risk mitigation strategies, which have been adopted by the FAO/WHO, detailed in the above reference include the following:
post-harvest chilling of fish,
gutting and gilling of susceptible fish,
freezing and refrigerated storage,
use of decarboxylase free starter cultures for fermented fish products.
Green Pastures has assured me that an HACCP plan is in place for the harvest and handling of their cod livers from the harvesting of the fish all the way through the fermentation process.
In addition to utilizing a HACCP plan by Green Pastures, the most important question to ask is whether FCLO is expected to be at risk for containing high levels of biogenic amines. The answer is no. This is due to the way in which Green Pastures produces FCLO. Since biogenic amines are highly water soluble and not fat soluble, any biogenic amines formed during fermentation, would be extracted to the aqueous layer of the fermentation vat. After removal of the water and sediment layers from the FCLO by centrifugation, biogenic amine formation is completely eliminated in the finished product since none of the required precursor amino acids would be present.
Analysis of FCLO Samples by Midwest Laboratories and Eurofins Nutritional Analysis Center
In response to the article published by Dr. Daniel, Green Pastures sent FCLO samples to Midwest Laboratories and Eurofins Nutritional Analysis Center to determine the rancidity profile of FCLO based on FAO guidelines and testing recommendations. I would like to make special note of the quality and authenticity of the reports provided for this investigation by Green Pastures. The reports were provided as prepared by the testing labs and include sample lots, testing dates, and by whom the testing was performed. It is unethical to publish redacted test results or to omit these key identifiers. By reporting her findings in this way, Dr. Daniel has made it impossible for any other labs to corroborate or discredit her testing and by doing so invited the authenticity of the data to be called into question. All lab tests results from Midwest Laboratories and Eurofins Nutritional Analysis Center clearly demonstrate that FCLO is not rancid with the exception of acid value which is higher than recommended by the FAO. This can easily be explained by the nonselective nature of this test and the fact that all fermented products are expected to contain elevated levels of acids, not necessarily free fatty acids, but likely acid by-products of carbohydrate fermentation (see attached reports).
Testing of FCLO produced by Green Pastures clearly demonstrates that FCLO is safe for human consumption and displays no signs of product rancidity or significant levels of biogenic amines. Moreover, after reviewing Dr. Daniel’s arguments regarding the safety of Green Pasture FCLO, it is clear that her analysis of the issue is flawed and her conclusions are incorrect and have misled the public. Whether this was intentional or unintentional is yet to be seen. Furthermore, the authenticity of her data is questionable since the key identifiers that would allow independent testing by other labs have been withheld. It is recommended the Dr. Daniel provide full transparency in her testing and data, and review the nature of FCLO and the Green Pastures fermentation process. On the other hand, Green Pastures has provided more than adequate testing to prove that FCLO is in fact not rancid, as well as to support their claims regarding the levels of fat soluble vitamins in FCLO.
1. Fermented Meat Products: Production and Consumption. Herbert W. Ockerman and Lopa Basu, Ohio State University, Columbus, OH
2. Meat Processing Technology, Food and Agriculture Organization of the United Nations, 2007.
3. Fats and Oils in Human Consumption: Report of a Joint Expert Consultation. FAO 1994.
4. Public Health Risks of Histamine and other Biogenic Amines from Fish and Fishery Products: Meeting report, 23-27 July 2012, Rome, Italy
5. Scientific opinion on risk based control of biogenic amine formation in fermented foods. EFSA J., 2011, 9(10): 2393.