HOME > Chowhound > General Topics >


Sardines and PCBs

Hi all,

Does anyone know which canned sardines are lowest in PCBs?

Grateful for any/all info.

  1. Click to Upload a photo (10 MB limit)
  1. Sardines have the least pcb's of any fish, if any. The fish don't live long enough to pick up a lot of junk in their short lives. If you wade through these threads, there are links along the way about how safe sardines are to eat

    The Great Sardine Taste-off - the Sardine Saga continues (cans 31-39)

    From Seafood Choices Alliance:

    "The KidSafe Seafood program recommends sardines. They are sufficiently low in mercury and PCBs to be safe for children age three and up to eat at least once a week. "

    My guess if this is a concern for you would to be to look out on the can for where the fish were harvested ... that is harvested ... not packed. Often it is two different areas.

    3 Replies
    1. re: rworange

      Thanks, rworange.

      However, I quote from the original "great sardine taste-off thread":

      "depending on where they are caught some sardines are high in pcbs."

      Just wondering if anyone knows which areas of the world's oceans are the most clean. Sigh, what a world that such questions need be asked.

      1. re: sardinefan

        Ah, you deserve to know just for reading all that ... even I won't re-read it and forgot. When I was doing a quick Google, it seems most think the pcb's are low in sardines due to short life-cycle, though there are occasional mentions of high pcbs. Didin't look too closely, but there was some study on Spanish sardines ... can't remember if it was the males or females that picked up fewer pcbs ... and there was something about mating season that upped the fat in sardines and since that is where the pcbs accumulate, the pcb level.

        1. re: rworange

          Sardines are not at the top of food chain like tuna so they do not accumulate PCB's....

    2. Thank you, rworange and Pollo.

      I quote from an article posted on msnbc.com re safe fish consumption:

      Though low in mercury, this fatty fish contains high levels of other toxins.

      Nutritional overview: Sardines (Atlantic, canned in oil, drained solids with bone, 2 sardines [24g]
      ) Based on a 3.5 ounce serving
      Calories: 206.25
      Protein: 24.5g
      Total Fat: 11.3g
      Omega-3s: 2.3 grams

      Toxic Topline
      Sardines have their own paradox: very high in calcium due to their bones, rich in Omega-3s, low in mercury — but high in PCBs and other toxins.
      Consumption recommendation: The Environmental Working Group suggests avoiding sardines due to the high level of toxins.


      Who to believe?

      1. And now over to the Environmental Working Group:

        "According to the guidelines, to limit their intake of PCBs, women and children should eat no more than 1 to 2 servings per month of salmon, sardines, herring and bluefish."


        I give up. There's nothing left to eat.

        4 Replies
        1. re: sardinefan

          Discouraging. I'm afraid to even look for specific studies on North Sea herring...the new catch just came in here and I'd like just a few guilt-free moments with my little buddies...

          1. re: sardinefan

            Unless they actually show how the data was generated this kind of info is worth as much as last year's snow....to be polite....

            1. re: Pollo

              Which is why I posted the original question in the first place... wondering if anyone has concrete information.

              There is this study, which suggests that, in a sampling of Spanish fish, "Tuna, hake, and sardine were the species with the highest contribution [of PCDD/PCDFs and PCBs]."

              ...and there are other studies like this as well.

              1. re: sardinefan

                Like I said before...unless you know how the study was conducted (details) that kind of abstract information is useless....

          2. It looks like Pacific caught sardines and anchovies have high levels fo PCBs and DDT:


            1 Reply
            1. re: nwgirl

              This 2007 US Govt study of fish caught off California says sardines had 3 parts per billion of PCBs. The PCB level was below the limit set for wildlife exposure, meaning it was safe for wildlife to eat the sardines. And the Pacific sardines studied were the longlived variety (3 to 7 years) vs. the short life of most tinned sardines.

              This coastal California area was especially studied because it had known high levels of DDT and PCB pollution. Sardines from elsewhere are presumably much lower in PCBs.

              This study was done in 2007. It is called CHLORINATED HYDROCARBONS IN PELAGIC FORAGE

              Enjoy your sardines!

            2. There's both good and bad information in this thread. First, PCBs can accumulate in virtually any animals, and some plants - it doesn't matter where they are on the food chain, although it is true that, in general, all things being equal, species higher on the foodchain have greater exposure due to biomagnification, and older fish will have higher contaminant concentrations than younger fish because of the longer exposure time.

              Of much greater importance is the lipid (fat) content of the fish - as non-polar organic contaminants, PCBs partition virtually exclusively into lipids, so (again, all things being equal) fatty fish will have higher levels than lean fish. Sardines are comparatively fatty fish, so they have the potential to bioaccumulate PCBs, and other lipophilic contaminants, to a greater extent than leaner species with similar exposure. Note that in the 2007 Southern California Bight study Pacific sardines are mentioned as living from a few to several years - that's not all that short-lived and certainly gives them time to come into near-equilibrium with whatever concentrations they're being exposed to. At the PCB hazardous waste site that I've worked at for the last decade, fish hatched in any given year (so called "young-of-the-year") essentially reach equilibrium within that year and then remain in equilibrium throughout their lives.

              With respect to PCBs, it's very difficult to provide a simple answer to the question of whether it's safe to eat sardines, and in what quantity, because there's really no way of knowing the PCB content of the food that's on the market shelves. The current FDA action level of 2.0 mg/kg ww (parts per million, wet weight) is laughable, bordering on criminal - my recommendation is that anyone who's eating sardines or anything else with that concentration of PCBs should put their affairs in order. At the other extreme, the 3.0 parts per billion cited in the 2007 SCB study, approximately 1,000 times less than the FDA action level, is almost certainly safe for regular consumption. (I would not regard that study to be representative of sardines in general, however.) In between . . . well, as the concentration of PCBs increases, the risk of getting cancer due to consuming sardines increases (the so-called linear low dose response model used to calculate these types of risks doesn't really have a "zero" risk, though the real response may well have some sort of threshold), perhaps to something on the order of one excess cancer in 10,000 people, which would be considered a rate at a hazardous waste site at which action would be taken to remediate the problem. However, it's good to remember that 4,000 of those people are going to get cancer even if they never eat a sardine, so we're talking about an increase, for example, of from 4,000 to 4,001, which is comparatively small in the statistical sense (but not if you're the one person, obviously).

              I have a colleague who is highly regarded internationally for his knowledge of the factors that contribute to human health risk. He limits himself to one fish meal of any kind per week. That's probably a reasonably prudent thing to do.

              2 Replies
                1. re: FlyFish

                  Dear FlyFish,

                  THANK YOU. I wish you would write to Tara Parker-Pope and share your knowledge with her, or at least cross-post your informative response at the New York Times site.

                  After reading her blog, I asked my question there too (see comment #6), and no one seemed to know much of anything.


                  Thanks again.

                2. I was under the impression that this whole PCB in fish concern was raised when it turned out that farmed salmon was high in PCBs, and they were high in PCBs because of the fishmeal they were being fed:


                  Mercury accumulation is a different thing, isn't it?--and sardines are low in mercury b.c. they are low in the food chain. Seems by this thread that information about how fish are contaminated with PCBs is not really well-established.

                  1 Reply
                  1. re: Mandymac

                    You are correct concerning the issue with farmed salmon and PCBs, however that was specific to salmon farming and really has little to do with PCBs in the environment and how they enter and move through the food chain.

                    We actually know a great deal about how and why PCBs accumulate in fish (and other biota) and the risks that those PCBs pose to fish consumers. For example, it is possible now to build numerical models that accurately simulate the fate and transport of PCBs into and through the components of a complex food web. The situation is complicated to a large degree by the fact that PCBs are not a single chemical, but actually a family of 209 related chemicals known as congeners (perhaps 170 or so are actually present in the environment) that vary greatly in their physical and chemical properties and, consequently, in their toxicity and propensity to bioaccumulate. In addition, PCBs often co-occur with chemicals such as dioxins and furans that pose their own threats, in many cases much greater threats than from PCBs. Studies that deal with PCBs as a large group but fail to consider congeners and co-contaminants are now considered rather simplistic and are generally of limited value.

                  2. “PBCs" is the general term used to refer to one member of the "dioxin" family of compounds which consist of dioxins, furans, and PCBs. This family of chemicals are sometimes referred to as "dioxin-like compounfs" (DLCs). There are 75 dioxins, 135 furans, and 209 PCBs. Each different form is called a "congener." Not all of these "dioxin-like" chemicals have dioxin-like toxicity, and the toxic ones are not equally toxic. The least toxic of these compounds is estimated to be about 10,000 times less toxic than the most toxic. Only 7 of the 75 dioxins, 10 of the 135 furans, and 12 of the 209 PCBs have dioxin-like toxicity.
                    There are seven dioxin and ten furan congeners thought to be most toxic to humans. Most studies measuring human exposure to dioxin and furans focus on this group. The term "dioxin" is sometimes used to refer to this group of 17 congeners.
                    Because dioxin congeners are not all equally toxic, scientists have developed toxic equivalency factors (TEF) to compare the relative toxicity of congeners. The dioxin congener 2,3,7,8-tetrachlorodibenzo dioxin (TCDD) is the most toxic form of dioxin. It is assigned a toxic equivalency factor (TEF) of 1. Each of the 17 toxic dioxins/furans and 12 PCBs is then assigned a "toxicity factor" that estimates its toxicity relative to TCDD. Congener TEFs range from 0.0001 to 1.0. The 12 PCBs with a toxicity factor are not included in the group of 17 as their toxicity is lower.
                    The Toxic Equivalent (TEQ) of a food is its toxicity. TEQs are obtained by multiplying the concentration of each congener found by its relative potency (its "toxic equivalency factor," or TEF) and summing up the results for all of the congeners. The toxic equivalent of a foods is most often given as pg/g (picograms / gram) which is equal to parts per trillion. The TEQ system is not perfect, but it is a reasonable way of estimating the toxicity of a mixture of dioxin-like compounds.
                    The FDA’s Total Diet Study collects residue data for approximately 300 pesticides, radionuclides, and industrial chemicals in approximately 280 core foods in the U.S. food supply. Four market baskets are generally collected each year, once in each of four (West, North Central, South, Northeast) geographic regions of the U.S. For each market basket, food samples are collected from grocery stores and fast food restaurants in three cities within the region, The samples from the three cities within each region are combined into one ‘‘market basket.’’ The foods sampled include processed foods (bottled, canned, and frozen), fresh foods including fruits and vegetables, baby foods, dairy products, fresh meats, cereals, peanut butter, and prepared foods such as pizza. The three samples of like foods are combined into one sample, prepared as for consumption, and analyzed for contaminants. FDA's Dioxin Monitoring Program analyzed selected TDS samples from one market basket each year in 2001 to 2004. In addition to TDS food samples, FDA's Dioxin Monitoring Program involves the collection and analysis of additional non-TDS food samples to complement TDS findings by further estimating dietary DLC exposure from specific types of foods (e.g., species of fish). Non-TDS food samples collected and analyzed by FDA include dairy products, eggs, fats/oils, fruits/vegetables, grains/cereals, seafood (finfish/shellfish), tree nuts/peanuts, and dietary supplements. Samples were assigned and collected by one of ten FDA District Offices at retail, as domestic imports or at the grower for certain domestic aquaculture finfish samples (e.g., catfish, salmon, striped bass, tilapia, trout). Unlike TDS foods, non-TDS food samples were not prepared as consumed (i.e., cooked) prior to DLC analysis though only the edible portion was analyzed.
                    Three TEQ values were generated for each food, reflecting assignment of zero, half the limit of detection (LOD), or LOD. Some contaminates are below the level of detection and are referred to as "non-detects." The assigning of zero to the non-detects means that the non-detects are not taken into account and the results given are the actually results of the test done. The assigning of LOD means that any non-detects are fully contaminated. Using zero and the LOD provide lower and upper bounds on the range of contamination.
                    In 2003 the FDA collection and analyzed canned sardines. The seven dioxin and ten furan congeners thought to be most toxic to humans were analyzed. The 12 PCBs with a toxicity factor are not included in the group of 17, as their toxicity is lower. The total toxicity for the 7 dioxins total 2.31, the total for the 10 furans is 1.07, while the total for the 12 PCBs is .136.

                    The results (TEQ (pg/g) ND=0 and ND=LOD) were:
                    2003 176166 "Sardine, Canned" 0.5206 0.5557
                    2003 180879 "Sardine, Canned" 1.3861 1.3877
                    2003 180880 "Sardine, Canned" 0.2670 0.2675
                    2003 183769 "Sardine, Canned" 0.0061 0.0247
                    2003 184696 "Sardine, Canned" 1.5655 1.6271
                    2003 192530 "Sardine, Canned" 0.3111 0.3190
                    2003 200492 "Sardine, Canned" 0.1980 0.4014
                    2003 200534 "Sardine, Canned" 0.0943 0.1451
                    2003 200535 "Sardine, Canned" 1.1017 1.1236
                    2003 200536 "Sardine, Canned" 0.2559 0.2644
                    2003 208241 "Sardine, Canned" 0.5971 0.5975
                    2003 211491 "Sardine, Canned" 0.1315 0.2729
                    2003 211492 "Sardine, Canned" 0.2710 0.4419
                    2003 217013 "Sardine, Canned" 0.8647 0.8654
                    2003 218749 "Sardine, Canned" 0.0371 0.2009
                    2003 219412 "Sardine, Canned" 0.0130 0.2443
                    2003 219413 "Sardine, Canned" 0.0120 0.2254
                    2003 221597 "Sardine, Canned" 1.8935 1.8937
                    These sardines would have been purchased in Albuquerque, New Mexico, Riverside-San Bernardino, California, and Salem, Oregon.
                    Information such as how canned (in oil or water), where canned or where harvested is not given.
                    The EPA set their "virtually safe dose" for dioxin at 0.006 pg/kg bw/day.
                    And raised it in 1994 to 0.01 pg TEQ/ kg bw/day.
                    The World Health Organization sets their dioxin limit at 1 to 4 pg / kg bw/day.
                    So, for a 160 pound man the originally EPA limit would be .43 pg/day.
                    Their revised 1994 figure would allow the same man to consume .72 pg/day.
                    And the WHO says its ok to eat 72 to 290 pg/day.
                    I get it!!! Just keep raising the limit and we will all be fine.
                    For comparison, their average results for:
                    Flounder 0.0147 0.0275
                    Halibut 0.0010 0.0126
                    Swordfish 0.0493 0.1689
                    Tuna - Canned 0.0021 0.1790
                    Beef Sirloin Steak- Broiled 0.1095 0.1297
                    Butter 0.5545 0.4135
                    The highest single results they have for salmon is in 2002:
                    Salmon - Farmed 0.5545 0.9270
                    Salmon - Wild 0.4220 0.6458
                    The Natural Resources Defense Council
                    PCBs are a complex and hazardous group of chemicals. While their acute toxicity is much lower than many of the organochlorine pesticides and the dioxins, they have serious long-term health effects at relatively low levels. High levels of exposure can cause effects in infants ranging from low birth weight; to abnormalities of the skin, hair and nails; to hearing loss.
                    Dioxins and Furans are among the most hazardous chemicals known - extremely tiny doses have been shown to cause negative health effects. These chemicals are listed by several governmental agencies as known causes of cancer in humans. Indeed, studies have linked dioxins and furans to many types of cancer, as well as to reproductive problems, abnormalities in fetal development, immune alterations, and disruption of hormones. Because dioxins and furans are attracted to fat and are resistant to metabolism, they are notorious for accumulating in the animals humans eat, and by that route accumulating in humans.
                    Environmental Defense Fund
                    PCBs are not highly toxic with a single dose (as in a single meal), but continued low levels of exposure (for example, eating contaminated fish over an extended period of time) may be harmful. EPA rates PCBs as "probable human carcinogens," since they cause cancer in laboratory animals. Other tests on laboratory animals show damage from PCBs to their circulatory, nervous, immune, endocrine and digestive systems.
                    Dioxins are highly toxic byproducts of industrial processes. Like many other contaminants found in fish, these chemicals are slow to break down and they accumulate in the bottom sediments of streams, rivers, lakes and coastal areas. Dioxins can build up in the fatty tissues of fish and other animals, and in high enough concentrations pose serious health risks to people who frequently eat contaminated fish.
                    Based on studies done on laboratory animals, exposure to dioxins and furans is known to cause a variety of cancers and can harm the immune system. Effects on reproductive, endocrine, circulatory and nervous systems have also been observed. TCDD (the most hazardous of the dioxin compounds) exhibits the highest cancer potency of any chemical ever studied in animals.

                    Just Thinkin: Why all the commotion is over PCBs when its the dioxins and dioxin-like chemicals that are going to hurt or kill us. All the reports I see are about how many PCBs are found in farmed salmon, not the toxicity of those PCBs.

                    FDA’s Food Contaminants & Adulteration

                    3 Replies
                    1. re: JustThinkin

                      Your information is correct, but it is restricted to cancer risk from PCBs, which can be attributed to the 12 dioxin-like (also called co-planar) congeners. PCBS also have serious non-cancer toxicity, however, and that isn't restricted to the co-planars. Also, the TEF values are a subject of some debate in scientific circles currently, as is the dioxin slope factor (slope factor can be thought of as the carcinogenic potency).

                      1. re: FlyFish

                        Thanks for post - I did not fully understand that the toxicity factor was solely based on cancer risk. I was in no way trying to downplay the risk associated with ingesting PCBs, but most of the reporting on farmed salmon is so hyped that you would think it is the only problem with our food supply. Dioxins are nasty and TCDD is very nasty. Over half of the sardines in the test contained TCDD. One can tested at .065 pg/g. Out of 46 farmed salmon that they tested - 9 tested positive for TCDD and only 3 on those were over .015 pg/p.
                        I assumed as other did that because sardines were such a small fish that they must be low in contaminants and had read the KidSafe advice that they were fine. I could fine very little information of PCBs in sardines but after seeing their dioxin levels - I aint going be eatin no more sardines!!!
                        Hopefully you can answer a question. I believe that when I see a report that says farmed salmon has 40 ppb of PCBs and then I see a report that says that wild salmon from Puget Sound has 30 ppb of PBCs that I have learned very little. Ok- they are both high in PCBs, but it is my understanding that PCBs vary greatly in their toxicity so I really know nothing about which fish is better. Am I right or am I missing something? Is the toxicity of PCBs only a factor in cancer risk?

                        1. re: JustThinkin

                          You're absolutely correct - knowing the concentration of total PCBs in a food is only half of the story; the specific mix of congeners that make up that concentration is the other half. That applies to both cancer and non-cancer risk, though perhaps more to cancer risk because virtually all the cancer potency is due to the 12 dioxin-like congeners. The real bad actor is PCB-126 (all of the PCB congeners have been assigned numbers to make talking about them easier), which has a WHO TEF of 0.1, i.e., is 10% as effective as 2,3,7,8-dioxin (also referred to as TCDD) in promoting cancer. As I mentioned before, however, there is ongoing debate about the TEFs and also about the toxicity of dioxin itself.

                          On a related issue that may be of interest, you sometimes see reference to the 2-ppm FDA action level, which is about the only concentration the government has published regarding PCBs in foods, as being a "safe" concentration of PCBs. Nothing could be further from the truth - the FDA level was never toxicity-based and is now considered to be an absolutely screamingly high concentration. Based on risk assessment work that I've personally been involved in, a "safe" level for unlimited consumption, generally regarded as a concentration that would cause one additional cancer in a population of a million consumers, would be at least a couple orders of magnitude lower than that.

                    2. Sorry I am not a scientist!!!! Just stick with King Oscar brand for taste and quality
                      I'm not a fan of BumbleBee and Polar - i find them mushy and flavorless

                      To get your Omega 3's and forget worrying about PCB's:

                      Place 2 sardines- butterflied on 2 slices of lightly toasted whole wheat bread
                      add lettuce, tomato, onion and bacon- add a little mayo .....and enjoy....one of life's little great sandwiches....serve with a lil homemade potato salad, Ba-Tamp-Te kosher pickle, and some sliced red onion and greek olives on the side

                      1 Reply
                      1. re: paulispumonti

                        What .... with a name like paulispumonti what are you doing messing with Norwegian sardines. Get yourself some nice Italian Angelo Parodi sardines. More info here

                        The Great Sardine Taste-off – the Spanish Saga (cans 40-49)

                        I agree with you and Bumble Bee (# 29, 32, ,35, 39) out of the fifty brands I tried. Howver the best that King Oscar did was #17. Haven't tried Polar. Sounds skippable.

                      2. We have a local salvage store "Renys" or Chez Reny for those in the know. I can get off brands of sardines from Malasia and other places 2 0r 3 cans for a buck. We eat them twice a week for breakfast and equally for apps of evening snacks. Used to get cases of 48 King Oscar for 4 bucks from our local Prospect Harbor cannery, before Bumble Bee bought them out.
                        The fishies are so small, I understand that they are low in pcb's and mercury.