Document wrOVDYkE29ZdxedgE3mNKBa46

PCB RESIDUES: THEIR SIGNIFICANCE TO ANIMAL AGRICULTURE C. F. FRIES OLYCHLORINATED biphenyls, more com not recogtlized as widespread environmental con Pmonly knotvn as PCB's, arc chlorinated hydro* taminants until 1966. The major use of PCB's has carbon compounds that have a variety of industrial been as a dielectric in transformers and capacitors. uses. Residues of these compounds have been found Other significant uses have included heat transfer to be widely distributed in the environment. and hydraulic fluids. Minor uses, making up leu These residues have occurred in a number of than 6 percent of sales, have included such items as foods but more frequently in seafoods than any other surface coatings, plasticizers, sealants, adhesives, food product. (Contamination of animal products and printing inks. lias not been widespread, but several significant The sole producer of PCB's in the United States localised problems have arisen with poultry and is the Monsanto Chemical Co.1 whose production of dairy products. these compounds peaked in 1970 at about 36,500 It is my purpose to review the known incidents tons. The company has restricted sales of PCB's to of poultry and milk contamination by PCB's and to use in sealed products such as a dielectric in trans discuss future problems that may arise. formers and capacitors. Foreign producers of PCB's Rackprouiui F* CB's have been produced since 1930 hut were Trad# mnn art uord in tbl* publication aotrlr for tbo purport oI providing tptoilo information. Mention of trade namoa done not eonrtitute guarantor or warrant? of tbo prudurl b? the. U.J. Drpertmrnt of Aplral. turt or on codominant ortr other product* not ntrntiunrd. Third Quarter 1972 19 HONS 069930 iirc m>l a Mgriiluaut factor in U S, u!.:u<r but an* important when considering global contamination problems. |\>|vliloriuntcl biphenyls and polychlorinated triphemls are sold by Monsanto under the trade name Arocloi. They arc described by a four elicit number. The lirst two digits indicate the chemical stiiutun- of 11it: material, while the second two digits indicate (lie percent chlorine that the material contains. For example, Aroclor 1254 is a polychlori nated biphenyl containing 54 percent chlorine. Aroclors arc mixtures but do not contain all of the 210 theoretically possibly polychlorinated bi phenyl compounds. A single formulation may con tain as many as 40 individual PCB's. Chemically and biologically, PCB's are similar to oilier chlorinated hydrocarbon compounds. PCB's can he isolated by the same analytical procedures used in routine surveying for chlorinated hydrocar bon pesticides. Some components have the same gas chiomotographie retention times as chlorinated hy drocarbon pesticides. These interferences led to the discovery and identification of PCB's in the environ ment. Some values listed in the literature as DDT prior to 1966 may have been PCB's. PCB's persist in the environment and resist degra dation by organisms. However, in some species the less chlorinated components are degraded and con verted to water soluble compounds. PCB's are stored in the body fat of animals where the concentrations undergo biological magnification. Thus, animals at the higher levels of the food chain will have the highest residue levels. The toxicological significance of PCB residues has not been determined. The only clinical experience with acute human exposure occurred in Japan in 1968. About 1,000 persons had eaten rice oil con taminated with PCB's. They developed darkened Tins, eye discharge, severe acne, and symptoms of what came to be called Yusho disease. Average intake was 2 grams of PCB, with symptoms appear ing at a minimum (lose of 0.5 g. A variety of subacute effects at much lower in takes have been observed in many other species. The long-trrm significance of these effects and their implications for human health has not been deter mined. Toxicological evaluation of PCB's is difficult be cause they all are mixtures. Proportions of the compmu'iits air different among the various Aroclors. 20 One cannot evaluate a single formulation anil a*, sumo, that the result .applies to other formulations. There is also evidence that PCB's, particulaily those of European ami Japanese manufacture, may inntain highly toxic impurities such .udihenzofuruvs. The PCII ptobleius in the United States imolv mg animal agi icultme have been of tun distinct types The poultry incidents were short-term, high-level exposures causing not only high residues in the prod uct but also adverse effects on the birds. The milk incidents were chronic exposures involving only residues. Milk Rexidue* PcH residues in milk were first detected in 1970 in Ohio. The source was traced to a sealant used on the inside walls in the manufacture of concrete stave silos. Aroclor 1254 was included in the formula of this sealant which was used extensively by one company. All incidents of milk contamination have involved the silos of this company and its successors. Other companies have in their files formulas for PCB-containing sealants. It has not been deter mined whether or not any of these were produced and used. However, no contaminated silos originat ing from other companies have been detected. The sealant containing PCB's was developed in the 1930's and used until about 1968. Estimates of the number of silos in which it w as used are not available. In addition to Ohio, contaminated silos exist in Kentucky, Tennessee, and adjacent States. Scrapings from silo walls contain as much as 1 to 2 percent PCB. While PCB's are not very solu ble in water, their concentrations as high as 100 p.p.m. in the silage dry matter can occur adjacent to the wall. Levels drop rapidly as the distance from the wall increases and significant residues are seldom found more than 2 or 3 feet from the wall. The diffusion towards the center is greater at lower level* in the silo. Despite the relatively low transfer of PCB from the wall to the silage, there are two factors in addition to the high-wall concentrations that con tribute to the milk residue problem. First, it U impractical to separate the high-residue silage ad jacent to the wall from the other silage and second, PCB's arc transferred veiy efficient!) from the silage to the cows' milk. We surveyed a number of farms that were using contaminated silo*. Levels of PCB rrsidurs in milk Agrieufbirnl Scifnec Riiiru HONS 069931 fat writ* approximately 10 p.p.m. when the silage was fed. Other workers have found individual farms with residue level* as high as 28 p.p.m. The Food ami Drug Administration iFD.\> guideline for maximum PCB residue is equiv alent to 2.5 p.p.m. in milk fat. Low-level silage feeding would theoreti cally allow production of milk within the guide lines, but the level is loo low to he practical. We have studied the metabolism of PCB's by cows both in the field and under controlled condi tions in the laboratory. Transfer of PCB's from the feed lo the milk fat and body fat is very efficient. Rates of transfer are approximately equal to the rates for DDE, the most persistent metabolite of DDT. The total diet of the cow must contain Jess titan 0.5 p.p.m. PCB in order to prevent milk from exceeding FDA guidelines. If excessive residues occur in milk the time re quired to return the herd to a marketable condi tion may be quite long. PCB's ate removed from the body fat at a rate proportional to the con centration in the body fat. It takes approximately 6 In 8 weeks to reduce the concentration in milk fat hy one-half. Thus, if the PCB concentration in milk fat is 10 p.p.m. (four times the FDA guideline', it will take approximately 16 wreks from the time (hr contaminated feed is removed from live row's diet for the milk to be within the guideline. Since the PCB concentration in (lie body fat is closely related to that In the milk fat, cows exceed ing the guideline for milk would also exceed the guideline for meat, If the cows are producing milk with excessive PCB levels, one can either destroy the animal or wait until the residue level in milk returns within the guideline. In either rase the loss (n the dairyman can be substantial. The mmtbcv of contaminated *ils delected hy regulatory activity lias been small compared to the estimated number of contaminated silos. This is lamely due to problems of analysis anti survey tech niques that are followed by regulatory agencies. Generally the lower level of detection has been about I p.p.m. in milk fat. Thus, if the milk supply is surveyed using eumposiic sample' of many farms, it would he difficult lo delect an individual farm with contaminated milk if the other farms in the sample produced clean milk. I he amount of I'CB-rnsilaminuird milk produced is iciativrly sm,all comp.iicd to the tola) milk supply of the country. The larins producin': contaminated I'hitrl Qua'lst IU77 milk arc scattered and it is greatly diluted by clean milk. Thus, there is no reason to be particularly concerned about the effects of the residues on the health of the general public. However, there is one segment of the population that could receive significant exposures fioin this source: Farm families who drink milk from their own cows fed on contaminated silage. Using typical residue and milk consumption levels, intakes of PCB for over 15 to 20 years would approximate the level of short-term intakes that caused serious effects in Japan. However, the problem is probably not as serious as implied since a dose consumed over a long period may be less serious than the same dose ronsuined in a very short period. Also the PCB in the con taminated silos was more chlorinated than the Jap anese product which also may have contained highly toxic contaminants. Thus, the situations are not directly comparable. Silo Dmonlaminulion A NY method devised lo correct the PCB problem in milk should have two desirable characteristics: Effectiveness and economy. Because of the high efficiency of transfer of PCB's from a cow's diet to its milk there is no known effective remedy short of elimination of most PCB from the cow's diet. The most effective and most costly method of elimi nating the problem would be complete abandon ment of the contaminated silos. Several alternatives to complete abandonment have been suggested and partially tested, l'itc sug gestions generally involv e resealing the silo to block the diffusion of PCB's from the silage. In some cases the rescaling may be preceded by partial removal of the PCB's. Partial removal of PCB has been attempted bv either sandblasting the surface or washing the sur face with a suitable solvent. Sandblasting followed by rescaling has been tried but no studies of its effec tiveness have been made. The cost of this procedure is greater than half the cost of a new silo and may exceed the depreciated value of the silo. Sandblasting has i nnsequences that have not been lulls explored. The opetalion is duMv and thus it is possible that other areas of the farm eould lx* con taminated. The PCB's also penmate quite deeply into ronriete. It is not known Iku much of the surface would have to he irinovid lo cnmplrtrlv 21 HONS 069932 remove or reduce (he PCB's (o a satisfactory level. Solvent washing has been tried on a few silos, Hitt the cost of the procedure is not established. Pre sumably it would he similar to the cost of sandblast ing. In held experiments we have found that the process of solvent washing and reseating has not been effective. Whether this was due to the specific procedure used has not been determined. Rescaling the silo without prior treatment would be a desirable method because of its lower cost. For this method to be effective the PCB should not pene trate the new sealant and the new sealant must ad here to the walls for the remainder of the silo's life. Several materials have been tested and some show promise. However, the work that has been reported has only covered the first years' experience. It is not known how well these materials will continue to seal. Temporary expedients have been attempted on a fesv farms where contaminated silos were iden tified by regulatory agencies. Lining the interior of the silo with plastic was effective enough to keep the residue levels near the guideline. Discarding the silage within about 6 inches of the wall was less effective. Neither method is compatible with mech anized unloading of the silo and neither eliminates the problem. Poultry SOME of the most spectacular and publicised incidents of PCB contamination have involved poultry. In the past 2 years there have been serious incidents in North Carolina, New York, Maine, and Minnesota. The source of contamination was not established in all cases. , The North Carolina incident was the most exten sive and is the best documented. A large poultry pu>duccr noted a severe decrease in hatchability of eggs. The problem was tracked to fishmeal con taminated by Aroclor 1242. The fishmeal was con taminated during processing when the PCB leaked from a corroded heat exchanger. In addition to this incident, contaminated feed from the same source had been shipped to 11 other States before the situ ation was brought under control. Significant num bers of broilers and eggs had to be destroyed because of the high residue*. Lowered hatchability is one of the more impor tant toxicological effects of PCB's in birds. In con- tr.iu to some of the chlorinated hydrocarbons such 22 as DDT, PCB's apparently have little or no effect on eggshell thickness. The effect on hatchability may occur at levels as low as 10 p.p.m. in the diet. Gen erally, there is full development of the embryo hut the chicks are unable to break the shell. Subdemi.nl edema is the most frequent abnormality. Only the Aroclon with intermediate degrees of chlorination affect hatchability. We found that Aroclor 1232 lias some effect, while the most severe effects are encountered with Aroclon 1242 and 1248. In contrast, Aroclon 1221 and 1254 have no effect at levels as high as 20 p.p.m. in the diet. No significant effects of PCB's on mammalian embryo development have been reported. This may indicate a species difference. However, most mammalian studies have used the more highly chlorinated PCB's and it is possible such effects could be obtained with the less chlorinated compounds. The FDA has recently prohibited use of PCB's in feed plants and the use of recycled paper for feed HONS 069933 containers. Presumably these restrictions should eliminate short-term exposures of the type that have occurred with poultry. However, since PCB's have been so widely used and alt of their possible uses have not been established, it is conceivable that a few incidents of this type will occur in the future. Other /ininicU Products contamination of poultry and dairy products suggests that contamination of other animal prod ucts may have occurred. However, there are circum stances that minimize its significance and chances for detection. It is possible that some cattle raised for beef have received silage from contaminated silos. However, silage, particularly during fattening, is not as im portant a component of beef rations as it is of dairy rations. From levels in milk contamination, it can be assumed that some carcasses of surplus dairy animals have residue levels exceeding the FDA guideline. However, surplus dairy animals account for a small part of the meat supply. There would only be a small chance for a carcass of either type to be selected for routine analysis. The incidents involving poultry suggest that con tamination of feeds for other livestock could also have occurred. Some of the fishmeal from the North Carolina incident may have been used for pigs, yet residues in pork have not been reported. Some of the fishmeal also was ultimately destined for catfish food. To date there have been no revere losses of species other than poultry due to PCB's. This could indicate that other species are not as sensitive to PCB's as poultry and if exposure occurred, (here was no noticeable effect on productivity. IHtcnssion PCI! residues have been accumulating for 40 years hut were not delected until G years ago. It can be assumed that other undetected contamination exists. It usually comes to light by one of two ways. One is the release of high levels of a material and subse quent observation of severe problems, as illustrated by the poultry incidents. The other, and more fre quent, is thr development of analytical instruments and methods capable of detecting smaller residues than was previously possible, as illustrated by the silo incidents. Third Qutitltr 1972 PCB's have the two characteristics of a compound most likely to cause contamination problems. These are resistance to degradation and solubility in fat. Water soluble compounds, even if not degradable, would tend to disperse in the environment and would be of less potential hazard. The poultry feed and silo problems also illustrate some practical problems associated with regulator)' activities. With the poultry feed there was no ques tion of the seventy of the problem. Through an intensive short-term effort, regulatory agencies lo cated all of the contaminated feed and products which were then removed from food marketing channels. The situation involving silos is considerably dif ferent. Mixed samples of milk from several farms rarely show significant residues. Thus, only a small percentage of the affected farms have been identi fied in routine surveys. In order to completely eliminate the problem it would be necessary to sample individual farms. This would involve a very large sample load and is prob ably beyond the capability of most regulatory labora tories. This problem could be overcome to a large extent if a preliminary survey was made and the individual samples were limited to the farms having the suspected type of silo. To my knowledge only Ohio is following such a program at the present time. Regulatory efforts that are not completely effec tive raise a number of questions. The first is whether or not the undetected farms pose a threat to public health. TMfey probably do not because dilution would reduce the residues to a negligible level. How ever, the "dilution principle" has not been accepta ble to many regulatory agencies. The second question is whether or not the oc casional removal of a contaminated dairy from the market provides a sufficient incentive for other dairymen to determine whether they have silos that are contaminated and to correct the problem if it exists. Removal from the market is a major eco nomic loss but the cost of correcting the problem without market loss can also lie substantial. If a dairyman feels the chances of being detected arc very small he may take the chance and ignore the possibility that he has contamination. Problems arc also posed for others, such as re search and extension personnel, who may be aware 23 MOWS 06993** of the nature of the situation hut who have no direct rvtjxWibility for its resolution. They should not condone the production of milk with residue levels above the guidelines. On the other hand, it is diffi cult to recommend a large economic outlay to a dairyman unless it prevents a larger outlay in the future. Also confusing the issue is the determination of responsibility for the contaminated silos. Litigation involving dairymen, the silo builder, and the PCU manufacturer is now in progress. When responsibil ity is established, it should be easier to decide upon and to implement a course of action. ON ANSWERING "WHAT ARE YOU DOING, MR. RESEARCHER?" sometimes imagine myself discussing my own I have worked out a method for rapid identification I (research) project with several kinds of people of heat-tolerant individuals within a breed. These and they each ask the question, "What are you individuals should have better feeding efficiency in doing." our climate, thus making your business more profit Let us say that this question is asked by a fellow able." scientist, a Texas beef cattle producer, a man in To the man in the street I would say, "I am the street, and a member of the legislative Appro working on a scheme which, if successful, will give priations Committee. the consumer better beef at lower cost." To my fellow scientist I would say, "I have a To the legislator I would say, "Texas cattle theory based partially on a feedback control model feeders can use this information to their economic that the regulation of internal temperature in cattle can be studied by treating the hypothalmus as a sensor-controller amenable to testing. To achieve advantage and thus contribute further to our indus trial economy." Now I know all of this sounds kind of awful. . . . such tests I have to install a camula permanently in the anterior hypothalmus to permit repeated trying to be everything to everybody, but every word of what I said, I think, is true and illustrates the entry. Thus the tissue can be heated artificially to idea that if the researcher cannot diseuss these put a step input into the controller or 1 can measure matters with people having all shades of opinion, its temperature change in response to a step change then maybe he shouldn't be doing that kind of work. in environmental temperature. In either case the R. F. Stewart. data for entering the control model can be secured Distinguished Professor of Engineering, and from this the thermal time constant can be Texas A tit M Vniversity derived. Thus a small advance in our knowledge of bovine temperature regulation might be made." To the beef cattle producer I would say, "I believe (From a talk at a workshop on research planning and decision making held in Washington, D C. by the Cooperative State Research Service.) 24 Agricultural Science Review HONS 069935