A Reply to Jepsen, N., K. Aarestrup and S.J. Cooke. Tagging Fish in the Field: Ethical and Procedural Considerations. A Comment to the Recent Paper of D. Mulcahy; Legal, Ethical and Procedural Bases for the Use of Aseptic Techniques to Implant Electronic Devices, (Journal of Fish and Wildlife Management 4:211–219)

I thank Jepsen et al. (2014) for a second opportunity to comment on their position relative to the use of aseptic surgical techniques for the implantation of transmitters into fish. The first opportunity was their publication “testing” whether aseptic technique affected surgical implantation of transmitters into fish (Jepsen et al. 2013), to which I coauthored a response (Mulcahy and Harms 2014). Where possible, herein I will avoid repeating arguments already made in our reply to Jepsen et al. (2013). For clarity, I use italics to identify major quotations from Jepsen et al. (2014) or Jepsen et al. (2013) to which I respond. Quotations for emphasis or from other sources are not italicized.

Jepsen et al. (2014) largely ignore the legal and procedural components of my paper (Mulcahy 2013), and concentrate on the ethical content, specifically on my stance that aseptic surgical technique is the standard of care for transmitter implantation surgeries with few exceptions. Jepsen et al. (2014) extend their criticisms to a previous paper of mine (Mulcahy 2003).

Jepsen et al. (2014) question my qualifications to engage in the debate about the use of aseptic techniques based on an apparent lack of experience, as judged by the lack of my name as a coauthor on fish implantation-surgery papers. They would not do so had I agreed with them. Here are my qualifications: I have a Ph.D. degree in Microbiology and did research on infectious fish diseases for about 15 y. After a further 2 y of involvement with wildlife diseases, I attended veterinary medical school and obtained a Doctor of Veterinary Medicine degree and was soon employed at a federal research center as a wildlife veterinarian. After additional study and professional performance, I attained board-certification in the American College of Zoological Medicine, indicating specialized knowledge and qualification in nondomestic animal medicine and surgery. Most of my work over the past 22 y as a wildlife veterinarian has been with free-ranging birds and mammals (much of it performing thousands of surgeries in the field to implant transmitters), but I have also done fish transmitter-implantation surgeries. I do not believe that merely doing implantation surgeries on a project meets the requirements of coauthorship without further contribution, hence the absence of fish transmitter papers in my vitae. I have authored and coauthored >90 scientific papers on fish, birds, and mammals (including marine mammals), some dealing with surgery, anesthesia, and the effects of transmitter implantations. I have served on three Institutional Animal Care and Use Committees, all of which dealt largely with free-ranging animals. I am knowledgeable about animal welfare, pathogens, infections (including the detection of infections), fish diseases, and surgical techniques (including field surgical techniques).

Jepsen et al. (2014) misquote me as follows: “ ‘The surgical implantation of a transmitters into the coelom of a fish is an inhumane act’ (wma-05-02-10-Mulcahy1Mulcahy 2003 ) is based on moral conjecture and in our opinion does not belong in a valid scientific discussion.” Publication of this inaccurate and inflammatory statement places into the scientific record the idea that I oppose all intracoelomic implantations of transmitters into fish, which is untrue and a gross misdirection of my concerns. There are two places in my article that could have been the source of this misquotation. There, what I actually wrote was, “The surgical implantation of a nonsterile transmitter into any animal is an inhumane act, which should not be performed.” (p. 298), and “The surgical implantation of a transmitter into the coelom of a fish without anesthesia is an inhumane act, which should not be performed.” (p. 299). In neither place can my words be construed to mean that I oppose implantations and I stand by both of the actual statements that I made.

Here is the basic issue concerning the role of aseptic surgery in fish implantation surgeries that reduces the debate to mere argument: no one knows what the postsurgical infection rate is following transmitter implantations in fish. There simply have been no adequate scientific investigations of the prevalence of postsurgical infections in implanted fish. But absence of proof is not proof of absence. Postsurgical infection rates cannot be obtained from one or two studies because they undoubtedly vary widely with fish species, details of the surgical technique, location, water quality, the general health of the fish before implantation, stress levels, season, temperature, and more.

In addition, the argument between Jepsen et al. (2014) and me defies resolution because we are discussing two different things. They cite studies that show no difference in behavior or survival of control fish and fish implanted using “dirty” techniques and claim that proves there is no need for aseptic techniques, but almost none of these studies adequately tested for the presence of infections. I say that postsurgical infections occur at some rate in implanted fish—as they do in all other animal taxa and in humans—but I can find almost no studies that actually and adequately tested for the occurrence of pathogens in the fish following implantation. Fish biologists do nearly all implantation surgeries in fish, but they do not perform the tests required to determine the presence of infections in implanted fish. They cannot claim that infections do not occur.

Jepsen et al. (2014) and others claim that postsurgical infections do not occur or are not important because their fish do not all die and because they cannot see infections with their naked eyes. Jepsen et al. (2013) admit, “However, no bacterial cultures were collected from the fish so we cannot exclude the possibility of a bacterial infection,” but then cite that paper as proof of the lack of benefit of aseptic surgical technique for preventing infections (Jepsen et al. 2014). Those who don't believe in postsurgical infections cite papers showing no difference in behavior or survival of implanted and nonimplanted fish (Bridger and Booth 2003; Cooke et al. 2011). But the possibility for infections in implanted fish is acknowledged in both of these papers: “…and may result in the greatest amount of attachment-associated infection.” (p. 23, Bridger and Booth 2003) and “Any surgical procedure…has the potential to impair health, introduce infection….and even lead to mortality…or indirectly via infection…” (p. 2, Cooke et al. 2011). The references cited in these papers may suggest that massive mortalities or behavioral alterations did not occur or were not detected—at least in the projects that were published. These findings do not address the occurrence of postsurgical infections because observing behavior and survival does not detect infections. The importance of postsurgical infections is not simply whether or not they cause mortality. It includes the costs to the fish caused by sublethal contamination, inflammation, and infection. It is unethical to expect experimental animals to deal with iatrogenic infection and inflammation that could be prevented or greatly reduced by the application of aseptic surgical techniques. At least one of the authors (Cooke) of Jepsen et al. (2014) has strongly supported the use of sterile surgical instruments: “Also recommended is the veterinary practice of surgical instrument sterilization prior to and between surgical procedures on fish” and “In the case of fish surgery, the risk of infection is considered high for any instrument used in the surgical procedure. As such, these tools are considered critical items and require the most stringent of sterilization procedures because any microbial contamination could result in disease transmission.” (Wagner et al. 2011, p. 72). Most of the fish implanted with transmitters are healthy and can usually overcome the iatrogenic introduction of some bacteria during surgery. In fact, the principle of aseptic surgery is that the numbers and types of microorganisms introduced during surgery are reduced to levels that do not cause infection because the patient can deal with them. Jepsen et al. (2014) misconstrue that principle to mean that if aseptic techniques cannot prevent the introduction of all bacteria, then they do not have to use aseptic techniques at all.

My position that the standard of care for the quality of surgeries on fish is at the same level as that for other taxa is held for several reasons. First, the presence of infectious disease in fish is confirmed by the existence of thousands of scientific papers and many books on the subject, and it is inconceivable that postsurgical infections do not occur in fish at some prevalence. Second, most laws, regulations, and societal guidelines mandate the use of aseptic techniques—clarifying this was one of my primary objectives in writing Mulcahy (2013). Third, we must ensure that the data harvested from fish experimentation are valid and are consistent with the reliability of such data from other taxa. Fourth, it is possible to attain a high level of aseptic technique for most fish surgeries in the field, and, given that possibility, it should be done except in the cases of a few carefully considered projects. The asepsis opponents will immediately raise the impossibility of employing aseptic techniques in a project in which thousands of fish have to be implanted in a week's time, but they want to apply the same exception to all projects, even those that might implant only a few fish in a month.

The closest example I can find of an actual characterization of postsurgical infections in transmitter-implanted fish is Walsh et al. (2000), who found infection rates of 46–51% of their fish at high water temperature and 3–5% of fish at low water temperature. Most importantly, Walsh et al. (2000) submitted material from their fish to a laboratory that reported the isolation of bacteria from the samples. Determination of the prevalence and effects of postsurgical infections in fish will require proper necropsies, with some or all of the following: culturing for pathogens; histopathology of organs and tissues; clinical pathology of blood, including white blood cell counts; and determination of biomarkers of inflammation and infection. Those techniques are uniformly absent from purported studies of the usefulness of aseptic techniques to implant fish.

Jepsen et al. (2014) state, “However, when we consider tagging of wild fish, animals are from the same population, living in the same water with relatively free transfer of pathogens. Thus, sterilization of tagging equipment will not prevent transfer of pathogens among fish from the same water.” This quotation highlights the authors' lack of understanding of microbiology and the epidemiology of infectious disease. Not every fish in a body of water necessarily shares every microbe that is present with every other fish. Each pathogen occurs at a prevalence in fish in a population. Despite the statement quoted above, Jepsen et al. (2013) tacitly acknowledge that every fish in a population is not infected with a pathogen and that pathogens can be transmitted between fish from the same population when they cite a paper that demonstrated the transmission of Renibacterium salmoninarum by injection of coded wire tags and transponders into salmon Oncorhynchus tshawytscha (Elliott and Pascho 2001). If all the fish in that population already shared the pathogen, there could not have been transmission.

Jepsen et al. (2014) repeatedly argue that fish are “fundamentally different” to dismiss any comparisons to other taxa of animals. When they say that fish are different, they mention only that fish skin is different from bird and mammal skin in that it is “…sensitive to most chemical disinfectants and sterilants and thus attempts of aseptic practice can be counterproductive or even harmful to fish,” and that drapes cannot be used because of the need to keep fish skin moist (Jepsen et al. 2014). Apparently, they feel that if skin cannot be disinfected and drapes cannot be used then the rest of the components of aseptic technique (sterilized instruments and transmitters, sterile gloves, etc.) that actually enter the coelom can be neglected. If fact, the use of sterile plastic drapes can aid in keeping fish skin moist during surgery and prevent contamination of suture. The paper they cite regarding the sensitivity of fish skin to disinfection actually showed no harm from the use of povidone–iodine (Wagner et al. 1999). I support the statement by Wagner et al. (2011, p. 72): “Fish surgeries should entail as many sterile practices as possible within the limitations of the species and field conditions involved.” The selection of those procedures should begin with the materials that enter the coelom (gloves, transmitters, surgical instruments, suture); materials that do not enter the coelom (e.g., surgery boards) are of less concern.

In their final paragraph, Jepsen et al. (2014) make their main arguments about fish being different: “The fact that fish live (and breathe and eat and defecate) in a pathogen-rich environment and that their integument (including mucus) differs markedly from mammals and birds emphasizes the issues with drawing lessons from other taxa.” I see no logic in the “pathogen-rich environment” being the justification for not using aseptic surgical techniques to implant devices into fish. It seems that Jepsen et al. (2014) are saying that, because implanted fish come from and are returned to a contaminated environment, there is no need to use aseptic techniques because fish are either contaminated before surgery or they will be contaminated when returned to the water after surgery. The American Fisheries Society guidelines essentially say the same thing (American Fisheries Society, American Institute of Fishery Research Biologists, and American Society of Ichthyologists and Herpetologists 2004). In other taxa, these statements would be used as arguments to maximize aseptic technique. I point out that marine birds and mammals also live in the same water environment in which they eat and defecate (and immediately above which they breathe), and yet aseptic surgical techniques are routinely used to implant electronic devices into them. I believe Jepsen et al. (2014) and others dislike comparisons of fish to other animals, not because fish are “different,” but because their basic arguments are not supported by such comparisons.

Jepsen et al. (2014) state: “…“general feelings” or experience from mammal and bird studies, may not be very relevant to field studies on fish and should be refined to become evidence-based.” That statement itself is a “general feeling.” A survey of those active in fish implantation surgeries found that 73% believed using sterile equipment between fish was important and that 78% viewed it important to keep water from entering the incision to avoid introduction of pathogens (Walker and Cooke 2005). Given that the “general feeling” of a large majority of those active in fish implantation surgeries is that at least those aseptic principles are important, I invoke the precautionary principle. I argue that there is insufficient good science to support the neglect of aseptic techniques for fish surgeries (Mulcahy 2003, 2013; Mulcahy and Harms 2014).

Jepsen et al. (2014) argue that, “In fish, it is well-known that “surrogates” should not be used for tagging evaluation studies (wma-05-02-10-Ebner1Ebner et al. 2009 ), so drawing information from the mammalian and avian literature must be done with caution.” Of course, any extrapolation carries the possibility of being in error. However, to cite one paper to justify ignoring all of the accumulated experience of surgeries done on birds and mammals (including humans) lacks substance as an argument and does not rise to the level of “it is well-known”.

Jepsen et al. (2014) make several statements that I question. First, “…without the relevant veterinary training…” This statement wrongly suggests that I am opposed to anyone without veterinary qualifications performing transmitter implantations in fish. I did not question who performs the surgeries, but rather only the standard of care used in those surgeries. They make the next two statements in defense of the status quo of fish surgery, because it is a “…popular, productive, and well-described method…” and is “…widely used as the basis for important management decisions.” Improper techniques do not acquire legitimacy by how often they are used, the perception of their productivity, the quality of their descriptions, or by how often the resulting data are used by managers.

The asepsis opponents repeatedly state that there is no evidence that aseptic techniques are necessary in fish surgeries, but the experiments they cite are marked by small sample sizes, an inability or refusal to actually use aseptic surgical techniques, and an inability or unwillingness to use proper techniques to detect the presence of infections in implanted fish (see Mulcahy and Harms 2014). Jepsen et al. (2014) state that “…in most European Union countries, no person will be allowed to implant fish without following intensive courses including aseptic techniques.” Despite such training, they were unable to produce an adequate simulacrum of aseptic technique in the work they reported (Jepsen et al. 2013), and so they did not test what they purported to test (Mulcahy and Harms 2014). Further, if their required “intensive courses” included aseptic techniques, was that not an implication that the authorities requiring the classes expected them to be used? Both Chomyshyn et al. (2011) and Jepsen et al. (2013) report that their fish suffered from uncharacterized or minimally characterized “water quality problems” that caused 10–50% mortality in their experimental and control groups. The investigators cannot assume that their measurements were unaffected by something in their system that could cause sufficient physiological disruption as to produce death in sizable proportions of their experimental animals. I support the call by Cooke et al. (2011) for an increase in experimental rigor and design with adequate statistical power for evaluation of all aspects of fish surgeries.

Jepsen et al. (2014) challenge several of my statements: “We would also argue that statements such as, ‘few biologists have been formally trained in aseptic techniques,’ ‘I maintain that biologists find it difficult to place the concept of asepsis into practice in their work because of confusion about what constitutes aseptic technique, a lack of surgical knowledge and training,’ ‘Biologists do not know what microorganisms persist on the disinfected instrument and devices they use,’ and ‘The privilege of using animals in research is accompanied by an obligation to minimize their pain and distress’ are not very fruitful and are potentially counterproductive to a constructive scientifically based debate about aseptic vs. nonaseptic tagging techniques. The first three quotes could equally apply to all personnel performing implantation on fish (including veterinarians).” The quotation about the privilege of using animals in research and the obligation to minimize their pain and distress is a basic tenet of the welfare of research animals. They are wrong to say that the first three statements could equally apply to veterinarians. Veterinarians have been formally trained in aseptic techniques, and they do know how to place the concept of asepsis into practice in their work. The third comment (“Biologists do not know…”) is incomplete as quoted and is taken out of context. It was meant to reinforce the difference between “aseptic” and “sterile” in the context of fish surgery, and appeared in a more general discussion about the misuse of terms such as “sterilized” and “disinfected.” I invite the reader to view the paragraph as originally published (p. 212, Mulcahy 2013). I have no doubt that Jepsen et al. (2014) have observed veterinarians with little idea of how to implant a fish. It is not part of classical veterinary training, which covers fish only very briefly, if at all. I did not raise the question of who should do fish implantation surgeries.

Jepsen et al. (2014) state, “We will not discuss the legal issues, given that they extend beyond our realm of expertise, not to mention that they vary widely among jurisdictions.” Dismissing the legal requirements for doing research (including transmitter implants) on animals is unwise. All of us, even fish biologists, are required to follow the laws and regulations governing our jurisdictions; and clarification of those requirements, at least for the United States and Canada, was one of the goals of my paper (Mulcahy 2013). The laws, regulations, and guidelines statements quoted therein are written in clear language, not in legalese, and are easily understandable. Before writing our commentary (Mulcahy and Harms 2014) on Jepsen et al. (2013), we attempted to obtain pertinent documentation from the Danish Experimental Animal Committee that they mentioned, but we could find nothing online. Because much of the literature opposing asepsis in fish surgeries seems to originate from only a couple of laboratories in the United States and Canada, I focused on the laws governing those jurisdictions (Mulcahy 2013). That local animal care and use committees have allowed the use of “dirty” surgical techniques for transmitter implants in fish in jurisdictions where the use of such techniques violates legal regulations is an indictment of the animal care and use committees, not an endorsement of such techniques.

Jepsen et al. (2014) state that Chomyshyn et al. (2011) “revealed” that attempts at aseptic technique “…doubled the time…” for the surgery, but they do not mention that the extra time amounted to only 60 s. Likewise, Jepsen et al. (2013) express concern about the greater length of time required by their version of aseptic technique compared with their traditional method (8 min vs. 4 min), but they admit that there was no difference in the actual fish handling time. The increase came from the time spent by the surgeon preparing for the next surgery. From an animal welfare standpoint, it is only the time the animal spends in surgery that matters, not the time the surgeon spends between surgeries, and the surgery times would have to be much longer before there are grounds for concern (see Mulcahy and Harms 2014). Jepsen et al. (2014) state, “This is not a matter of cost, as suggested by wma-05-02-10-Mulcahy2Mulcahy (2013),—it is a matter of evidence.” I was not the one who questioned the affordability of aseptic surgical techniques. For example, Chomyshyn et al. (2011) compared expenses and came up with a capital cost of CN$2,135/fish using the aseptic technique vs. CN$95/fish using the nonaseptic surgical technique. The CN$2,135 figure was reached by including the cost of an autoclave in the costs for every fish in a single small project, without amortization over the course of multiple fish in multiple projects over multiple years. This comparison does not rise to the level of “evidence.” Others mention cost as an issue (e.g., Cooke et al. 2013) or dismiss or fail to mention the option of purchasing sufficient instrument sets as an alternative (e.g., Walker et al. 2013a, 2013b).

The arguments Jepsen et al. (2014) and others present do not support a claim that aseptic surgical techniques are not needed or cannot be done for fish. Aseptic surgical techniques as the standard of care for fish surgeries is a position supported in most venues by laws and regulations, and by most professional society guidelines. What is needed next are not more attempts to prove that aseptic techniques are not required for fish surgeries, but the use of aseptic techniques in good-quality research to improve the overall effects of capture, handling, holding, and surgery that are all parts of fish biotelemetry projects.