A Review of Nonsteroidal Anti‐Inflammatory Drugs for Food‐Producing Animals With a Focus on Potential Applications for Farmed Finfish

ABSTRACT

In finfish aquaculture, there are several inflammatory diseases impacting productivity and animal welfare, however there are limited options available to veterinarians to treat inflammation and pain in fish. Nonsteroidal anti‐inflammatory drugs (NSAIDs) are widely used in terrestrial animals raised for human consumption to treat a range of diseases and to manage pre‐ and postoperative pain as they offer anti‐inflammatory, antipyretic and analgesic properties. This review examines the use of NSAIDs in terrestrial food animals with the pharmacological and safety learnings to the possible use in aquaculture, specifically for oral administration of farmed salmonids. Across Norway, United Kingdom and Australia, which are salmon producing regions burdened by inflammatory disorders such as complex gill disease, seven NSAIDs are authorised for use in terrestrial food‐producing animals. The NSAIDs considered in this review have a good safety profile for users and the targeted terrestrial animal species, but more research is needed to develop safe and effective treatment regimens for farmed fish. Some NSAIDs have been shown to reduce pain sensitivity in fish but there is no published data demonstrating that they reduce inflammation. However, the targeted immune pathways of NSAIDs appear conserved between fish and terrestrial animals so it is highly likely that NSAIDs will have anti‐inflammatory effects in fish if they can be delivered effectively through in‐feed treatments. This review informs prescribing fish veterinarians and fish health professionals, as well as researchers, national licensing authorities and other stakeholders, to assist with consideration of the use and potential applications of the NSAIDs in finfish aquaculture.

1. Introduction

Acute inflammation is an essential and coordinated immune response to infection and/or injury (Chen et al. 2018). A targeted and rapid inflammatory response can contain infection and activate innate and adaptive immunity, and tissue repair (Eming et al. 2017; Sun 2024). However, inflammation is detrimental if the response goes beyond what is necessary to counter the immediate threat or it is not resolved, leading to excessive tissue damage, prolonged recovery, or in severe cases, sepsis and mortality (Adams 2022; Broom and Kogut 2018; Griffith et al. 2016). Inflammatory diseases in animals raised for human consumption are widespread and have diverse causes (Broom and Kogut 2018). Some examples of economically significant diseases include porcine respiratory disease complex (PRDC) caused by infectious organisms, either bacteria, viruses, or both, in combination with husbandry practices and impacting the swine industry worldwide (Assavacheep and Thanawongnuwech 2022). Mastitis, inflammation of the mammary glands, is a multifactorial disease that is highly prevalent and costly for the dairy industry (de Vliegher et al. 2018). Nutritional imbalances, such as high carbohydrate in fish, low vitamin B in pigs and low selenium in chickens, can also cause inflammation in farmed animals (Chen et al. 2022; Wu et al. 2016; Zhao et al. 2021). In addition to the impact of disease on animal welfare, subclinical cases of chronic inflammation should be managed, as this immune dysfunction is costly to production, suppressing appetite and diverting nutrients away from productive purposes such as growth and milk or egg yields (Gouvea et al. 2022; Przyborowska et al. 2024; Shuster et al. 1991).

Aquaculture is an increasingly important source of protein globally, accounting for more than 50% of aquatic animal production (FAO 2024). Salmonids are the most valuable traded aquatic species (20% of total aquaculture value) and the fourth most produced finfish based on annual tonnes of live weight harvested (FAO 2024). Such large‐scale production is challenged by inflammatory diseases caused by both infectious and noninfectious agents and is of significant concern for salmon health, productivity and animal welfare. Some of the major diseases with inflammation as a primary pathology in salmonid aquaculture are heart and skeletal muscle inflammation (HSMI), dermatitis, cardiomyopathy syndrome (CMS), enteritis and complex gill disease (CGD) (Boerlage et al. 2020; Kristoffersen et al. 2013). To date, there are limited options available to prescribing veterinarians to treat inflammation and pain in aquaculture species (Sladky 2023). Developing treatments that provide symptomatic relief of inflammatory diseases would be a particularly valuable tool for multifactorial diseases, such as CGD, where treatment targeting just one causal factor (such as via vaccine or antimicrobial/antiparasitic) often does not resolve disease, or provide symptomatic relief.

Gill disease is a significant health challenge in salmon aquaculture (Mitchell and Rodger 2011; Oldham et al. 2016; Rozas‐Serri 2019). There are regional differences in the presentation and description of marine salmonid gill disease, but there appears to be some related syndromes (Boerlage et al. 2020). Complex gill disease (CGD) is described in the European salmon producing countries such as Norway, Ireland and Scotland. Chilean salmon aquaculture often refers to proliferative gill disease (PGD), which is a non‐specific term describing gross gill lesions and is encompassed in the definition of CGD according to Boerlage et al. (2020). In Australia, amoebic gill disease (AGD) is the primary gill issue, however other proliferative gill lesions unrelated to AGD do occur (Clark and Nowak 1999). Complex gill disease (CGD) is a multifactorial marine disease with no clear single aetiology (Herrero et al. 2018). Several pathogenic organisms have been associated with CGD including Desmozoon lepeophtherii, salmon gill poxvirus, Candidatus Branchiomonas cysticola and Neoparamoeba perurans (Boerlage et al. 2020; Gjessing et al. 2019). Noninfectious gill insults such as gelatinous zooplankton (hydroids, siphonophores and jellyfish) and harmful algal blooms can also contribute to CGD and in some cases may precede infection by pathogenic organisms (Davidson et al. 2009; Ferguson et al. 2010; Rodger et al. 2011). Co‐infections involving two or more of these agents are common and histopathological interpretation can be challenging (Gjessing et al. 2019). Alongside a variety of histological changes in the gills such as epithelial cell hyperplasia, necrosis and mucous cell hyperplasia, local inflammation indicated by subepithelial leukocytes or filament branchitis, is often described in cases of CGD (Gjessing et al. 2019; Herrero et al. 2018; Noguera et al. 2019). Since this is a consistent finding, it is a promising target for symptomatic treatment. Aside from freshwater or hydrogen peroxide bath treatments which are labour intensive, expensive and not always effective, there are no other treatments for marine gill disease currently available (Parsons et al. 2001; Wynne et al. 2020).

Nonsteroidal anti‐inflammatory drugs (NSAIDs) are widely used in food‐producing animals for their anti‐inflammatory, antipyretic and analgesic properties (Reader et al. 2020; Schoos et al. 2019). There is a range of NSAID formulations authorised to treat several diseases and manage pre‐ and postoperative pain in terrestrial food animal species. Benefits of NSAIDs have been demonstrated in the management of mastitis in cows (Fitzpatrick et al. 2013), lameness in cattle, chickens and pigs (Hadipour et al. 2011; Mustonen et al. 2011; Thomas et al. 2015), diarrhoea in calves (Todd et al. 2019) and respiratory disease in cattle and pigs (Georgoulakis et al. 2005; Guzel et al. 2010). For infectious diseases such as acute respiratory infection in pigs and cattle or bovine mastitis, NSAIDs are often used as adjunctive therapy alongside antimicrobial treatments as they help recovery by reducing pain sensitivity, fever and inflammation (Fitzpatrick et al. 2013; Friton et al. 2006; Thomas et al. 2015). Ultimately, the benefits of short‐term treatment regimens (1–5 days) in terrestrial food‐producing animals include reduced mortality, maintained feed intake and improved pain management, which are significant productivity and animal welfare outcomes (van Soest et al. 2018; Todd et al. 2019). Since NSAIDs have proven effective with established safety profiles in meat‐producing animals across multiple terrestrial farmed species, they are worthy of investigation for applications in farmed fish.

In‐feed NSAID treatment of inflammatory diseases in fish could be a non‐invasive method to provide symptomatic relief and increase recovery rates. While few veterinary medicinal products have been developed specifically for farmed fish, a wider range could be available under the veterinary cascade, including some NSAIDs (Loeb 2019). The efficacy of NSAIDs as anti‐inflammatories in fish is not yet well understood, but the targeted immune pathway of NSAIDs appears conserved between a diversity of animals (Chandrasekharan and Simmons 2004). Reduced pain sensitivity has been demonstrated in fish, where, for example, ketoprofen ameliorated the effects of a noxious stimulus in goldfish ( Carassius auratus ) (Ward et al. 2012) and carprofen decreased the time to resume feeding in rainbow trout ( Oncorhynchus mykiss ) injected with acetic acid as a painful stimulus (Mettam et al. 2011). There is a long history of research into the use of NSAIDs in terrestrial food animals and insights into their pharmacodynamics, pharmacokinetics and safety are needed to guide dosing, formulations and risk assessments in fish. If proven effective and safe in salmonids, oral administration of select NSAIDs has the potential to improve recovery from disease that has a strong inflammatory response, such as complex gill disease (CGD), and reduce mortality rates, thereby improving performance and animal welfare during high‐risk periods of CGD.

The aim of this review was to examine the use of NSAIDs in food‐producing animals including fish and investigate the pharmacological and safety learnings to the possible use in finfish aquaculture, with an emphasis on salmon marine gill disease. It summarises the NSAID products that are already approved for veterinary use in food animals across the salmonid producing regions of Norway, United Kingdom and Australia, and the pharmacokinetics of these drugs in fish, poultry, pigs, cattle and sheep. Safety risks for each drug are documented, including safety to the target animals, the people administering the treatment, food and environmental safety. This review is expected to contribute to the background information required to assess the anti‐inflammatory and analgesic properties, as well as the safety of NSAIDs in farmed fish. It will inform fish veterinarians prescribing under the cascade system and provide background to licensing authorities and other stakeholders assessing the potential use of such medicines in finfish aquaculture.

2. Pharmacodynamics of NSAIDs Approved for Food Producing Animals

There are over 17 different NSAIDs used in veterinary medicine globally but not all are approved for use in animals that produce products consumed by humans (e.g., meat, milk, eggs) (Edwards 2021). To expand the use of NSAIDs into aquaculture, it is advantageous to trial the drugs that are already approved for use in food‐producing animals. Such drugs have established safety profiles, with defined maximum residue limits and withdrawal periods to ensure that drug residues in meat products are below safe levels for consumption. The approved medicines are already licensed in different jurisdictions and manufactured to a set quality standard (i.e., stability, consistent formulation) making it easier to upscale a reliable treatment option if effective in salmonids. Ultimately, using existing drugs will lower development costs and allow faster market access compared to developing novel drugs.

This review focusses on three Atlantic salmon ( Salmo salar ) producing regions: Norway, United Kingdom (UK) and Australia, all of which are burdened with marine gill disease and would greatly benefit from additional treatment options. Table 1 summarises NSAIDs that are approved for treatment of food‐producing animals as listed by each region's veterinary medicine regulatory agency as of June 2025 (Norwegian Medicines Agency https://www.legemiddelsok.no/sider/default.aspx, UK Veterinary Medicines Directorate https://www.gov.uk/check‐animal‐medicine‐licensed, Australian Pesticides and Veterinary Medicines Authority https://portal.apvma.gov.au/pubcris). The summary includes the number of injectable and oral products by region and the prescribed applications and treatment regimen for each species listed in the product information sheets. The complete list of NSAID products currently available for each species and region is provided in Table S1.

TABLE 1.

NSAIDs approved for use in food‐producing animals in Norway, United Kingdom, and Australia.

Region	Approved NSAIDs	No. of products	Species	Applications	Dose (mg/kg)	Treatment regimen
Norway	Meloxicam	22 inject	Cattle	Respiratory infection, diarrhoea, acute mastitis, postoperative pain	0.5	Single SC or IV injection
			Pigs	Lameness and inflammation, puerperal septicaemia and toxaemia	0.4	Single IM injection, if required a 2nd dose can be given after 24 h
		2 oral	Pigs	Lameness and inflammation, puerperal septicaemia and toxaemia	0.4	Mixed with small quantity of feed, if required a 2nd dose after 24 h
	Ketoprofen	4 inject	Cattle	Mastitis, fever associated with respiratory disorders	3	IV or deep IM injection once daily for up to 3 days
			Pigs	Fever in respiratory disorders, postpartum dysgalactia	3	Deep IM injection once a day for up to 3 days
		1 oral	Cattle	Inflammation and reduction of fever in individual animals	4	Orally once daily for 1–3 days
	Flunixin	3 inject	Cattle	Respiratory disease, endotoxemia, mastitis, lameness, postoperative pain	2.2	IV or IM injection once daily for up to 3 days
			Pigs	Respiratory disease, postpartum dysgalactia, lameness, postoperative pain	2.2	IV or IM injection once daily for up to 3 days
	Sodium salicylate	1 oral	Cattle	Respiratory disease, fever	40	In drinking water once daily for 1–3 days
			Pigs	Anti‐inflammatory agent, respiratory disease	35	In drinking water once daily for 1–3 days
United Kingdom	Meloxicam	24 inject	Cattle	Respiratory infection, diarrhoea, acute mastitis, relief of post‐operative pain	0.5	Single SC or IV injection
			Pigs	Lameness and inflammation, puerperal septicaemia and toxaemia	0.4	Single IM injection, if required a 2nd dose can be given after 24 h
		2 oral	Pigs	Lameness and inflammation, puerperal septicaemia and toxaemia	0.4	Mixed with a small quantity of feed, if required a 2nd dose after 24 h
	Ketoprofen	12 inject	Cattle	Post‐partum inflammation/pain, lameness, respiratory disease, mastitis	3	IV or IM injection for 1–3 days
			Pigs	Respiratory disease, postpartum dysgalactia	3	Single SC or IV injection, if required a 2nd dose can be given after 24 h
		1 oral	Pigs	Respiratory disease, fever	3	In drinking water for 1 day, 1–2 additional administrations if required
	Carprofen	4 inject	Cattle	Mastitis, respiratory disease, fever	1.4	Single SC or IV injection
	Flunixin	12 inject	Cattle	Respiratory diseases, endotoxemia, mastitis, lameness, postoperative pain	2	Once daily by IV or IV injection for 1 to 3 days
			Pigs	Mastitis, respiratory disease, fever	2	Once daily by IV or IM injection for 1 to 3 days
	Sodium salicylate	2 oral	Cattle	Respiratory disease in calves	40	Orally once daily for one to three consecutive days
			Pigs	Inflammation	35	Per day orally for three to five consecutive days
			Turkeys	Inflammatory respiratory diseases	86.2	Orally daily for three consecutive days
	Acetylsalicylic acid	1 oral	Pigs	Fever	30	Orally twice daily for 3 days
Australia	Meloxicam	26 inject	Cattle	Analgesic, anti‐inflammatory, diarrhoea, respiratory disease, mastitis	0.5	Single SC or IV injection
			Pigs	Puerperal septicemia, toxaemia, lameness	0.4	Single IM injection, if required a 2nd dose can be given after 24 h
			Sheep	Analgesic, anti‐inflammatory	1	Single SC injection
		2 oral	Cattle	Analgesic, anti‐inflammatory, lameness, mastitis, surgery	0.5–1	Once orally or immediately before surgery procedure
			Sheep	Analgesic post‐surgery	1	Once orally immediately before the procedure
	Ketoprofen	10 inject	Cattle	Anti‐inflammatory agent, colic	3	IV or IM injection for 1 to 3 days
	Flunixin	16 inject	Cattle	Analgesic, anti‐inflammatory, antipyretic, lameness, respiratory disease	1.1–2.2	IM injection daily for 3 to 5 days
			Pigs	Agalactia, analgesic, anti‐inflammatory, antipyretic, mastitis, metritis	1.1–2.2	Daily for up to 3 days, by deep IM injection
	Tolfenamic acid	4 inject	Cattle	Anti‐inflammatory, mastitis, pneumonia	2 4	IM injection for pneumonia, may be repeated once after 48 h. Single IV injection for mastitis
			Pigs	Agalactia, anti‐inflammatory, mastitis, metritis	2	Single IM injection

There are no NSAIDs currently approved for use in fish in the countries reviewed, but several are available for cattle and pigs and a few for sheep and turkeys. In total, seven NSAIDs can be prescribed for food animals in Australia, Norway and the UK with only meloxicam, ketoprofen and flunixin approved across the three regions considered in this review. Most NSAID products are available in injectable formulations and there are a few powder formulations available for oral administration and include meloxicam, ketoprofen, sodium salicylate and acetylsalicylic acid (also known as aspirin). Bespoke formulations including these NSAIDs can also be made by veterinary compounding pharmacies. The NSAIDs are often prescribed to reduce pain and inflammation in combination with antimicrobial, antiparasitic or antiviral therapy for infectious conditions, such as respiratory infections. Treatment regimens for each drug and species generally align between different regions. Some products have multiple active ingredients, for instance an NSAID combined with an antibiotic (i.e., florfenicol, ceftiofur, tulathromycin or oxytetracycline). These products are included in the list of products (Table S1) but are unlikely to be prescribed in the context of CGD due to differences in disease aetiology and the focus to reduce antibiotic use in order to combat potential development of antimicrobial resistance.

All the NSAIDs listed in Table 1 have similar mechanisms of action and work by inhibiting cyclooxygenase (COX) enzymes. In mammals, inflammation is controlled by the conversion of arachidonic acid into prostaglandins by two main paralogous COX enzymes, COX‐1 and COX‐2 (Smith et al. 1998). Prostaglandins are essential in triggering inflammatory responses and their biosynthesis is significantly increased in inflamed tissue (Hahn et al. 1999). While the role of COX genes in the inflammatory response is well established in many mammals, our understanding of prostaglandin biology in fish remains comparatively limited. Available evidence suggests that fish cells, particularly leukocytes and macrophages, can produce prostaglandin E (PGE) in response to diverse pathogen‐associated molecular patterns (PAMPs), often in combination with calcium ionophores (reviewed by Gómez‐Abellán and Sepulcre (2016)). In some cases, increased prostaglandin production has been associated with the upregulation of COX‐2 expression following lipopolysaccharide (LPS) stimulation (Boltaña et al. 2014). However, in Atlantic salmon macrophage‐like SHK cells, the relationship between COX‐2 expression and prostaglandin production appears more complex, with PGE₂ downregulating COX‐2 via a negative feedback loop (Fast et al. 2005). Consistent with a central role for COX‐2, LPS‐induced PGE production is suppressed in the presence of COX‐2 inhibitors (Zou et al. 1999). Collectively, these findings suggest that the core functions of COX pathways are likely conserved across fish species and thus remain a logical target for therapeutic intervention with NSAIDs.

Although numerous studies have examined the structure and regulation of COX genes in fish, topics largely beyond the scope of this review (see Wang et al. 2016), it is noteworthy that most salmonids possess at least two COX‐2 isoforms, commonly referred to as cox‐2a and cox‐2b (Ishikawa and Herschman 2007; Olsen et al. 2012). In both rainbow trout and Atlantic salmon, these isoforms appear to be inducible under distinct conditions. Further gene duplication has been identified in Atlantic salmon, with two paralogs of cox‐2a (cox‐2a1 and cox‐2a2) reported, of which only cox‐2a2 is inducible by bacterial PAMPs (Abushweka et al. 2026). In the gill, cox‐2b shows the highest expression among COX genes, suggesting a potentially prominent role in this tissue and highlighting it as a candidate therapeutic target in the context of gill stress or injury.

Different NSAIDs have different selectivity for COX enzymes and the therapeutic and toxic properties of NSAIDs directly relates to the type and amount of prostaglandin production that is inhibited (Curry et al. 2015). Many NSAIDs non‐selectively inhibit both COX‐1 and COX‐2 and other preferentially bind to one isoform (Edwards 2021). Blocking COX‐2 is generally considered the target anti‐inflammatory pathway because it is the inducible isoform that is primarily involved in inflammation (Curry et al. 2015; Morita 2002). Whereas COX‐1 prostaglandins are constitutively expressed in many tissues and play key roles in various physiological processes such as gastrointestinal protection, haemostasis and renal function, in addition to some involvement in inflammation (Morita 2002; Sakamoto 1998). If COX‐1 suppression is too high, it increases the risk of unwanted side effects such as gastrointestinal ulcers (Becker et al. 2004). Alternatively, high COX‐2 selective inhibitors can increase cardiovascular risks such as thrombosis due to disruption of prostacyclin–thromboxane balance (Funk and Fitzgerald 2007). For example, in humans, meloxicam preferentially inhibits COX‐2 over COX‐1 which is why it has a much lower risk of causing gastrointestinal ulcers (Del Tacca et al. 2002), whereases ketoprofen nonselectively inhibits both COX‐2 and COX‐1 creating a higher risk of gastrointestinal and renal side effects (Afrin et al. 2023). The selectivity of NSAIDs for COX‐1 and COX‐2 is primarily controlled by the chemical structure of the NSAID molecules that bind to the active COX enzyme sites and the strength of these bonds, which can differ between animal species (Blobaum and Marnett 2007). For instance, carprofen is COX‐2 selective in cats but nonselective in horses (Brideau et al. 2001). This highlights the need for species‐specific pharmacological assessments when assessing NSAIDs. The COX‐selectivity of each NSAID in salmonids in unknown. It also remains unclear whether selective COX‐2 inhibitors target different COX‐2 isoforms with equal efficacy. A more comprehensive understanding of the regulation and functional roles of these isoforms will be critical for optimising the therapeutic application of NSAIDs in fish.

3. Medicated Feeds for Finfish Aquaculture

One of the major differences in any proposed use of NSAIDs in finfish aquaculture compared to farmed terrestrial animals is the strong preference for oral administration over the injectable route of administration (Table 1). Medicated feed treatments are advantageous for aquaculture because they are a less laborious method to simultaneously treat a large group of animals without handling stress. The route of administration greatly impacts NSAID pharmacokinetics, particularly in the absorption phase, when drugs cross the gastrointestinal tract into the bloodstream (Coskun et al. 2022; Pairis‐Garcia et al. 2013; Uney et al. 2021). NSAIDs are weak acids with medium to high lipid solubility, so they are considered good candidates for oral administration as they should absorb efficiently through cell membranes in the gastrointestinal tract (Lees et al. 2004). However, NSAIDs are often bitter (Yoshida et al. 2014) and may require masking flavours, bitter blockers or technologies such as microencapsulation to overcome palatability issues. NSAID diet formulations that are both palatable and efficiently absorbed have not yet been developed for fish. Once proven clinically effective, the formulation of medications into fish diets is a specialised and important area of research that can maximise the efficacy in target hosts while reducing the environmental impact by minimising leakage from pellets.

There are several examples of in‐feed medications that are effectively delivered to and benefit the health of farmed fish. Emamectin benzoate (SLICE, MSD Animal Health) has been routinely used as oral in‐feed treatment by the salmon industry for over 25 years. After administering emamectin benzoate to farmed Atlantic salmon at 50 μg/kg body weight for 7 days, parasitic sea lice ( Lepeophtheirus salmonis ) abundance reduced by an average of 94% at 21 days post treatment (Ramstad et al. 2002). Antibiotics like florfenicol are also delivered as in‐feed treatments of bacteria diseases such as tenacibaculosis (Irgang et al. 2021) and piscirickettsiosis (San Martín et al. 2019) in salmonids and are generally administered over short periods of up to 10 days. Many factors must be considering when preparing the correct dosage in medicated feed, including the average body weight of the fish population, number of fish in that population, their feed intake rate which changes with life‐stage (Azevedo et al. 2025) and health status (Damsgård et al. 2004) and the concentration of the active ingredient in the premix to be incorporated within the feed.

The in‐feed dose and treatment length of NSAIDs for management of gill disease in salmonid can only be hypothesised with our current level of understanding. Short courses of NSAIDs (1–5 days) are common in terrestrial food producing animals (Table 1) however our current understanding of NSAID pharmacokinetics and the long‐term effects of exposure in salmonids remains limited. Ideally this intervention will transiently shift the immune response to promote gill repair and prevent chronic inflammation, without detrimentally supressing the hosts inflammatory response. Depending on the diagnosis of gill disease there may be different treatment options in terms of medicine, dosage and duration and whether repeat treatment is needed. The stage of the disease in the pathogenesis will likely be an important consideration. For example, jellyfish damage causes acute inflammation initially, characterised by a predominant neutrophilic infiltrate, and then can develop into a more chronic lesion (Marcos‐López et al. 2016). However, several pharmacokinetics, pharmacodynamics and target‐animal safety studies are required before NSAID treatment regimens can be optimised for farmed fish species.

4. Pharmacokinetics

Pharmacokinetics considers the absorption, distribution, metabolism and elimination of a drug. Understanding how the drug moves through the body helps determine optimal dosing regimen, and encompasses the dose, route, interval and duration of drug exposure. By comparing parameters like the maximum concentration a drug reaches in the blood (Cmax) or the half‐life (the time it takes for the active ingredient to reduce by half), it helps predict how long therapeutic concentrations may be sustained. Pharmacokinetics differ depending on drug, species, life‐stage, administration route and temperature for ectotherms (Baert and De Backer 2003; Rairat et al. 2019; Uney, Corum, Durna, et al. 2024; Warner et al. 2020), however, interspecies comparisons are a useful starting point for animals like fish where no approved NSAID products exist. This data provides an initial range for estimating safe and effective doses in alternate species.

To begin to understand the pharmacokinetics of NSAIDs in fish, intravenous (IV) data provides the baseline with the most direct understanding of how drugs behave in the body, free from the complexities of absorption. IV pharmacokinetics are also required to calculate oral bioavailability, that is the fraction of an orally administered drug that reaches the systemic circulation in an active form. Table 2 summarises pharmacokinetic parameters of the seven food‐animal NSAID candidates administered intravenously and orally in fish (including elasmobranchs), chickens, pigs, cattle and sheep. It is important to note that all the studies with oral administration of NSAIDs to fish were via gastric gavage which bypasses factors like palatability and feed intake rates, which is critical in the context of medicated feeds. Additionally, drug absorption can change significantly when given with or without food (Moore et al. 2015) so the gavage studies may not be reflective of the pharmacokinetics of an NSAID‐feed treatment.

TABLE 2.

Comparative pharmacokinetic parameters of nonsteroidal anti‐inflammatory drugs (NSAIDs) administered via intravenous (IV) and oral routes in five animal species (fish, chickens, pigs, cattle, sheep). Key parameters include maximum plasma concentration (C_max), time to reach C_max (T_max), area under the plasma concentration–time curve (AUC), elimination half‐life (T¹/₂), mean residence time (MRT), volume of distribution (V_d), total body clearance (Cl) and oral bioavailability (F). All values are representative of single‐dose pharmacokinetics. Oral administration was via gastric gavage or syringed into the mouth (oral) or in feed treatments (oral food). Dashes (−) indicate parameters not reported or not applicable.

NSAID	Species	Route	Dose (mg/kg)		Cmax (μg/ml)	Tmax (h)	AUC* (h*μg/ml)	t 1/2 (h)	Vd (ml/kg)	Cl (ml/h/kg)	MRT (h)	F (%)	Adverse reaction	References
Meloxicam	Nile tilapia	IV	1		—	—	11.26	1.36	127	88.5	1.43	—	Not rep	(Fredholm et al. 2016)
	Rainbow trout	IV	1		7.97	—	21.47	3.63	200	50	4.24	—	No	(Corum et al. 2022)
	Nursehound shark	IV	0.5		2.05	—	11.37	10.71	670	48.6	15.01	—	Not rep	(Morón‐Elorza et al. 2022)
	Rainbow trout	Oral	1		0.84	1.78	4.61	2.95	—	—	5.13	21	No	(Corum et al. 2022)
	Tilapia	Oral	1		0.07	2	0.4	1.91	—	2494.81	3.59	—	No	(Martin et al. 2021)
	Nursehound shark	Oral	0.5		Not dete	—	—	—	—	—	—	—	Not rep	(Morón‐Elorza et al. 2022)
	Yellow stingrays	Oral	2		0.42	13.12	17.36	15.46	—	—	—	—	No	(Kane et al. 2022)
	Chicken	IV	0.5		—	—	40.8	3.21	58	13	4.41	—	No	(Baert and De Backer 2003)
		IV	0.5		—	—	20.2	3.2	117	25	4.41	—	No	(Kris Baert and De Backer 2002)
		IV	1		—	—	38.63	5.45	206.5	26.48	5.63	—	No	(Shao et al. 2022)
		Oral	1		3.04	3.08	27.09	5.79	—	—	8.46	70	No	(Shao et al. 2022)
		Oral	1		7.21	2	37.92	2.79	—	—	—	—	No	(Souza et al. 2017)
	Pig	IV	0.4		3.28	—	8.04	2.7	190	61	3.5	—	Not rep	(Fosse et al. 2008)
		IV	0.5		—	—	13.27	6.15	160	—	4.26	—	No	(Pairis‐Garcia et al. 2015)
		Oral food	0.5		1.07	2.4	11.61	6.83	—	43.08	9.01	87	No	(Pairis‐Garcia et al. 2015)
	Cattle	IV	0.5		—	—	82.34	20.35	171	6	28.19	—	No	(Coetzee et al. 2009)
		IV	0.5		—	—	237.4	79.2	243.8	2.15	114	—	Not rep	(Jokela et al. 2024)
		Oral	0.5		1.77	10	230.5	84.8	—	2.32	122	97	Not rep	(Jokela et al. 2024)
		Oral	1		3.1	11.64	164.46	27.54	—	—	44.9	100	No	(Coetzee et al. 2009)
	Sheep	IV	0.5		—	—	49.26	14	179	10.2	17.6	—	No	(Stock et al. 2013)
		Oral	1		1.72	19	75.09	15.4	—	—	31.3	72	No	(Stock et al. 2013)
Ketoprofen	Rainbow trout	IV	3		24.07	—	135.69	3.91	120	22.11	5.64	—	No	(Greene et al. 2020)
	Chicken	IV	3	R—	—	—	3.75	0.16	200	960	0.29	—	Not rep	(Neirinckx, Croubels, De Boever, et al. 2011)
				S+	—	—	1.58	0.21	580	1980	0.23	—	Not rep
		Oral	3	R—	1.36	0.2	1.15	0.6	—	—	0.96	32	Not rep	(Neirinckx, Croubels, De Boever, et al. 2011)
				S+	0.83	0.19	0.85	0.58	—	—	1.06	53	Not rep
	Pig	IV	3	R—	—	—	4.71	0.39	410	690	0.63	—	Not rep	(Neirinckx, Croubels, De Boever, et al. 2011)
				S+	—	—	38.39	2.53	300	80	3.51	—	Not rep
		Oral	3	R—	3.85	0.27	4.34	0.78	—	—	1.28	86	Not rep	(Neirinckx, Croubels, De Boever, et al. 2011)
				S+	7.26	0.58	31.3	2.23	—	—	3.85	84	Not rep
		IV	3		—	—	33.86	2.66	350	94	3.4	—	Not rep	(Raekallio et al. 2008)
		Oral	3		5.09	1.25	32.04	3.52	—	—	5.54	97	Not rep	(Raekallio et al. 2008)
		Oral	6		12.03	1.19	69.51	3.22	—	—	5.15	105	Not rep	(Raekallio et al. 2008)
	Cattle	IV	3	R—	—	—	9.49	0.42	200	330	1.69	—	Not rep	(Landoni et al. 1995)
				S+	—	—	9.49	0.42	220	330	2.67	—	Not rep
	Sheep	IV	3	R—	—	—	10.54	0.63	110	160	0.77	—	Not rep	(Arifah et al. 2001)
				S+	—	—	5.82	0.63	180	280	0.64	—	Not rep
Carprofen	Rainbow trout	IV	2.5		14.71	—	159.49	30.66	600	20	38.55	—	No	(Uney et al. 2021)
		Oral	2.5		2.52	4	125.46	41.08	—	—	58.78	79	No	(Uney et al. 2021)
	Pig	IV	4		15.11	—	620.07	36.34	342	6.86	—	—	Not rep	(Gómez‐Segura et al. 2022)
	Cattle	IV	0.7		—	—	294.3	30.7	91	2.4	—	—	No	(Lohuis et al. 1991)
	Sheep	IV	4		—	—	2025	43.36	121.36	1.98	61.45	—	No	(Coskun et al. 2022)
		Oral	4		15.9	11.73	1277	46.71	—	—	70.04	63	No	(Coskun et al. 2022)
Flunixin	Rainbow trout	IV	2.2		—	—	322.03	8.37	55.81	6.83	8.17	—	No	(Uney, Corum, Durna, et al. 2024)
		Oral	2.2		6.24	4	82.9	8.76	—	—	12.38	26	No	(Uney, Corum, Durna, et al. 2024)
	Chicken	IV	1.1		—	—	118.6	5.45	84	10	6.66	—	No	(Kris Baert and De Backer 2002)
		IV	1.1		—	—	236.9	5.52	43	9	6.66	—	No	(Baert and De Backer 2003)
	Pig	IV	2.2		—	—	30.44	7.06	2900	540	5.4	—	No	(Kittrell et al. 2020)
		IV	2.2		—	—	21.63	6.29	300	—	3.01	—	No	(Pairis‐Garcia et al. 2013)
		Oral food	2.2		0.95	1	4.84	7.08	—	—	5.58	22	No	(Pairis‐Garcia et al. 2013)
		Oral	3.3		4.88	0.93	31.01	11.38	—	—	8.33	99	No	(Kittrell et al. 2020)
	Cattle	IV	2.2		—	—	20	5.2	782	115	6.4	—	Not rep	(Odensvik 1995)
		Oral food	2.2		0.9	3.5	12	6.2	—	—	12.7	60	Not rep	(Odensvik 1995)
	Sheep	IV	1		—	—	—	3.83	166.2	36	—	—	Not rep	(Welsh et al. 1993)
		IV	2		—	—	—	3.43	151.8	42	—	—	Not rep	(Welsh et al. 1993)
		Oral food	4		1.78	6	37.68	7.95	—	—	13.59	—	Not rep	(Marini et al. 2016)
Tolfenamic acid	Rainbow trout	IV	2		—	—	70.65	3.47	90	30	3.11	—	No	(Corum et al. 2024)
		Oral	2		1.24	2	17.68	9.19	—	—	13.34	25	No	(Corum et al. 2024)
	Cattle	IV	2		—	—	8.66	2.5	980	—	2.57	—	Not rep	(Landoni et al. 1996)
	Sheep	IV	2		—	—	7.81	1.97	410	260	1.59	—	No	(Corum et al. 2018)
		Oral	2		1	4	8.85	5.13	—	—	8.09	107	No	(Corum et al. 2018)
Sodium salicylate	Chicken	IV	25		—	—	124.9	3.13	950	210	4.59	—	No	(Baert and De Backer 2003)
		IV	50		—	—	1047.7	4.04	390	70	4.59	—	No	(Baert and De Backer 2002)
		IV	50		—	—	941	3.35	320	50	6.08	—	Not rep	(Poźniak et al. 2013)
		Oral	50		96.93	1.64	946	2.92	—	—	6.95	101	Not rep	(Poźniak et al. 2013)
	Cattle	IV	50		—	—	—	0.54	240	—	—	—	Not rep	(Gingerich et al. 1975)
		IV	50		—	—	201.7	0.62	200	258	0.89	—	not rep	(Kotschwar et al. 2009)
	Sheep	IV	10		2.29	—	1.42	0.46	5070	9290	0.66	—	Not rep	(Mathurkar et al. 2018)
		IV	50		17.05	—	14.94	0.54	2860	3890	0.79	—	Not rep	(Mathurkar et al. 2018)
		IV	100		20.82	—	25.95	0.74	4160	3990	1.07	—	Not rep	(Mathurkar et al. 2018)
		IV	200		27.72	—	47.11	1.16	7190	4520	1.67	—	Not rep	(Mathurkar et al. 2018)
		Oral	100		4.22	—	12.26	1.89	12,480	4620	2.69	53	Not rep	(Mathurkar et al. 2018)
		Oral	200		8.27	—	24.45	1.9	16,450	5790	2.75	64	Not rep	(Mathurkar et al. 2018)
Acetylsalicylic acid	Chicken	IV	50		—	—	15.4	0.25	780	3350	0.24	—	Not rep	(Poźniak et al. 2013)
		Oral	50		6.9	0.54	5.09	0.39	—	—	0.78	35	not rep	(Poźniak et al. 2013)
	Cattle	IV	26		—	—	106	0.51	199.5	263.9	0.81	—	Not rep	(Whittem et al. 1996)
		Oral	50		20	3	—	3.7	—	—	—	—	Not rep	(Gingerich et al. 1975)
		Oral	100		47	3	—	3.7	—	—	—	—	Not rep	(Gingerich et al. 1975)
	Sheep	IV	20		—	—	106.9	0.52	202.1	261.1	0.81	—	Not rep	(Ali 2003)

^*AUC_0–∞ was reported by default when available; if not reported, AUC_0–last was used instead.

One of the main physiological differences in fish that will affect the pharmacokinetics of NSAIDs compared to other food‐producing animals is their ectothermic nature. Body temperature and hence metabolic rate are largely determined by the environmental temperature. Salmonids are cool‐water species with a much slower basal metabolic rate compared to tropical fish and mammals, which can lengthen half‐life due to slower enzyme activity metabolising the drugs (Brett 1972; White and Kearney 2013). While there are no published examples of how temperature impacts NSAID pharmacokinetics in fish, this concept is well documented for orally administered antimicrobials (Rairat et al. 2019; Sohlberg et al. 1994; Zanuzzo et al. 2022). In Atlantic salmon held at 12°C, 16°C, and 20°C and fed a dose of either tetracycline or florfenicol, the plasma and muscle residues were lower in fish held at the higher temperatures for both antibiotics (Zanuzzo et al. 2022). Several factors could have influenced this result, with temperature influencing gut transit rates (Mock et al. 2022) and intestine membrane fluidity (Ge et al. 2021) in salmonids which could alter gastrointestinal absorption. Temperature can also alter plasma protein conformation impacting drug‐protein binding affinity (Cunha et al. 2024; Xu et al. 2023), as well as enzyme activity which affects metabolism of the molecule (Andersson and Koivusaari 1985) and renal or branchial excretion rates (Rairat et al. 2019).

There are several other physiological factors that influence pharmacokinetics in fish. Gastrointestinal physiology will affect how a drug is absorbed, with salmonids having a less compartmentalised and shorter digestive tracts adapted to higher protein diets compared to terrestrial invertebrates (Løkka et al. 2013). For instance, drug stability and solubility will be affected by stomach pH, which is an average of 4.8 in Atlantic salmon (Krogdahl et al. 2015) compared to pigs and chickens ranging 1.2 to 4 (Henze et al. 2021; Lee et al. 2021). Gut microbiota also influences how certain drugs are metabolised, and there are large interspecies differences (Ikeda‐ohtsubo et al. 2018; Maseda and Ricciotti 2020). Elimination of pharmaceuticals is typically renal and/or hepatic in mammals, while in fish, the gills can serve as an additional site for systemic waste removal (Cravedi et al. 1993; Pichai and Lakshmanan 2019). Finally, rainbow trout of different body sizes (25 vs. 108 vs. 227 g) that were orally administered 10 mg/kg bw the antimicrobial danofloxacin had a higher Cmax and longer half‐life with larger body size (Uney, Corum, Marín, et al. 2024). Hence, NSAID pharmacokinetic could differ with body size and/or life stages of Atlantic salmon (smolt vs. market size adult), as seen in juvenile and adult sheep (Coskun et al. 2023) and pigs (Millecam et al. 2019).

The health status of the host could also influence NSAID pharmacokinetics, as it does with antimicrobials administered to fish (Shiry et al. 2019; Uno 1996; Wang et al. 2015). For example, orally administered florfenicol in rainbow trout infected with Lactococcus garviea had lower oral bioavailability and increased total body clearance compared to healthy controls (Shiry et al. 2019). While rainbow trout infected with Streptococcus iniae in the same experiment had a higher Cmax and lower volume of distribution compared with the healthy controls (Shiry et al. 2019). This demonstrates the influence of disease on pharmacokinetics in fish and shows these differences can be pathogen specific. NSAID pharmacokinetics can also be impacted by the health or inflammatory status of vertebrate hosts (Ikuta et al. 2017; Lohuis et al. 1991; van den Ouweland et al. 1987). Carprofen IV administered to cows with Escherichia coli endotoxin‐induced mastitis had lower systemic clearance and longer elimination half‐life compared to the healthy control cows (Lohuis et al. 1991). Different diseases and inflammatory statuses do not appear to affect pharmacokinetics parameters in a consistent manner, hence studies begin in healthy animals and then can be extended to the target disease models.

To date, there are no publications that show what concentration of NSAIDs in the blood is clinically effective at reducing inflammation in fish. Most of the terrestrial food animal pharmacokinetics studies listed in Table 2 dose the animals at an amount close to what is recommended by the pharmaceutical companies (see Treatment Regimen in Table 1). If the assumption is made that therapeutically effective plasma concentrations align between species, then the maximum plasma concentration can be compared to see if NSAIDs are being sufficiently absorbed at clinically relevant concentrations in fish. However, given differences in animal physiology, this could be too presumptive. Separate efficacy studies are required in fish. The remainder of Section 3 outlines inter‐species comparisons of the major pharmacokinetics parameters for the seven NSAIDs of interest based on the data collated in Table 2.

4.1. Meloxicam

Meloxicam is the most studied NSAID in fish to date (Table 2). The half‐life of meloxicam is highly variable across species, ranging between 1.36 to 84.8 h (h) in the animals covered in this review (Table 2). The half‐life was shortest in finfish (1.36 to 3.63 h) compared to elasmobranchs (10.71 to 15.46 h) chicken (3.2 to 5.79 h), pigs (2.7 to 6.83 h), cattle (20.35 to 84.8 h) and sheep (49.26 to 75.09 h) which suggests finfish quickly eliminate meloxicam from the bloodstream. Indeed, fish had the fastest clearance rate of meloxicam, ranging from 49 to 2495 mL/h/kg compared to other species with a rate between 2.15 to 43 mL/h/kg. Meloxicam is prescribed orally in farmed animals and has high oral availability in chickens, pigs, cattle and sheep of between 70 to100%. However, in rainbow trout ( Oncorhynchus mykiss ) given 1 mg/kg by gavage, the oral bioavailability compared to 1 mg/kg administered intravenously was only 21% (Corum et al. 2022). The peak plasma concentration (Cmax) of orally administered meloxicam was also relatively low in rainbow trout and tilapia compared to the terrestrial species (Corum et al. 2022; Martin et al. 2021). A low Cmax aligns with poor absorption or alternatively could indicate the drug is distributed rapidly into the tissue (Tozer and Rowland 2006). The lower peak plasma concentration, faster clearance and shorter half‐life means overall exposure to meloxicam in fish is lower compared to other food producing animals, that is, low area under the curve (AUC). While meloxicam demonstrates low oral bioavailability and low systemic exposure in fish, which is generally considered a suboptimal pharmacokinetic profile, these limitations are not necessarily prohibitive. Adjusting the dose, increasing dosing frequency, or developing more efficient in‐feed or sustained‐release formulations may help achieve therapeutic concentrations.

4.2. Ketoprofen

Ketoprofen is marketed as a racemic (50:50) mixture of R(−) and S(+) enantiomers (Ghezzi et al. 1998). The two enantiomers have the same chemical formula, but their three‐dimensional arrangements are mirror images of each other. These structural differences alter pharmacodynamics, toxicity and pharmacokinetics (Ghezzi et al. 1998). It is unknown which form is the most therapeutically active in teleosts, but in Nile tilapia ( Oreochromis niloticus ) injected with 8 mg/kg intramuscularly, fish plasma was exposed to the S‐enantiomer much more than the R‐enantiomer (Greene et al. 2020). Several studies report the pharmacokinetics of each enantiomer because it can provide information about whether one enantiomer is more responsible for the therapeutic and/or toxic effects. Differences between S‐ and R‐enantiomer pharmacokinetics can be due to differences in absorption or clearance, or chiral inversion from R‐ to the S‐enantiomer, which can occur in mammals (Aberg et al. 1995; Neirinckx, Croubels, De Boever, et al. 2011; Neirinckx, Croubels, Remon, et al. 2011).

Across all species considered, ketoprofen had a relatively short half‐life compared to the other NSAIDs, which aligned with a moderate to high clearance rate and short mean residence time. Rainbow trout dosed with 3 mg/kg of ketoprofen IV had the longest half‐life at 3.91 h and the slowest clearance rate of 22 mL/h/kg (Greene et al. 2020) compared to a range of 80 to 1980 mL/h/kg when considering chickens, pigs, cattle, and sheep (Arifah et al. 2001; Landoni et al. 1995; Neirinckx, Croubels, De Boever, et al. 2011; Neirinckx, Croubels, Remon, et al. 2011). Rainbow trout also had the highest total drug exposure over time, indicated by the higher AUC (Greene et al. 2020). Oral bioavailability of ketoprofen appears variable across species, being high in pigs (84%–100%) (Neirinckx, Croubels, De Boever, et al. 2011; Neirinckx, Croubels, Remon, et al. 2011; Raekallio et al. 2008) but low to moderate in chickens at 32% and 53% for the R‐ and S‐enantiomers, respectively (Neirinckx, Croubels, De Boever, et al. 2011; Neirinckx, Croubels, Remon, et al. 2011). The oral bioavailability of ketoprofen in fish is unknown.

4.3. Carprofen

The number of published carprofen pharmacokinetics studies is more limited compared to the other NSAIDs considered, possibly because veterinary use of carprofen in food animals is only approved in the UK for cattle. Of the NSAIDs considered, carprofen had the longest half‐life ranging from 30 to 47 h, which was a similar duration across the different species (Table 2). It had prolonged elimination from the body, with slow clearance rates less than 20 mL/h/kg and a mean residence time (MRT) greater than 38 h. This drug profile could extend the duration of action but may also lengthen the withdrawal periods required to reach residue limits. The apparent volume of distribution (Vd) of carprofen in rainbow trout was 600 mL/kg, which was the largest for finfish of the seven candidate NSAIDs (Uney et al. 2021). This Vd is also greater than values reported for carprofen administered to pigs, cattle and sheep (Coskun et al. 2022; Gómez‐Segura et al. 2022; Lohuis et al. 1991). A higher Vd indicates the drug has a higher propensity to distribute from blood into body tissue, which is the case with drugs that are not highly bound to plasma proteins and more readily cross cell membrane (Lees et al. 2004; Lin et al. 1987). NSAIDs generally have low Vd because they have high protein binding ratios (Lees et al. 2004). The protein binding ratio of carprofen in fish is unknown. Oral bioavailability of carprofen in rainbow trout was reported high at 79%, showing it was well absorbed by the gastrointestinal tract (Uney et al. 2021). The long half‐life of carprofen and good oral availability in fish theoretically makes this a suitable drug profile for in‐feed treatment.

4.4. Flunixin

Flunixin is the pharmacologically active substance but flunixin meglumine is a form often used in veterinary products. Flunixin meglumine is a salt compound that is more water‐soluble than flunixin alone and once administered it dissociates to release flunixin. The half‐life of flunixin was of moderate duration compared with the other NSAIDs and similar between species (Table 2), from 3.4 to 11.4 h, with rainbow trout within this range at approximately 8 h for both IV and oral administration (Uney, Corum, Durna, et al. 2024). The Vd was smallest in rainbow trout at 56 mL/kg in contrast to cattle with a Vd of 782 mL/kg (Odensvik 1995; Uney, Corum, Durna, et al. 2024). Clearance rates varied across each species (6.83 to 540 mL/h/kg) and rainbow trout was the slowest to clear flunixin from the blood. The total flunixin exposure in rainbow trout dosed with 2.2 mg/kg was significantly reduced when administered orally, indicated by an AUC of 332 and 83 h*μg/ml for IV and oral gavage respectively (Uney, Corum, Durna, et al. 2024). Indeed, oral bioavailability was only 26% which could mean the flunixin is not easily absorbed within the salmonid gut. Flunixin oral availability seems highly variable even within the same species, as 22% (Pairis‐Garcia et al. 2013) and 99% (Kittrell et al. 2020) have been documented in separate pig studies. However, the differences are likely caused by Pairis‐Garcia et al. (2013) dosing adult pigs with NSAID‐laced food and Kittrell et al. (2020) dosing piglets via oral gavage without food.

4.5. Tolfenamic Acid

Tolfenamic acid is a structural analogue of flunixin (Edwards 2021), but the pharmacokinetics are not as well documented in food‐producing animals. From the limited studies available, the half‐life of tolfenamic acid is short when IV administered across species (1.97 to 3.47 h) but significantly increases with oral administration in rainbow trout (IV T¹/₂ trout = 3.5 h oral T¹/₂ trout = 9.2 h) (Corum et al. 2024). In rainbow trout, the peak plasma concentration of tolfenamic acid was only 1.24 μg/mL at 2 h post oral administration; then it was slowly eliminated, corresponding to a long mean residence time of 13.34 h (Corum et al. 2024). Tolfenamic acid had a low Vd in trout at 90 mL/kg compared to cattle at 980 mL/kg and sheep at 410 mL/kg (Corum et al. 2018; Landoni et al. 1996). The lower Vd indicates the drug is more likely to stay within the blood rather than distributing into tissue, reducing the likelihood of the drug reaching local inflammation in peripheral tissue, although this may be beneficial for a well perfused organ such as the gills. However, the low oral availability (25%) and low plasma concentration in orally administered rainbow trout may indicate that this is an unsuitable drug profile for in‐feed treatments.

4.6. Sodium Salicylate

Sodium salicylate is a sodium salt of salicylic acid that rapidly dissociates after administration (EMA 2024). Hence salicylic acid is the active agent measured in the blood for pharmacokinetics studies administering sodium salicylate. To our knowledge, there are no sodium salicylate pharmacokinetics studies in fish. From the studies in chickens, cattle and sheep, sodium salicylate has a short half‐life (0.46 to 4.04 h) compared to the other NSAIDs reviewed (Table 2). In sheep it has a very large Vd, over 2860 mL/kg in all studies and is cleared quickly from the body at rates faster than 3890 mL/h/kg (Mathurkar et al. 2018). Oral availability varies, with sheep absorbing 53 to 64% and chickens 100% (Mathurkar et al. 2018; Poźniak et al. 2013). In chickens, oral salicylate was rapidly absorbed from the gastrointestinal tract, with a Cmax of 96.93 μg/mL at 1.64 h post administration (Poźniak et al. 2013).

4.7. Acetylsalicylic Acid

Acetylsalicylic acid is the acetylated form of salicylic acid and has a similar pharmacokinetics profile as sodium salicylate in terms of rapid absorption and short half‐life across multiple species, with sheep and cattle having very similar profiles (Table 2). Compared to sodium salicylate in ruminants, acetylsalicylic acid has a much smaller Vd and slower clearance rate. There are no pharmacokinetics studies of acetylsalicylic acid administered in fish.

5. Safety

Safety, in the context of expanding the use of NSAIDs into finfish aquaculture, refers to the assurance that these drugs do not pose significant harm to the target animal being treated, the people administering the treatment, the consumers of the animal‐derived products and the environment.

5.1. Target Animal Safety

Target animal safety evaluates the risks of the veterinary drug when administered to the species for which it is intended. NSAIDs have been approved for treatment of cattle and pigs in the UK for decades, and during this period several NSAIDs have proven effective and safe. Despite this long track record of safety, a few side effects are possible. From the Summary of Product Characteristics (SPC) documents that accompany each marketed NSAID, adverse reactions are listed for four of the seven NSAIDs covered here, but in all cases the frequence of occurrence is rare, impacting between < 1 and 10 animals per 10,000 animals treated (APVMA https://portal.apvma.gov.au/pubcris; EMA https://medicines.health.europa.eu/veterinary/en/search‐medicines; (NMA n.d.) https://www.legemiddelsok.no/sider/default.aspx; (VMD n.d.), https://www.vmd.defra.gov.uk/ProductInformationDatabase/). Gastrointestinal irritation or ulcers can occur in cattle and pigs treated with ketoprofen, flunixin, sodium salicylate and acetylsalicylic acid, with increased risk after repeated administration according to the SPC documents. NSAIDs that inhibit COX‐1 can cause gastrointestinal issues because this enzyme is responsible for producing prostaglandins that protect the stomach lining from the corrosive effects of stomach acid (Becker et al. 2004). The risk of renal disorders is listed in the SPC for ketoprofen and flunixin. NSAIDs that select for COX‐1 and COX‐2 can have renal side effects because both isoforms play a role in kidney function (Davies et al. 2006). COX‐1 inhibition increases sodium excretion followed by a decrease in blood pressure, whereas COX‐2 inhibition promotes sodium retention and increases blood pressure in mammals (Drożdżal et al. 2021). Prolonged bleeding can also occur after the use of sodium salicylate and acetylsalicylic acid as salicylate may inhibit normal blood clotting, but this effect is reversible within 7 days (Undas et al. 2007). Aside from injection site swelling, no adverse reactions are listed for meloxicam, carprofen and tolfenamic acid in cattle and pigs in the SPC documents.

While some gastrointestinal and renal side effects infrequently occur in mammalian food‐producing animals, this does not mean such adverse reactions will occur in salmonids because COX selectivity can differ between species (Brideau et al. 2001). There are very few published adverse reactions in fish injected or orally dosed with the food‐approved NSAIDs. Of those, the analgesic effect of carprofen was tested in rainbow trout and while it ameliorated some effects of a noxious stimulus, it had the unwanted side effect of making the fish less active 3 h post treatment (Mettam et al. 2011). However, the carprofen dose was high, at 5 mg/kg administered intramuscularly, compared to most other studies that injected terrestrial animals with between 0.7 and 4 mg/kg carprofen (Table 2) and prescribed dose of 1.4 mg/kg for cattle. Salicylates (such as sodium salicylate) have been shown to impair the adaptive cortisol response to acute stress in trout (Gravel and Vijayan 2007), although the trials were conducted with high levels of active ingredient. Of the six pharmacokinetic studies on fish summarised in Table 2 that reported an adverse reaction, none observed any significant side effects, indicating low to no acute toxicity at the doses studied. Additionally, in rainbow trout undergoing surgery with flunixin, there were no reported negative effects in terms of skin healing of incisions when assessed through histopathology (Rizzo et al. 2017). Though there are a limited number of studies on the treatment of fish with NSAIDs, and of the examples available, many do not specifically investigate acute or chronic toxicity. More research is required to confirm safe and effective dosing regimens in farmed fish.

5.2. User Safety

User safety relates to the protection of people handling the pharmaceuticals such as veterinarians, technicians and farmers. The main hazards listed on the SPC documents common to the seven NSAIDs are possible irritation to the skin, eyes and respiratory tract. During preparation and mixing of the product, it is recommended to wear gloves, safety glasses and a dust mask to avoid direct contact. Ketoprofen carries an additional warning that the product may cause dizziness and drowsiness and to avoid self‐injection and dermal exposure. Pregnant women should be particularly cautious with meloxicam and flunixin as laboratory studies have shown evidence of foetotoxic effects. People with known hypersensitivity (allergy) to NSAIDs should avoid contact.

5.3. Food Safety

Consumer safety must be assessed for all pharmacologically active substances, including NSAIDs, which are intended for use in food‐producing animals. To manage the public health risk of pharmaceuticals that remain in the final food product, regulatory agencies set withdrawal periods and maximum residue limits (MRLs) to ensure the amount of residue is below the amount safe to consume daily over a lifetime without health risks. In Europe, the European Medicines Agency (EMA 2025) Committee for Veterinary Medical Products (CVMP) is responsible for recommending MRLs and withdrawal periods. These recommendations become legally binding food safety standards when adopted by the European Commission. While Norway is not a member of the European Union (EU), it is part of the European Economic Area and conforms to EU legislation concerning food safety. In Australia, the Australian Pesticides and Veterinary Medicines Authority (APVMA 2025) is responsible for the scientific assessment and setting of MRLs and withdrawal periods. These limits are then incorporated by the Food Standards Australia and New Zealand into the food standards code that can be enforced by State and National law.

An MRL is the highest amount of an agricultural or veterinary chemical residue legally allowed in food products. Separate MRLs are set for all major food‐producing species (i.e., cattle, pigs, chickens, sheep, salmonids) for each drug approved for treatment in that animal. The MRLs for the seven NSAIDs considered in this review for Europe and Australia are listed in Table 3. In some cases, the pharmacologically active substance is metabolised into a different form and that metabolite serves as the marker residue to be quantified. For example, carprofen can undergo glucuronidation, hence the marker residues are the sum of carprofen and carprofen glucuronide conjugate (Rubio et al. 1980) (Table 3). Currently no MRLs exist for NSAIDs in salmonids. For veterinary use, MRLs can be extrapolated from ‘major’ to ‘minor’ species (e.g., sheep to goat), however, salmonids are considered a major species so the existing MRLs in cattle and pigs cannot be applied to aquaculture. If NSAIDs prove to be effective at aiding the management of inflammatory disease(s) in salmonids, separate studies will be required to generate the safety and residue data required to establish MRLs, using the scientific guidelines provided by the EMA (EMA 2025a).

TABLE 3.

Maximum residue limits (MRLs) for food animal approved NSAIDs in Norway, United Kingdom and Australia. All MRLs are listed in μg/kg.

Active substance	Marker residue	Norway and United Kingdom	Australia
Meloxicam	Meloxicam	Bovine, caprine, porcine Muscle 20 Liver 65 Kidney 65 Milk 15	Cattle kidney 200 Cattle liver 100 Cattle meat * 10 Cattle milk 5 Pig fat/skin 100 Pig kidney * 10 Pig liver * 10 Pig meat 20
Ketoprofen	Ketoprofen	Bovine and porcine no MRL Poultry muscle, liver, kidney 10 Poultry skin and fat 30	Cattle, edible offal * 50 Cattle meat * 50 Cattle milk * 50
Carprofen	Sum of carprofen and carprofen glucuronide conjugate	Bovine muscle 500 Bovine fat 1000 Bovine liver 1000 Bovine kidney 1000	Not listed in Food Standards Australia and New Zealand Schedule 20
Flunixin	Flunixin	Bovine muscle 20 Bovine fat 30 Bovine liver 300 Bovine kidney 100 Porcine muscle 50 Porcine fat and skin 10 Porcine liver 200 Porcine kidney 30	Cattle kidney 20 Cattle liver 20 Cattle meat (in the fat) 20
Tolfenamic acid	Tolfenamic acid	Bovine and porcine muscle 50 Liver 400 Kidney 100 Milk 50	Cattle kidney * 10 Cattle liver * 10 Cattle meat 50 Cattle milk 50 Pig kidney * 10 Pig liver 10 Pig meat * 10
Sodium salicylate	Salicylic acid	Bovine and porcine no MRL Turkey muscle 400 Turkey skin and fat 2500 Turkey liver 200 Turkey kidney 150 Poultry (not turkey) muscle 400 Poultry (not turkey) skin and fat 250 Poultry (not turkey) liver 500 Poultry (not turkey) kidney 1000	Not listed in Food Standards Australia and New Zealand Schedule 20
Acetylsalicylic acid	Salicylic acid	No MRL required	Not listed in Food Standards Australia and New Zealand Schedule 20

^*Indicates that the maximum residue limit is set at the limit of determination.

Withdrawal periods are the minimum time after the last drug administration before the animal can be harvested to ensure that residues are below the established MRL. The withdrawal period is set individually for each veterinary product as part of the marketing authorisation process. They are species‐specific and vary with different administration routes, different drug formulations, the type of food product (e.g., meat or milk) and regional legislation. For example, the meat withdrawal period for oral administration of meloxicam is 5 days for pigs in Europe, 10 days for sheep in Australia and 14 to 35 days for cattle in Australia depending on the product used. Since there are no withdrawal periods specified for the use of NSAIDs in salmonoids, new residue depletion studies will be required to establish these limits. The EMA offers scientific guidelines on determination of withdrawal periods for edible tissues (EMA 2022).

5.4. Environmental Safety

In‐feed NSAID treatments of farmed fish could introduce residues into the water and sediment through uneaten feed and fish waste. These residues must be managed so they do not accumulate to concentrations that affect nontarget organisms, such as wild fish, invertebrates or algae. Ideally the input of NSAIDs is less than the rate of natural attenuation of pharmaceuticals through processes such as microbial degradation, photodegradation and hydrolysis (Mohapatra et al. 2020). NSAIDs are already extensively used by human and veterinary medicine and are detected in water bodies around the world (Lin et al. 2023; Świacka et al. 2021; Wilkinson et al. 2017). The most frequently detected NSAIDs in environmental water are acetaminophen, diclofenac, ibuprofen, indometacin, naproxen, mefenamic acid, ketoprofen and salicylates (Castro‐Pastrana et al. 2020). Of these NSAIDs frequently detected in environmental water, three are candidate treatments of inflammatory diseases in salmonids: ketoprofen and salicylates (which are the active constituents of sodium salicylate and acetylsalicylic acid).

Recently the European Medicines Agency assessed the environmental safety of substances used in veterinary medicine that are authorised by the EU for use in food‐producing terrestrial animals and may be used in food‐producing aquatic species (EMA 2025b). This environmental assessment considered the hazards listed in the European Chemical Agency, United States Environmental Protection Agency, safety data sheet and scientific literature for each chemical. Based on these documents, a risk to the environment was not identified for carprofen and tolfenamic acid. However, an environmental risk could not be excluded for meloxicam, ketoprofen, flunixin, sodium salicylate or acetylsalicylic acid. This does not mean the latter NSAIDs cannot be used safely in aquaculture under well‐defined and managed treatment regimens. Rather, it serves as an early indicator that these NSAIDs require more in‐depth review of their acute and chronic toxicity to nontarget organisms.

Of the NSAIDs considered in this review, ketoprofen is most often cited as a pharmaceutical pollutant of significant environmental concern (Tyumina et al. 2023; Wang et al. 2018). It has been detected in many water bodies globally, with high environmental detection of 260 μg/L in South India, 159 μg/L in South East Africa (Tyumina et al. 2023) and 6 μg/L in the Mediterranean Sea (Togola and Budzinski 2008). However, most records are much less than 1 μg/L (Świacka et al. 2021). In adult zebrafish ( Danio rerio ) exposed to 10 and 100 μg/L ketoprofen for 42 days, ketoprofen accumulated in organs over this time period, reduced antioxidant capacity and at 100 μg/L caused structural changes in the liver, kidney and gut histology (Madesh et al. 2025). Ketoprofen can also be developmentally toxic at very high concentrations, with a higher mortality rate in zebrafish embryos exposed to 1000 μg/L and increased incidence of malformations at 10000 μg/L after 4 days of exposure (Rangasamy et al. 2018). In the molluscs, Mytilus galloprovincialis , ketoprofen did not accumulate to detectable levels over 60 days exposure at 2.5 μg/L (Mezzelani et al. 2018). Growth inhibition has been shown in algae; however, the test concentration of ketoprofen was not environmentally relevant at 100000 μg/L (Krawczyk et al. 2023). Ketoprofen appears reasonably well tolerated at low concentration, but the environmental concern stems from its frequent and sometimes high concentration detected in water bodies globally.

Salicylic acid is a longstanding environmental residue, since acetylsalicylic acid (aspirin) was the first NSAID to be developed over 120 years ago, and is widely used in human and veterinary medicine (Montinari et al. 2019). A systematic literature review reported the global average and maximum concentration of acetylsalicylic acid in surface water is 0.92 μg/L and 20.96 μg/L, respectively (aus der Beek et al. 2016). Acute effects of salicylic acid, such as oxidative stress, have been demonstrated across a diversity of animal taxa but occur at high, unrealistic concentrations (Nunes et al. 2015; Parolini 2020; Singh and Pflugmacher 2025; Zhao et al. 2024). Some chronic effects at low, environmentally relevant concentrations have been documented. For example, reduced acetylcholinesterase (AChE) activity, an indication of neurotoxicity and chronic stress, occurred in mussels ( Mytilus galloprovincialis ) exposed to 5 μg/L salicylic acid for 12 days (de Marco et al. 2022). In common carp (Cyprinus carpio) larvae, the lowest concentration of salicylic acid that caused significant adverse effect was 4 μg/L after 34 days exposure (Zivna et al. 2015).

The scientific literature concerning the aquatic toxicology of flunixin and meloxicam is limited. Evidence of developmental toxicity was shown in medaka (Oryzias melastigma) embryos submerged in 1 μg/L of meloxicam for 31 days which had a shorter hatching time and higher malformation rate compared with the control (Liang et al. 2025). Meloxicam also caused oxidative damage in common carp ( Cyprinus carpio ) fingerlings after being submerged for 28 days at 2000 μg/L (Sheikhlangi et al. 2023), but this concentration is not environmentally relevant, as the most common NSAIDs are typically detected in aquatic environments at levels ranging from a few hundredths to a few tenths of a microgram per litre (Świacka et al. 2021). If any of the candidate NSAIDs prove effective at reducing inflammation in salmonids, more detailed investigations into ecotoxicology are warranted, particularly for chronic effects at low, environmentally relevant concentrations.

6. Conclusion

There are no NSAIDs approved for use in farmed fish, but seven are authorised for use in terrestrial food animals across Norway, United Kingdom and Australia. Of these, meloxicam, ketoprofen and flunixin are approved in all three regions considered. Norway has four approved NSAIDs: meloxicam, ketoprofen, flunixin and sodium salicylate. These four NSAIDs are also approved in the United Kingdom, in addition to carprofen and acetylsalicylic acid. Australia has approval for the use of meloxicam, ketoprofen, flunixin and tolfenamic acid. These NSAIDs are administered to manage several inflammatory diseases, with both infectious and noninfectious causes, in farmed terrestrial animals. In fish, some NSAIDs have been shown to reduce pain sensitivity but there is no published data demonstrating that they reduce inflammation. However, the targeted immune pathways of NSAIDs appear conserved between fish and terrestrial animals so it is highly likely that NSAIDs will have anti‐inflammatory effects in fish if they can be delivered effectively.

The NSAIDs considered in this review have a good safety profile for users and the targeted terrestrial animal species. The main risk to the target animals is gastrointestinal irritation and renal disorders, but the occurrence is rare. In fish administered with NSAIDs, minimal adverse side effects have been reported. However, research is limited in this area and very few of the existing studies focused on acute and chronic toxicity. With respect to food safety, all these NSAIDs have established maximum residue limits (MRL) and withdrawal periods for target species. However, salmonids are classed as a ‘major’ species so separate studies will be required to establish species‐specific MRLs and withdrawal periods. In terms of environmental safety, the NSAIDs with the lowest environmental risk are carprofen and tolfenamic acid. Environmental risk could not be excluded for meloxicam, ketoprofen, flunixin, sodium salicylate or acetylsalicylic acid. This does not mean the later NSAIDs cannot be used safely in aquaculture with further ecological toxicology studies and under well‐defined and managed treatment regimens.

The pharmacokinetics of NSAIDs differ between species, but in general NSAIDs are rapidly absorbed after oral administration and have a low Vd due to high plasma protein binding. Elimination rates differ between species, reflected by the highly variable half‐life. In salmonids, the oral bioavailability of most NSAIDs is significantly reduced, with the exception of carprofen. This contrasts with other species, where NSAIDs generally exhibit high oral bioavailability. Variability in absorption and elimination of NSAIDs in fish compared to terrestrial animals likely stems from physiological differences such as ectothermic regulation and a gastrointestinal tract that is simpler and adapted to high protein diets. Most studies that orally administer NSAIDs use gastric gavage, providing no insight into palatability which is needed in the context of medicated feeds.

NSAIDs are a potential new treatment for inflammatory conditions in farmed fish; however, several major knowledge gaps exist before they can become commercially available. There is a need for multiple pharmacokinetics studies of NSAIDs in the target species, in this case Atlantic salmon, focusing on oral administration across different temperatures and commercially relevant life stages to ensure the medication can be effectively administered. While COX pathways appear relatively conserved across species, an increased understanding of the immunological pathways leading to inflammation, and of which NSAIDs are most suitable for targeting these pathways, is needed in fish. The anti‐inflammatory properties of NSAIDs should then be demonstrated in vivo, and the treatment regimen optimised for the specific disease of interest. Once the treatment is deemed effective and safe for Atlantic salmon, several food safety trials will be needed to determine muscle residue depletion. Then, according to the acceptable daily intake, food safety risk assessments and depletion studies, maximum residue limits and withdrawal periods can be established specific to the exact NSAID formulation, dose and treatment duration. If any candidate NSAIDs reduce inflammation in salmonids, further ecotoxicological studies and environmental release assessments will be required.

Author Contributions

William Roy: conceptualization, writing – review and editing, funding acquisition. James W. Wynne: conceptualization, writing – review and editing, funding acquisition. Christine Huynh: conceptualization, writing – review and editing, funding acquisition. Hamish Rodger: conceptualization, writing – review and editing, funding acquisition. Chloe J. English: conceptualization, data curation, writing – original draft, writing – review and editing.

Funding

This review was supported by Fiskeri—og havbruksnæringens forskningsfinansiering (FHF) grant 902028.

Disclosure

The authors used Copilot to improve the grammar, structure and clarity of selected sentences within the text following the completion of the initial draft. The authors confirm that they have reviewed and verified the content generated by the AI tool for accuracy and integrity.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: NSAIDs products approved for use in food‐producing animals in Norway (NOR), United Kingdom (UK) and Australia (AUS). Product ID or VM number refers to the official regulatory identification or veterinary medicine number assigned by the relevant authority and product name is the marketed name of the product. MA Holder refers to the marketing authorisation holder responsible for the product.

Data Availability Statement

The raw data that supports the findings of this study are available from the corresponding authors on request.