Jill E. Maddison
A relatively straightforward definition of an adverse drug reaction (ADR) is "any response to a drug which is noxious and unintended and which occurs at doses used in animals for prophylaxis, diagnosis or therapy". In veterinary medicine, suspected ADRs observed in dogs and cats that are most frequently reported to spontaneous reporting schemes involve vaccines, antimicrobial drugs, non-steroidal anti-inflammatory drugs, ectoparasiticides, anthelmintics and anaesthetic agents. These are also the most commonly used therapeutic or prophylactic agents in these species and the higher frequency of reported ADRs related to these agents probably reflects this increased usage rather than increased ADR potential.
Classification of adverse drug reactions
Type A (augmented) ADRs are expected but exaggerated pharmacologic or toxic responses to a drug. This may be an exaggeration of the intended response to the drug, a secondary response affecting an organ other than the target organ but predictable based on the pharmacology of the drug, or a toxic response.
Most ADRs of this type are attributable to differences in drug disposition that result in higher plasma drug concentrations as a result, for example, of organ failure, reduced protein binding attributable to displacement by another drug or inappropriate dosage of a non-lipid soluble drug in an obese dog. They are usually dose-dependent and avoidable if sufficient drug and patient information is available.
Type B (bizarre) reactions are unexpected or aberrant responses that are unrelated to the drug's pharmacological effect. They are not dose dependent and are unpredictable and idiosyncratic. Type B ADRs include allergic reactions, direct toxic effects on organs that are associated with actions unrelated to any desired therapeutic effect (the mechanisms for which may be complex and obscure), and aberrant responses in different species.
Difficulties in diagnosing ADRs
One of the difficulties in determining the incidence of ADRs is the difficulty in diagnosing ADRs. Appropriate diagnosis of an ADR is heavily dependent on the expertise of the attending clinician and the quality of the information available to them. Even experienced clinicians have difficulty in determining causality and experts have been shown to agree less than 50% of the time when assigning causality to an ADR.
The clinical signs of an ADR are almost always non-specific and rarely if ever pathognomonic for an ADR. In human medicine the most common symptoms of ADRs (e.g., nausea/vomiting, diarrhoea, abdominal pain, rash, pruritus, drowsiness, headache) are also reported in 80% of healthy patients on no medication. Placebo administration in humans causes an increase in the percentage of patients with symptoms and the number of symptoms per patient. Although the existence of true placebo effects may inexplicably exist in animals, veterinarians are reliant also on the observations of owners who may be subject to various conscious or subconscious factors that may influence their interpretation of their pet's behaviour, providing another dimension of placebo effect.
Other factors that contribute to difficulties in determining whether a true ADR has occurred include multiple medications, underlying pathology and assuming that it is the active principle of a medication that is responsible for the ADR. Many reactions are due to excipients and some may be due to in vitro degradation products.
Factors that influence Type A ADRs
Species differences in drug disposition may occur due to differences in absorption (due to differences in the anatomy of the gastrointestinal tract), differences in metabolism, protein binding and many other factors.
Cats have a slow rate of biotransformation for many drugs which depend on glucuronyl transferases for glucuronide conjugation in the liver. They have substantially reduced ability to conjugate drugs such as paracetamol and aspirin with glucuronic acid. As a result hepatic clearance of aspirin in the cat is very prolonged resulting in a half life of 37.5 hours compared with 8.5 hours in dogs. However, the drug can be used safely provided the dosage interval is appropriately extended. In contrast, paracetamol is extremely toxic to cats and cannot be used under any circumstances because alternative metabolic pathways to glucuronidation produce toxic metabolites. Other drugs which are metabolised more slowly in cats include dipyrone, chloramphenicol and hexachlorophene.
Some sulphate conjugation pathways are well developed in cats which may represent an alternative pathway for drugs that are normally conjugated with glucuronides in other species. Cats also have unusual receptor site sensitivity for many drugs, for example high doses of morphine result in excitement rather than sedation in this species.
Cat erythrocytes are more susceptible to oxidative changes. The grooming habits of cats facilitate ingestion of substances on their fur, thus they can receive unpredictably larger doses of aerosols or dusts. It is believed that this is the major route by which cats develop lead toxicity. Cats are more susceptible to aminoglycoside neurotoxicity than other species. The differences between factors that influence drug disposition in dogs and cats has been reviewed extensively.
Dogs do not have the ability to acetylate drugs. Where this pathway is responsible for drug inactivation, e.g., sulphonamides, the drug may, if alternative metabolic pathways are unavailable, have a longer duration of action than in other species.
Body size and percentage fat
Metabolic rate (and ability to metabolize a drug) is more closely related to body surface area than body weight and may explain why small animals within a species often require a higher dose per kg than larger animals. This is particularly relevant to dogs where the body size within the species covers such a large range. Where there is a narrow therapeutic range for the drug, this factor can become very important. The dose of a drug with a narrow therapeutic ratio (e.g., digoxin, cytotoxic drugs) is usually calculated on body surface area rather than body weight.
Another consideration when adjusting dosages for body size is the fat component of the body weight. Drug dosages are usually expressed per weight within a particular species. One should attempt to estimate the appropriate lean body weight and use this to calculate an appropriate dosage for non-lipid soluble drugs even if using body surface area. Drugs which have a narrow margin of safety and are not lipid soluble include digoxin and the aminoglycoside antibiotics.
Neonates have a reduced capability for drug biotransformation and have underdeveloped hepatic and renal excretory mechanisms. Hence the increased sensitivity to and prolonged recovery from barbiturate anaesthetics that may be observed in dogs and cats younger than 4 months. Other factors that influence drug disposition in the paediatric patient include increased gastrointestinal permeability, differences in body water and protein binding (greater percentage of body water, less extensive protein binding) and increased blood brain barrier permeability.
Older animals may have reduced hepatic or renal function, less body water and reduced lean body mass and therefore often require lower doses of drugs compared to younger animals. However, it is important to be aware that the aging process varies greatly between individuals. Patient-specific physiologic and functional characteristics are probably more important than age per se in predicting ADRs in patients.
During pregnancy or lactation, caution should be observed in administration of drugs that might affect the foetus or neonate. Adverse drug reactions occur more commonly in female humans but it is not known if this phenomenon occurs in domestic animals.
Dosage recommendations are usually based on pharmacokinetic data obtained from healthy animals under controlled conditions even though many drugs will be given to diseased animals. Drug metabolism and excretion may be adversely affected by pathology of various organs in particular the liver and kidney. Adjustment in dosage may be required, depending on the site of metabolism and route of elimination of the particular drug.
Hepatic disease can alter the bioavailability and disposition of a drug as well as influence the pharmacological effects of the drug. The enhanced effects of drugs in patients with liver disease is primarily due to decreased drug metabolism. Fortunately, glucuronidation, a common method by which lipid soluble drugs are metabolised in dogs appears to be relatively unaffected by hepatic disease. The effect of hepatic disease on bioavailability and disposition of drugs is difficult to predict. In general, when administering drugs that are extensively metabolised by the liver to patients with liver disease the dosage interval should be prolonged.
The degree to which impaired renal function affects drug elimination is determined by the fraction of the dose that is excreted unchanged by the kidneys. Some drugs are nephrotoxic (e.g., aminoglycosides, amphotericin). The potential for nephrotoxicity is increased in patients with pre-existing renal disease and patients that are dehydrated due to water/sodium loss or diuretic (especially frusemide) usage.
Drug absorption and distribution may also be adversely affected by cardiac insufficiency. Regional blood flow will be altered in cardiovascular disease resulting in the brain and heart receiving more blood and the kidney, skeletal muscle and splanchnic organs receiving less. Infiltrative gut disease will alter the absorption of orally administered drugs. Dehydration or acidosis may alter the biotransformation or distribution of drug. For example a dehydrated animal does not absorb drugs or fluids well from subcutaneous sites.
From the above discussion it is apparent that the potential for a Type A ADR to occur is higher in animals with organ dysfunction, particularly renal, hepatic or cardiac; in very young or very old animals; in animals to whom a number of drugs are administered concurrently; in species for which safe use of the drug or class of drugs has not been established and in obese or cachectic patients. In general, type A ADRs should be avoidable if the above factors are considered and dosage regimens are altered appropriately.
Type B adverse drug reactions (hypersensitivity)
Type B ADRs are unrelated to dose, are hard to predict and difficult to avoid. The major example of idiosyncratic ADRs or Type B ADRs is allergic or hypersensitivity reactions. Drug hypersensitivity reactions are more common in patients with a prior history of allergic reactions to the drug or atopic patients. However, they can occur in any individual.
Penicillin-induced hypersensitivity is the most well characterised drug-induced hypersensitivity in small animals. Other drugs that have been reported to cause allergic reactions include sulphonamides, doxorubicin, penicillamine, dipyrone and quinidine. In human medicine, allergic drug reactions account for approximately 5-10% of ADRs.
Any component of a drug preparation may induce a hypersensitivity reaction and microbiological contamination may also stimulate a hypersensitivity reaction. Drug hypersensitivity should be considered in the differential diagnosis of any apparent immune mediated disease (e.g., polyarthropathy, haemolytic anaemia and vesicular/ulcerative dermatitis).
Allergic drug reactions may occur as a result of a number of different immunological mechanisms including immediate hypersensitivity (Type I), cytotoxic hypersensitivity (Type II), immune complex formation (Type III), and delayed hypersensitivity (Type IV). However, the pathophysiology of many drug reactions eludes precise characterization and some immune reactions are a result of a combination of mechanisms.
Relatively few drugs are responsible for inducing allergic drug reactions, as most drugs are not capable of forming covalent bonds with proteins, a requisite step to render a molecule immunogenic. The drug/drug metabolite-protein complex must have multiple antigenic combining sites to stimulate a drug-specific immune response and to elicit an allergic reaction. For those drugs that are capable of inducing an immunologic response, it is generally the metabolites of the drug that are chemically reactive and easily form covalent bonds with macromolecules.
Prior exposure to the drug is not essential as hypersensitivity may develop over the course of drug administration. In humans, 5-7 days is required for drug-drug hypersensitivity to develop in a patient previously un-exposed to the drug.
Allergic drug reactions should be managed by withdrawing the drug and treating with corticosteroids if needed. Adrenaline and fluid therapy may be needed for acute anaphylactic reactions.
Pseudoallergic drug reactions
Drug reactions may occur that resemble drug allergies but do not have an immunologic basis. These reactions are often termed anaphylactoid reactions and do not require prior exposure to the drug. They occur most frequently when a drug is given rapidly intravenously. Anaphylactoid reactions may be due to non-specific release of mediators of hypersensitivity or can be due to the direct effects of the drug on tissues.