Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most widely used therapeutic agents in both human and veterinary medicine. Although for many years they were chiefly considered the drugs of first choice for the management of chronic pain — a role they still hold today — the development of newer and increasingly efficacious compounds has dramatically expanded their use in the acute pain setting (Grisneaux et al, 1999; Mansa et al, 2007). While opioid drugs are still generally regarded as the ‘gold standard’ for management of moderate to severe pain, a number of studies have demonstrated that NSAIDs may provide analgesia of a similar magnitude and, in many cases, of longer duration than opioids (Reid and Nolan, 1991; Balmer et al, 1998). Current opinion, however, is that to achieve optimal analgesia in the acute pain setting the two drug groups should be used in combination as part of a multimodal approach (Flaherty, 2013).
Unfortunately, despite their extensive use for management of pain, NSAIDs exhibit a significant sideeffect profile characterised by adverse events such as gastrointestinal irritation, renal toxicity and interference with haemostasis (Bergh and Budsbergh, 2005). The incidence of side effects with NSAIDs is probably underestimated, since many of these adverse events may be sub clinical; in other patients, however, they may become life threatening.
A basic understanding of the pharmacology of NSAIDs is useful to appreciate how these drugs provide analgesia, but also to recognise why, and under what circumstances, side effects may occur.
Pharmacology of the NSAIDs
Prostaglandins are one of the major groups of inflammatory mediators involved in generation of pain, through both contributing to peripheral sensitisation, and also via a direct effect in the central nervous system (Flaherty, 2013). The synthesis of prostaglandins (Figure 1) is triggered by tissue membrane damage which causes the detachment of arachidonic acid from the membrane phospholipids into the cytoplasm of the cell through the action of the enzyme phospholipase A2. The arachidonic acid can then follow one of two metabolic pathways:
- Under the action of the enzyme 5-lipoxygenase (5-LOX), it can be broken down to form leukotrienes, a group of inflammatory mediators similar to the prostaglandins, but possibly of lesser importance
- Under the action of the enzyme cyclooxygenase (COX), arachidonic acid can be converted into a variety of prostanoids with a range of effects throughout the body (Khan and McLean, 2012). Prostanoids is a term used to describe a variety of arachidonic acid derivatives, including the prostaglandins, prostacyclins and thromboxane.
The principal action of NSAIDs is inhibition of the COX enzyme system; consequently, they prevent production of prostaglandins (and other prostanoids) within the body (Vane, 1971). Some NSAIDs (so-called ‘dual inhibitors’) may additionally inhibit 5-LOX activity, but few of these are commercially available. One such dual-inhibitor is the NSAID, tepoxalin (Zubrin®, Schering Plough Animal Health), which was licensed for the treatment of osteoarthritis in dogs; however, this product has now been withdrawn from the UK market. Thus, the vast majority of NSAIDs currently available exert their effects through COX inhibition.
Significance of different COX isoforms
Originally, COX was considered to be a single enzyme, but, over 20 years ago, two distinct forms (‘isoforms’) of cyclooxygenase were subsequently identified, and named cyclooxygenase 1 (COX-1) and cyclooxygenase 2 (COX-2) (Fu et al, 1990; Kujubu et al, 1991).
It had long been realised that, although prostaglandins were intimately involved in generation of pain, they were also responsible for some necessary functions within the body. In addition, side effects of the NSAIDs available were already well recognised. The identification of the two different isoforms of COX thereby offered a readily accepted (at that time) hypothesis, namely that COX-1 was a constitutive (‘housekeeping’) enzyme present in almost every cell type in the body and exerting an important role in physiological functions such as renal and gastric protection, while COX-2 was considered to be an inducible enzyme produced only under conditions of inflammation and was, therefore, responsible for producing ‘unwanted’ prostaglandins, i.e. those responsible for pain.
On this basis, it was presumed that NSAID-induced inhibition of COX-1 was responsible for adverse effects such as gastrointestinal or renal damage, whereas inhibition of COX-2 produced the therapeutic anti-inflammatory effects.
In the 1990s, the firm belief that most of the NSAIDs’ adverse effects were caused by COX-1 inhibition lead to studies with the intention of identifying drugs that could selectively inhibit COX-2 without interfering with COX-1 homeostatic functions. This research resulted in the introduction of many COX-2 selective drugs in both human and veterinary medicine (Bergh and Budsbergh, 2005). It was considered that, by sparing COX-1, these agents would be efficacious anti-inflammatory and analgesic agents with minimal side effects. Unfortunately, development of these drugs did not produce the expected results, in that adverse effects on various organ systems were still apparent. In recent years, studies have confirmed that while COX-1 is mainly constitutive, it also plays some role in producing inflammation and pain, and while COX-2 is inducible under conditions of inflammation, it also has some constitutive roles: it is now accepted that COX-2 is constitutively expressed in a variety of tissues, such as the kidneys, ovaries, spinal cord and gastric mucosa (Wilson et al, 2004; Lascelles et al, 2009), and, just as with COX-1, COX-2 appears to have a variety of protective physiological functions such as in maintaining gastrointestinal integrity, assisting uterine function and bone metabolism (Schmassmann et al, 1998). Accordingly, COX-2 inhibition is now recognised to have potentially detrimental effects in the same way as does inhibition of COX-1, although drugs that inhibit COX-2 while relatively sparing COX-1 do appear to have a lower incidence of side effects.
Following on from the discovery of the two different forms of the cyclooxygenase enzyme, COX-1 and COX-2, more recently, a third isoform, initially called COX-3, was identified in the central nervous system of dogs (Chandrasekharan et al, 2002). As yet, relatively little is known about this enzyme, but it is now considered an alternative spliced form (‘splice variant’) of COX-1 rather than an alternative distinct isoform; it may be a component of the central analgesic activity of NSAIDs, and it has been suggested that it may also be a target for the anti-pyretic analgesic, paracetamol (although paracetamol is often classed as an NSAID, it lacks anti-inflammatory effects, so is better classified as an antipyretic analgesic).
Classification of NSAIDs
NSAIDs can be classified according to numerous characteristics, including chemical and pharmacological properties, but the distinction based on their selectivity for COX-1 versus COX-2 is the one most commonly referred to. On this basis, NSAIDs can be divided into non-selective COX inhibitors, selective COX-1 inhibitors, selective COX-2 inhibitors and highly selective (specific) COX-2 inhibitors (also known as COXIBs) (Table 1).
Table 1. Classification of non-steroidal anti-inflammatory drugs (NSAIDs) based on cyclooxygenase (COX) selectivity
| Classification | Definition | Examples |
|---|---|---|
| Non-selective COX inhibitors | Inhibit COX-1 and COX-2 to a similar extent | Aspirin, ibuprofen |
| Selective COX-1 inhibitors | Significantly greater inhibition of COX-1 compared to COX-2 | Tepoxalin (now withdrawn from the UK veterinary market) |
| Selective COX-2 inhibitors | Significantly greater inhibition of COX-2 compared to COX-1 | Meloxicam, carprofen (though exact mechanism of action of carprofen is unclear) |
| Specific COX-2 inhibitors | Specifically inhibit COX-2 with minimal effects on COX-1 | Robenacoxib, firocoxib, cimicoxib |
Non-selective COX inhibitors
Older NSAIDs, such as acetylsalicylic acid (aspirin) and ibuprofen, act on both COX-1 and COX-2, and are hence defined as ‘non-selective’ COX inhibitors (Frolich, 1997). Due particularly to their effects on COX-1, these agents are associated with a relatively high incidence of gastrointestinal side effects.
Selective COX-1 inhibitors
There are currently no selective COX-1 inhibitors licensed for veterinary use in the UK, although one — tepoxalin — was previously marketed for management of musculoskeletal pain in dogs (Zubrin, Schering-Plough Animal Health). This product was withdrawn due to a high incidence of gastrointestinal side effects.
Selective COX-2 inhibitors
Given that COX-2 seems to have greater involvement in the inflammatory process than COX-1, and also that COX-1 appears to have a greater constitutive role than COX-2, most of the modern NSAIDs developed have a preferential effect on inhibition of COX-2. While these agents are certainly not free of potential side effects (given the constitutive role that COX-2 also plays in the body, and the fact that all of these drugs will also have a lesser effect on inhibition of COX-1), the COX-2 selective NSAIDs demonstrate a lower incidence of adverse effects (particularly gastrointestinal effects) than the non-selective COX inhibitors, such as aspirin.
Specific COX-2 inhibitors (COXIBs)
While the selective COX-2 inhibitors have their major effect on COX-2, all of them will additionally affect COX-1, albeit to a lesser extent. Specific COX-2 inhibitors (COXIBs), on the other hand, essentially ‘spare’ COX-1 and exert their effects purely on COX-2. The idea behind the development of these drugs was to maximise the anti-inflammatory effects while minimising side effects. However, bearing in mind the constitutive role of COX-2 in the body, adverse events are also seen with these agents, and, indeed, one — rofecoxib — was withdrawn from the human market due to an increase risk of myocardial infarction and stroke.
Therapeutic effects of NSAIDs
Analgesia and anti-inflammatory effects
NSAIDs have been utilised for decades as the ‘first line’ management drugs in chronic pain, and their efficacy in conditions such as osteoarthritis is well documented (Moreau et al, 2003; Hanson et al, 2006; Mansa et al, 2007).
More recently, probably due to improved understanding of the pathophysiology of pain as well as adoption of the concept of multimodal analgesia, NSAIDs have assumed a major role in perioperative pain management.
Peripheral and central analgesic effects
As previously highlighted, NSAIDs reduce inflammation at sites of tissue damage through the inhibition of prostaglandin production and, therefore, inhibit peripheral sensitisation and pain (Flaherty, 2013). Some prostaglandins, such as PGI2 (prostacyclin) and PGE2, dilate arterioles, resulting in increased blood flow to the affected area, which contributes to the redness and swelling that is part of the inflammatory process, as well as increasing the delivery of other inflammatory mediators to the site. In addition, while prostaglandins have a direct effect in reducing the firing threshold of peripheral nociceptors at sites of tissue damage, they also increase the sensitivity of these nociceptors to other inflammatory mediators such as histamine, bradykinin and substance P, which are responsible for localised pain and hypersensitivity (Stock et al, 2001). Peripheral sensitisation results in an increased sensitivity to subsequent noxious stimuli (hyperalgesia), and NSAIDs have been shown to be effective in reversing established hyperalgesia (Zhang et al, 1997).
Certain prostaglandins (in particular, PGE2) also play a role in spinal nociception, and may be involved in the development of central sensitisation. COX-2 appears to be the main isoform involved in generating these particular prostaglandins within the central nervous system. Given this contribution of COX-2 to the central development of pain, and also that COX-2 selective inhibitors tend to cross the blood–brain barrier more easily than do other NSAIDs (due to greater lipid solubility and a less acidic pH), COX-2 selective NSAIDs are considered to be more effective in attenuating the centrally-mediated aspects of pain (Buvanendran et al, 2002).
Multimodal analgesia
Given that nociceptive transmission involves a multiplicity of pathways, mechanisms and mediators, neither NSAIDs nor any other class of analgesic drug are able to act on every component (Muir and Woolf, 2001). This has led to the introduction into clinical practice of multimodal analgesia, which refers to the concept of using analgesic drugs from several different classes in order to ‘target’ different parts of the nociceptive system. Due to the additive and synergistic effect of administering different types of analgesic, this should be more effective than utilising a single drug treatment regimen (Ritchie, 2006). In addition, combining a number of drugs will allow effective analgesia to be achieved using lower drug doses than would be required to produce a comparable level of analgesia with only a single drug; this reduction should decrease the likelihood of side effects (Kehlet and Wilmore, 2002).
NSAIDs’ analgesic/anti-inflammatory characteristics, mainly related to the reduction of the prostaglandin synthesis, make them an important part of multimodal analgesia, both for acute/perioperative pain and for chronic pain. Several meta-analysis studies conducted in human medicine have demonstrated that inclusion of NSAIDs as part of a multimodal analgesic regimen is advantageous in reducing the doses of the other analgesic drugs and their side effects, and also improves patient pain control and overall clinical outcome (Elia et al, 2005; Marret et al, 2005).
Other therapeutic effects of NSAIDs
Antipyretic effects
NSAIDs are effective antipyretics, but the role of the different COX isoforms in relation to the process that leads to fever is not clear; different iso-enzymes appear to be involved, with each one acting in a specific way and under specific circumstances depending on the triggering factor for the fever. For example, COX-2 seems to act in cases of pyrexia due to endotoxaemia, whereas COX-1 and COX-3 in cases due to endogenous causes (Parrott and Vellucci, 1998; Botting, 2003).
Antiendotoxin effects
NSAIDs have been used as therapeutic agents in the treatment of endotoxaemia.
Endotoxin originates from the outer wall of Gram negative bacteria, and, following absorption into the blood, results in a systemic inflammatory reaction. The inflammatory mediators so released are responsible for several of the clinical signs of endotoxaemia, such as cardiovascular changes, gastrointestinal ileus and fever. Whereas in some species data regarding the efficacy of NSAIDs in the management of endotoxaemia is controversial, they are considered highly effective agents, and are commonly used, in toxaemic horses and cattle in conditions such as colic and septic metritis or mastitis, respectively (King and Gerring, 1989; Konigsson et al, 2002).
Antineoplastic effects
Numerous human and canine studies have demonstrated that NSAIDs, especially COX-2 selective inhibitors, may be effective in the treatment of tumours such as carcinomas, adenocarcinomas and adenomas (Knapp et al, 1994; Steinbach et al, 2000; Liao et al, 2012).
COX-2 ‘up-regulation’ has been associated with malignancy, with COX-2 activation leading to higher concentrations of PGE2 which is responsible for several detrimental pathophysiological effects in cancer, including increased cell proliferation, enhanced angiogenesis, and increased tumour cell resistance to apoptosis (Mohammed et al, 2002). The administration of NSAIDs with the consequent inhibition of COX, helps offset these PGE2 pro-neoplastic effects.
While there is evidence for the beneficial effect of NSAIDs as anti-neoplastic agents in human and canine studies, results in other species have been inconclusive (Mohammed et al, 2004).
Conclusion
NSAIDs are highly effective analgesics for both acute and chronic pain, exerting their actions both peripherally, at the site of tissue damage, as well as within the central nervous system itself. Their effects principally occur through inhibition of the enzyme, COX, which results in decreased production of the prostaglandin group of inflammatory mediators. However, prostaglandins have a range of ‘housekeeping’ effects within the body in addition to their role in producing inflammation and pain, and the majority of the side effects seen with NSAIDs are due to inhibition of these constitutive prostaglandins. Despite attempts to develop drugs which specifically target the production of inflammatory (inducible) prostaglandins while sparing those that are constitutive, this has so far proved elusive, and all currently available NSAIDs carry the potential for adverse effects to be observed. The second part of this article (to be published in an upcoming edition of The Veterinary Nurse) will examine in more detail the contraindications and potential side effects of NSAIDs.
Key Points
- Non-steroidal anti-inflammatory drugs (NSAIDs) are efficacious analgesics in the management of both acute and chronic pain.
- For acute pain, in particular, to achieve maximal benefit, NSAIDs should be used as part of a multimodal approach to analgesia, in conjunction with other drugs, such as opioids.
- Most NSAIDs exert their effects through inhibition of the enzyme, cyclooxygenase, which is responsible for the production of pro-inflammatory prostaglandins.
- Different isoforms of cyclooxygenase exist, and there is some overlap between their constitutive (‘housekeeping’) and inducible roles.
- In addition to their analgesic effects, some NSAIDs have other potentially useful therapeutic actions, such as anti-neoplastic activity.
- All currently available NSAIDs have the potential to produce side effects due to inhibition of some of the constitutive functions of cyclooxygenase.