Benefits and limitations of using NSAIDs in equine care

For fear of sounding old, when I qualified 10 years ago all I remember needing to know about NSAIDs were the words flunixin – for visceral and eye pain – and phenylbutazone, for musculoskeletal pain. Plus, maybe a bit about them inhibiting certain enzymes called cyclooxygenases (COX) – but that was pushing the grey cells.

On a good day, I may have been able to recall they did something nasty to the kidneys in some horses and to the gastrointestinal tract (GIT) of others. However, more words have been added to the pages of the equine NSAID literature. I read a review published in the Journal of the American Veterinary Medical Association¹ discussing the rationale of using firocoxib, a COX-2 selective NSAID, instead of flunixin or phenylbutazone as analgesics in horses. However, this article was aimed at a US audience, so I wondered what the British situation surrounding these “newer” NSAIDs was, given we have licensed versions of meloxicam, firocoxib and ketoprofen², and whether these newer NSAIDs offer any advantage over the two stalwarts?

UK NSAIDs

How do COX-inhibitors work as analgesics? In short, COX enzymes act to convert arachidonic acid, which is released when phospholipid membranes are disrupted, into prostaglandins and thromboxane (Figure 1). These two families of mediators act to enhance the inflammatory cascade and bring about the typical signs of inflammation and pain.

However, it isn’t quite that simple. Three main isoforms of COX exist: COX, COX-1 and COX-2. COX-2 is an inducible form associated with inflammation, whereas the other two forms are crucial for normal physiological maintenance of coagulation, blood flow in the kidney and the mucosal barrier in the GIT¹. Disruption of COX and COX-1 functions produces the unwanted side effects of renal impairment and GIT ulceration, so COX-2 selective inhibition would seem like a good idea. But you need to bear in mind COX-2 is constitutively expressed in the kidney, under normal physiological situations, so renal side effects can still be seen with these COX-2 selective NSAIDs³.

The NOAH Compendium² shows that flunixin and phenylbutazone are not the most commonly licensed NSAIDS for equids; in fact, more products contain meloxicam as their active ingredient than either of the two previous NSAIDs. Firocoxib and ketoprofen are also licensed for use in horses in the UK. All three of these are COX-2 selective, and licensed in both oral (not ketoprofen) and injectable forms. The oral forms of meloxicam and firocoxib are licensed for the alleviation of acute and chronic musculoskeletal pain in horses, and the injectable forms of meloxicam and ketoprofen for the “relief of pain associated with equine colic” and symptomatic relief of fever for ketoprofen.

Efficacy

Some vets have anecdotally reported the COX-2 selective NSAIDs are not as potent analgesics as the COX-inhibitors (phenylbutazone and flunixin)¹.

The COX-2 selective NSAIDs have been shown in many studies, in horses and donkeys, to provide analgesia when compared to administration of a placebo¹. Importantly, the bioavailability of the licensed forms of these COX-2 inhibitors appears to be adequate^4,5, when the licensed equine formulations are administered at the correct dosages and intervals. When administering firocoxib, suggestions have been made a loading dose is used to ensure adequate plasma levels are reached in a timely fashion to ensure inhibition of prostaglandins and thromboxane are achieved^5,6. However, this is not mentioned in the data sheet of the UK-licensed products.

Meloxicam and firocoxib analgesic properties have both been directly compared to that of flunixin or phenylbutazone. The results of such studies can be confusing and, potentially, difficult to apply to clinical cases in the field, which is why vets often need to take into account their experience of certain drugs in a clinical setting, as well as evidence-based medicine to ultimately drive their clinical choices.

For instance, in a model comparing the analgesic effects of meloxicam and phenylbutazone between a mechanical pain (heart bar shoe model) or inflammatory pain (lipopolysaccharide [LPS]-synovitis) model, phenylbutazone appeared to be a better overall analgesic in mechanical-pain situations, and meloxicam in the inflammatory model⁷. However, the dosage of phenylbutazone used was 4.4mg/kg PO bid, which may not reflect the actual “real world” dosage of 2.2mg/kg I routinely use in clinical practice.

Naylor et al⁸ investigated the clinical outcome of naturally occurring strangulating small intestinal lesions when treated with different NSAIDs (meloxicam [0.6mg/kg IV bid] and flunixin [1.1mg/kg IV bid]). They also indirectly looked at the NSAID effect on intestinal healing – indirectly by looking at intestinal permeability through blood LPS levels. Although no difference was reported in the analgesic properties between the two NSAIDs, six horses were withdrawn from the meloxicam group as they needed to receive additional pain relief in the form of flunixin.

Meloxicam is not licensed for twice daily administration; as such, the authors obtained a special treatment certificate to allow bid administration. This makes comparison between the two NSAIDs difficult, as a real-world dosing regime is not recapitulated here. Differences seen between the two NSAIDs, however, included a significantly lower blood LPS level in the horses receiving flunixin at 48 hours post-surgery than before surgery. This was not the case with those receiving meloxicam.

Side effects of COX-2 selective NSAIDs

The main reason I can see to use COX-2 selective NSAIDs is their potential reduction in side effects. As mentioned, these NSAIDs should result in lower incidences of gastrointestinal ulceration, and potentially in renal impairment – although this is somewhat questionable given the role of COX-2 in the normal kidney physiology.

The reduction in side effects is likely to be only present at the licensed dosages and dosing intervals, and, where these are exceeded, side effects will be more likely to be seen¹. Indeed, plasma levels of firocoxib can be maintained far longer than other non-selective NSAIDs¹, prolonging the dosing interval, but also meaning the drug’s effects may be maintained a lot longer than initially desirable.

Studies^9,10 have shown COX-2 selective NSAIDs to be less detrimental to postoperative intestinal mucosal healing than flunixin, supporting the idea that sparing COX-1 may reduce endotoxin-related complications in colic patients. However, the study by Naylor et al⁸ failed to recapitulate this. The benefit of the reduction in side effects with COX-2 selective drugs has not been thoroughly evaluated in horses with volume deficits as well (hypovolaemia or dehydration), although one study did not find any benefit in using meloxicam over phenylbutazone in artificially induced volume-depleted horses on renal parameters¹¹, thus this needs to be taken in to account when using them on clinical cases.

From the research available, it appears COX-2 selective NSAIDs definitely have their place in equine analgesia. However, a thorough understanding of their benefits and limitations are needed to ensure these potential benefits is maximised in clinical cases.