  |
|
|
|||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
22-September-2006, Vol 119 No 1242
|
|||||||||||||||||||||||||||||||||||||
|
Back to the future: postoperative pain management beyond
COX-2 inhibitors
Ole Naesh
Recent debate has highlighted severe cardiovascular side
effects from COX-2 selective nonsteroidal anti-inflammatory drugs (COX-2
inhibitors or coxibs). This has let to a withdrawal from the New Zealand (NZ)
market of rofecoxib and valdecoxib in accordance with international
recommendations.
The bulk of evidence for the potentially harmful effects of
COX-2 inhibitors is based on long-term trials in rheumatoid populations. An
extensive database is available on the analgesic efficacy of COX-2 inhibitors
for postoperative pain.1 While only a few studies have focused on adverse
effects of the short-term perioperative use (1–5 days) of these drugs, the
main impression seems to be of a high general tolerability apart from a cluster
of cardiovascular adverse events reported after coronary by-pass
surgery.2–6
Anaesthetists have been advocates of the COX-2 inhibitors as
they have allowed more patients to get the benefits from a nonsteroidal
anti-inflammatory drug (NSAID) as part of a multimodal perioperative analgesic
regimen offering superior analgesia with opioid sparing effect and reduced
opioid-related side effects.
In the aftermath of the recent COX-2 debate it is however
appropriate to revisit the rationale perioperative use of NSAIDs. The extent
upon the prolonged use of NSAIDs outside the perioperative period will not be
elaborated.
PharmacologyThe synthesis of prostaglandins is the primary target of all
NSAIDs. Prostaglandins are known to be involved in numerous physiological
systems (Table 1). The regulation of vascular tone and platelet aggregation is
affected by endothelial prostacyclin and platelet-derived thromboxane.
Prostaglandins of the E-series exert protective effects on the gastric
mucosa.7–11 Prostaglandins are also of major importance in the regulation
of the inflammatory cascade and they act as sensitisers of peripheral
nociceptors.12
The synthesis of prostaglandins (PG-series), thromboxane,
and the leucotrienes is initiated (e.g. after tissue trauma) by the conversion
of arachidonic acid to intermediate PGG2 and PGH2, which are the substrates for
the several other prostaglandins (Figure 1).
The two first
metabolic steps are catalysed by cyclo-oxygenase (COX) which is the enzyme
responsible for the velocity of the reaction and thus the rate limiting factor.
Cyclo-oxygenase is known to be present in at least two isomeric forms (COX-1 and
COX-2) with different physiological effects.9,10
Table 1. Prostaglandins: their organ-specificity and
effects
The
predominant COX-enzyme involved in their metabolism is marked ++; + indicates
that the enzyme is present but in lower concentrations under normal
non-inflammatory conditions; GFR=glomerular filtration rate;
GI=gastrointestinal; CNS=central nervous system.
COX-1 is a costitutive enzyme (i.e “daily
household”) and is involved in the production of "physiological"
prostaglandins. COX-2 is classically described as inducible and is expressed in
inflamed/traumatised tissues, but is lacking in others (e.g. platelets) (Figure
2). Recent evidence, however, points to a more complex picture, with the COX-2
enzyme being constitutively expressed in several tissues as e.g. brain and
kidney (cf. Table 1). A third isomeric form (COX-3) has recently been proposed
as being expressed in the restitutional phase of inflammation.13
The classic, non-selective NSAIDs are not more specific for
either isomeric form of the COX enzyme as opposed to the newer and selective
COX-2 inhibitors. A hydrophilic side-pocket unique to the COX-2 isoenzyme allows
the active site to accommodate only the coxibs due to their added side chain.
Classic NSAIDs block arachidonic acid access to both isoforms. However the
degree of COX-1 or COX-2 selectivity (i.e. COX-1: COX-2 inhibitory ratio)
warrants caution in the interpretation due to methodological differences of
currently available test systems, of which biological models have more clinical
relevance. Of the various coxibs, celecoxib has a ratio of 1:30 whereas
rofecoxib, for example, has a ratio of 1:276 and lumaricoxib a ratio of
1:433.
The analgesic effects of NSAIDs are ascribed primarily to
COX-2 inhibition, whereas several adverse effects are believed to be mediated by
COX-1 inhibition.
The inhibition of COX-1 prolongs the bleeding time due to an inhibition of TXA2
synthesis from platelets and may lead to the formation of gastric ulcerations
due to PGE2 inhibition. COX-1 inhibition may, under certain circumstances,
decrease renal glomerular filtration rate. COX-2 selectivity may theoretically
attenuate such adverse effects.14
Figure 1. Tissue injury
activates the arachidonic acid cascade through membrane bound phospholipase A2
(PA2).Through the action of cyclooxygenase various prostaglandins (PG) are
formed according to tissue specific pathways. Nonsteroidal anti-inflammatory
drugs (NSAIDs) inhibit the COX enzyme.
 Figure 2.
Cyclo-oxygenase (COX) exists in at least two forms (COX-1 / COX-2). In a
simplified model, various normal physiological stimuli induce COX-1 activity and
inflammation induces COX-2 activity. By the enzymatic action of COX, arachidonic
acid is converted into prostaglandins (TXA2=thromboxane; PGI2=prostacyclin;
PGE=prostaglandin E)
 Inhibition of COX-3 by the non-selective NSAIDs may
theoretically interfere with restitution after tissue trauma but the clinical
relevance remains to be elucidated.13 The above mentioned pharmacological
effects of NSAIDs are widely accepted. Central (CNS) effects have recently been
suggested but the exact mechanism has not been fully clarified. NSAIDs are
strongly ionized at physiological pH, and have difficulties passing the
blood-brain barrier. However endothelial cells in the brain were recently shown
to possess interleukin receptors coupled to COX-2 activity.
Provided the proper stimulus (e.g. interleukin 1), such
receptors would allow PGs to be expressed in the brain. An inhibition by NSAIDs
at this level could explain an intracerebral effect. Finally, an interaction
with opioid receptors in the CNS has been suggested.15–18
Selective COX-2 antagonism affects the balance between PGI2
and TXA2 in favour of TXA2, and might thus act as a prothrombotic principle due
to unopposed inhibition of endothelial PGI2.19,20 Early warnings of this
potential mechanism were given by several groups but it took larger patient
materials in order to realize that COX-2 antagonism might increase
cardiovascular events in predisposed patients.
Several recent large-scale studies have unraveled this side
effect and have led to a heated debate on the use of these new drugs and to the
withdrawal of some (but not all) coxibs.21–25
The inherent potential for serious cardiovascular events
with the use of coxibs seems to be a class effect but may still differ among the
coxibs (e.g. rofecoxib >> celecoxib). However, as would be expected, it
becomes evident in patients at risk of such events.
Recent debate has focussed on an interference of coxibs with
the mechanisms of myocardial preconditioning (i.e. a preceding minor ischaemic
episode offering protection towards a following ischaemic event) and on the
impact of coxibs on renal physiology as underlying coxib-induced adverse
cardiovascular events.24 Indeed, the increased cardiovascular risks are
acknowledged in the recent national and international recommendations on the
continued use of COX-2 inhibitors.26
To paraphrase George Orwell,
all NSAIDs are not created equal.
Indeed, possibly due to several variant forms of the COX enzyme,27 NSAIDs differ
in their effects and side effects profile. Of the four major pharmacological
groups of classical NSAIDs (salicylic acid, propionic acid, acetic acid, and
oxicams), there are differences in platelet inhibitory activity (e.g. diclofenac
being less active than comparator NSAIDs) and in gastrointestinal side effects
profile (e.g. ibuprofen showing best GI tolerability).28
Interaction (e.g. decreased effect) with low-dose regimens
of salicylic acid is seen with some but not with others (e.g. ibuprofen may
interact whereas e.g. diclofenac does not).29 Furthermore, NSAID-induced side
effects show a strong dose, time, and age dependence.
Clinical use of COX-2 inhibitorsSo in which patients are COX-2 inhibitors indicated for
perioperative pain management?
As an inflammatory tissue response to surgery is involved in
sensitisation of peripheral and central pain pathways, NSAIDs / coxibs should be
used as facilitating analgesics as part of a multimodal regimen. Coxibs should
be used at the lowest recommended dosage and for short periods of time only
(i.e. < 5 days). A lower “ulcerogenic” potential and a
platelet-sparing effect must be taken into consideration.
Where the nature of the surgical intervention
“contraindicates” use of classic NSAIDs due to risk of bleeding
(e.g. ENT surgery, plastic surgery, or neurosurgery) coxibs may well be
used.30
There is, however, no valid evidence for any superior
analgesic benefits of the coxibs as opposed to the classic NSAIDs in the
majority of surgical patients.1,31–34
In NZ, there is access to five COX-2 selective NSAIDs:
meloxicam, etoricoxib, lumericoxib, celecoxib, and parecoxib. The latter is for
parenteral use and is the only available intravenous (iv) COX-2 inhibitor in NZ.
Primarily used by anaesthetists during the immediate perioperative period, it
has not proven superior to any classical iv NSAIDs.
Meloxicam has a long record as anti-inflammatory drug with a
relatively long half-life (~20 h) and a favourable gastrointestinal (GI) profile
(ulcerogenicity and GI bleeding) in long-term studies in osteoarthritis. The
database for its perioperative use is limited, and its COX-2 selectivity at
clinically relevant perioperative doses has been questioned (Virtual Anaesthesia
Textbook: http://www.virtual-anaesthesia-textbook.com).
Etoricoxib has not been trialled in the perioperative
setting. Lumericoxib was recently introduced, but is only scarcely documented
for postoperative pain.
This leaves us with celecoxib as the only available true
COX-2 selective, oral NSAID for perioperative analgesia. A one-off premedication
dose of 400 mg (as opposed to 200 mg) of celecoxib (elimination T1/2 ~ 4–8
hr), followed by 200 mg once to twice daily for postoperative analgesia, has
recently been advocated as optimal in adults (cf. Straube et al 2005).
Intraoperatively, the only available intravenous COX-2 inhibitor is parecoxib
(elimination T1/2 ~ 8 hr) at a normal adult dose of 40 mg. No further NSAIDs
should be administered until after at least 12 hours.
Contraindications and adverse effects of NSAIDsAnecdotes and myths are often quoted when discussing NSAIDs.
Clear and upated guidelines for the perioperative use of NSAIDs (including
COX-2) are imperative to ensure that patients get the full benefit of their
inclusion into a multimodal analgesic regimen. There are, however, some clear
contraindications to the use of NSAIDs and some more controversial relative
contraindications (Table 2).
The national advisory board of NZ (Medsafe: http://medsafe.govt.nz/hot.htm), in
agreement with international consensus, recommends that COX-2 NSAIDs are
contraindicated in patients undergoing cardiac or vascular surgery, and in
patients at high risk of cardiovascular disease (including patients with
diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia,
hypertension, or smokers) who are undergoing major surgery.
Table 2. Absolute contraindications to NSAIDs
There is, however, no indications as to the severity of such
risk factors and the perioperative team is left with a recommendation to weigh
the risks and benefits in each individual case and an obligation to inform the
patient of any intended perioperative use of coxibs.
It is worth remembering that all NSAIDs hold a potential to
aggravate any pre-existing heart failure and hypertension due to fluid retention
through renal effects. It is also worth noting that patients should continue
low-dose acetylsalicylic acid if a COX-2 inhibitor is prescribed to maintain a
cardioprotective / antiplatelet effect, as this is not offered by COX-2
inhibitors.
Ibuprofen and indomethacin may impede access of aspirin to
platelet COX-1 enzyme and inhibit this protective effect. Furthermore, the
addition of low-dose acetylsalicylic acid will remove any
“gastroprotective” effect of the COX-2 inhibitor.
Patients with a history of peptic ulcer disease may well
benefit from the perioperative use of coxibs if a NSAID is indicated. An
alternative approach to coxib prescription for the patient with gastrointestinal
intolerance to NSAIDs is the concomitant use of a proton inhibitor or
misoprostol as mucosal protection.35 Patients on continued perioperative
acetylsalicylic acid should also receive gastroprotection during COX-2 inhibitor
treatment.
Another major concern of perioperative NSAID use is the
potential for renal impairment. Both COX-1 and COX-2 activity is expressed in
the kidney, and in the marginally failing kidney, any class of NSAIDs may cause
deterioration of such failure. It has been stated that the physiological
function of COX-1 in the kidney is mainly in maintaining the glomerular
filtration rate (GFR), whereas COX-2 is primarily involved in water and
electrolyte haemostasis.36
Side effects such as acute renal failure, papillary
necrosis, and nephrotic syndrome are extremely rare with NSAID therapy in the
uncompromised patient. Although mild side effects such as hyperkalaemia and
fluid retention, and minor increases in blood pressure, are seen they are
readily reversible with discontinuation of the drug. Furthermore, interactions
with antihypertensives and diuretics warrant caution in patients concomitantly
treated with these drugs.
Patients who should not receive either non-selective nor
COX-2 selective anti-inflammatories are the ones with preoperative renal
dysfunction or a renal perfusion compromise (e.g. hypovolaemia, severe liver or
heart failure, and advanced hypertension or diabetes).
Continuous perioperative monitoring of creatinine and urea
during NSAID therapy is warranted in patients with hypertension and mild
diabetes and also in the very elderly. There are some indications, however, that
COX-2 inhibitors may prove safer in these patient groups, but more data is
needed before any recommendations can be made.37
NSAIDs are contraindicated in patients with a known allergy
to this group of drugs. As celecoxib and parecoxib contain a sulphonamide moity,
they are contraindicated in patients with a known allergy to sulphonamides. Of
the asthmatic population, only 10–15% are actually reactive to the effect
of NSAIDs, partly related to their diversion of the arachidonic acid cascade
towards bronchoconstrictory leucotrienes.38
A simple questioning of the patient prior to surgery of any
NSAID usage in the past will often solve the concern. If the patient is an
asthmatic and never challenged with NSAID, the perioperative period is perhaps
not the ideal time to test the system. In addition, COX-2 selectivity does not
seem to confer any advantage in these patients.
The use of NSAIDs in orthopaedic surgery is
controversial—some experimental work points to NSAIDs having an inhibitory
effect on bone healing. However long experience and a widespread use of NSAIDs
after fracture surgery has not highlighted any clinical problem.
Some evidence points to a negative impact of NSAIDs after
spinal fusion surgery but most of the scarce literature has not corrected for
confounders such as smoking which has a major impact on bone and soft tissue
healing.39–41 Whilst awaiting prospective, randomised clinical trials, no
clear recommendations can be given, although the prudent healthcare professional
might consider avoiding NSAIDs in orthopaedic cases involving bone grafting
As the coxibs may interfere with the cytochrome P-450
enzymatic system, other drugs depending on this enzyme for their metabolism may
be adversely affected.42 Anticoagulation therapy with warfarin can thus be
potentiated and the dosage may have to be adjusted in those cases where a NSAID
(COX-2) is deemed of major benefit—but the classic NSAIDs are
contraindicated due to their antihaemostatic effect.
The risk of haematoma formation with the use of neuraxial
blocks (i.e. spinals and epidurals) and the concomitant use of NSAIDs has been a
matter of concern. International consensus holds that non-selective NSAIDs
(including acetylsalicylic acid) do not per
se contraindicate neuraxial techniques. But as combined with
low-molecular weight heparins and/or other “weak” anticoagulants,
the risks of bleeding does increase.
Full anticoagulation (e.g. coumarins) contraindicate
neuraxial techniques irrespective of NSAIDs. Combination of NSAIDs and the newer
anti-platelet drugs (e.g. clopidogrel, ticlopidine) markedly increase the risk
of perioperative bleeding and should be avoided. Interestingly, several
complementary and alternative medicines (e.g. garlic, ginko, ginseng) are
platelet inhibitors so it is currently recommended that their use is stopped
before surgery and that the are not used together with classic NSAIDs.43
Perioperative use of classic NSAIDsNSAIDs are an integral part of a multimodal and preventive,
perioperative analgesic regimen. It is, however, only a relatively small
proportion of patients who will benefit from selective COX-2 inhibition for
perioperative analgesia.
Gastric ulcer disease (GI intolerance to non-selective
NSAIDs) or surgical request for minimal platelet inhibition (e.g. plastic
surgery, neurosurgery, ENT surgery) may warrant the use of perioperative
selective COX-2 inhibition. The combined use of NSAIDs and paracetamol has
proven highly cost-effective and with a desirable opioid sparing effect, not
least in day-case surgery.1,37,44–45
Of the classic NSAIDs, ibuprofen in appropriate oral dose
(i.e. 400–800 mg tds), diclofenac (50 mg tds), and iv tenoxicam (e.g.
20–40 mg intraoperatively) have a long and well-established place in
perioperative analgesia. Together with paracetamol, a NSAID can be incorporated
into a cost-effective “take-home pack” for day-case surgical
patients (e.g. paracetamol 1 g qid with ibuprofen 400 mg tds for 3 days’
use).
It is notable that merely being a child is no
contraindication to the use of NSAIDs. Indeed, no evidence suggests that
paediatric surgery patients tolerate NSAIDs to any lesser extent than adults;
otherwise healthy children may well benefit from the perioperative use of
NSAIDs, often in combination with paracetamol.
Judicious consideration of indications, side effects, and
contraindications is as appropriate as in the adult surgical patient.46,47
Interestingly, a recent study showed less analgesic efficacy of a COX-2
inhibitor (rofecoxib) than ibuprofen in a paediatric tonsillectomy population
although any potential haemostatic advantage of a COX-2 inhibitor was not
further discussed.48 COX-2 selective NSAIDs are currently not recommended in NZ
for age groups under 18 due to lack of valid data on dose-effect relations.
Supplemental drugs such as α2-agonists (e.g.
clonidine), NMDA receptor antagonists (e.g. ketamine), tramadol, and gabapentin
are emerging as facilitating perioperative analgesics and their possible
combination with NSAIDs is under intense scrutiny. A more procedure-specific
approach to perioperative analgesia has recently been suggested.49 Eventually,
recommendations for analgesic regimens involving or not involving NSAIDs or
COX-2 inhibitors may emerge.
We may have gone two steps forward and one step back, but
the judicious use of the entire group of NSAIDs (selective and non-selective) in
postoperative pain remains to be determined in the future.
Author information:
Ole Naesh, Consultant Specialist in Anaesthesiology, Department of Anaesthesia
and Intensive Care, Timaru Hospital, Timaru, South Canterbury
Acknowledgement: I
am grateful to Dr R Rarity for his
valuable input and proofreading.
Correspondence: Dr
Ole Naesh, Dept of Anaesthesia, Helsingborgs Lasarett, S.Vallgatan 5, 251 87
Sweden. Email: OleNaesh.Hendriksen@skane.se
References:
|
|
|||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|