NONSTEROIDAL antiinflammatory drugs (NSAIDs) produce an antiplatelet activity by inhibiting platelet cyclooxygenase.1Their mechanism of action is thus close to that of aspirin. Consequently, NSAIDs have been used to prevent thrombosis in indications similar to those of aspirin use. For example, indobufen2and flurbiprofen3have been shown to be effective antithrombotic agents in patients with coronary syndromes. Moreover, different molecules exert different levels of maximal inhibition,1and inhibition within each NSAID molecule is typically dose dependent. In short, NSAIDs differ from aspirin not only by their reversible action on platelet activity but also by not having aspirin's dose-independent on-off mechanism. In view of the potential adverse effects of maintaining aspirin therapy before scheduled surgery, these considerations provide the rationale to replace aspirin with a drug with a similar, but much shorter, action on platelet activity. If normal platelet activity is recovered within 24 h, the patient would be protected during the substitution period, and platelet function and hemostasis would be back to normal during surgery. NSAIDs could be the ideal substitute for aspirin, but, despite their attractiveness, proof is lacking because studies in the field are few and far between. The study by Gonzáles-Correa et al. 4in this issue of Anesthesiology could be understood as a first step toward providing the evidence we require to justify this practice.
The NSAID ibuprofen is a racemic mixture of two enantiomers, S (+) ibuprofen (dexibuprofen), which is the active enantiomer, and R (−) ibuprofen, which is inactive. Several studies have demonstrated the antiplatelet activity of ibuprofen on platelet aggregation5and on occlusion time measured using the Platelet Function Analyzer 100.6In their study, Gonzales-Correa et al. 4show that the antiplatelet activity of dexibuprofen is similar to that of aspirin and that the administration of dexibuprofen, unlike that of aspirin, results in complete recovery of platelet function 24 h after drug withdrawal. Dexibuprofen might indeed be an appropriate alternative to aspirin in a perioperative setting.
All is not that simple, however. For many years, some Europeans have often prescribed NSAIDs, in particular indobufen or flurbiprofen, during the 10 days preceding surgery, after withdrawing aspirin. The NSAID is administered twice a day and discontinued 24 h before surgery so that normal platelet function is recovered in time for the intervention. Because these compounds develop a strong antiplatelet activity that is similar to that of aspirin, prescription is thought to carry few risks apart from the usual complications of NSAIDs (gastrointestinal toxicity, renal insufficiency). This preoperative use in coronary patients represents an officially recognized indication in the Summary of Product Characteristics for flurbiprofen in some countries. However, substituting an NSAID for aspirin has never been assessed in comparative trials of the two drugs. Moreover, a recent study by Collet et al. 7reported 47 cases of acute coronary syndrome among patients in whom aspirin was withdrawn before scheduled surgery. Some of these patients had received flurbiprofen as a substitution, but this did not prevent acute coronary syndrome. Clearly, prescribing even a powerful NSAID in this indication could be questioned.
The study of Gonzáles-Correa et al. 4is useful to confirm the potent antiplatelet activity of dexibuprofen and its reversible action. However, it is important to note that their study was performed in healthy volunteers and not in patients. No bleeding events were expected. Moreover, after 14 days, the reduction in prostacyclin synthesis was greater in the dexibuprofen group, potentially shifting the balance arms toward a prothrombotic risk. Therefore, even if this agent develops a reversible platelet activity, further explorations are mandatory to eliminate a theoretical prothrombotic risk.
Actually, it may be that it is the withdrawal of aspirin that should be reconsidered. A small number of fairly common clinical situations undoubtedly require aspirin withdrawal, because even slight worsening of bleeding could have severe consequences. The types of surgery concerned are mainly neurosurgery, prostate surgery, surgery for the posterior segment of the eye, and abdominal aorta surgery. However, the majority of operations can be conducted without withdrawing aspirin.8Any worsening in bleeding that has been observed so far has generally not led to an increase in the number of blood transfusions in patients taking aspirin, including those undergoing cardiac surgery. In addition, some surgical procedures must be performed while the patient is taking aspirin, and aspirin is introduced preoperatively. This is the case, for example, for carotid endarterectomy and prosthetic femoral-popliteal bypass surgery, as outlined in the evidence-based guidelines of the Seventh American College of Chest Physicians conference on Antithrombotic and Thrombolytic Therapy.9
Another factor to be taken into consideration is perioperative platelet activation in patients with vascular disease. Perioperative infarction occurs most commonly within 48 h of surgery. This has been widely confirmed because physicians use sensitive markers of cardiac ischemia such as troponin I. In such patients, even if platelets are not always activated, they are prone to be activated.10Consequently, antiplatelet treatment either should not be withdrawn or should be resumed as soon as possible.
Finally, yet another reason for reconsidering aspirin withdrawal is the increasing number of retrospective case series reporting acute coronary syndrome, acute peripheral arterial syndromes, or acute cerebral events in patients who discontinued aspirin before surgery.7,8,11–13
Routine withdrawal of aspirin must therefore be challenged, and practice may even have to change. A large randomized double-blind study (“Stratagem”) comparing continuing aspirin (75 mg) or discontinuing aspirin (placebo) during the 10 days preceding scheduled surgery is currently ongoing in France. The primary endpoint is a composite variable with thrombotic complications, bleeding, and death as components. Results are expected by the end of 2007 (verbal personal communication, June 2005, Jean Mantz, M.D., Ph.D., Professor, Department of Anesthesia and Critical Care, Beaujon University Hospital, Clichy, France).
Therefore, the dogma of aspirin withdrawal before surgery must be reconsidered, and we must move toward a case-by-case analysis. Available published data, which are admittedly still too scarce, seem now to favor the maintenance of antiplatelet treatment, in particular aspirin, in many clinical situations. NSAID substitution must be evaluated clinically with either dexibuprofen or other short-life potent NSAID agents on a risk-benefit basis. However, taking into account the iatrogenic risk of even short-term use of NSAIDs and the potential increase in perioperative bleeding,14the question could be, Preoperative aspirin substitution with NSAIDs: Isn't it already too late?
Department of Anaesthesiology and Intensive Care, Hotel-Dieu University Hospital, Paris, France. marc.samama@htd.aphp.fr
