Perioperative β‐blockade: how long does the perioperative period last?

Sir, We read with great interest the editorial of T. Oksanen and M. Hynynen1 on perioperative β-blockade. We agree with the conclusion that high-dose β-blockade in intermediate-risk patients may be harmful and should be considered only for selected patients. We remind that Lindenauer,2 in a retrospective cohort study of 782,969 patients, previously underlined that perioperative β-blockade was associated with a reduced in-hospital mortality among high-risk, but not low-risk, patients undergoing major noncardiac surgery. According to us, the results of the POISE study3 raise two problems. Firstly, POISE highlights the importance of perioperative hypotension on the patient's outcome. In 2005, Monk et al.4 showed that intraoperative hypotension was an independent predictor of increased mortality at 1 year. In POISE, hypotension appeared as an independent predictor of death and stroke. We regret that POISE did not present a multivariable analysis of the risk of myocardial infarction, including hypotension as a predictive factor. Indeed, in this study, myocardial infarction is the second cause of death after sepsis and infection whereas the cardiovascular death rates did not differ between the two groups. Several studies have found a strong correlation between hypotension and ischemia, both in the operative setting5–7 and in patients treated for hypertension.8,9 Hypotension could have reduced the global benefit of β-blockade by increasing the rate of myocardial infarction and cardiovascular death in the metoprolol group. Two previous randomized-controlled trial10,11 had already shown that the incidence of hypotension was higher with perioperative β-blockade and it is important to point out that β-blockade was initiated on the day or the day before the surgery. This pleads for an earlier onset of β-blockade in order to achieve a titration of the dosage regimen and to minimize the risk of several effects such as hypotension or bradycardia. Secondly, we would like to raise a concern that Oksanen and Hynynen did not mention: the length of follow-up. We wonder whether the benefit of a perioperative β-blockade could have been higher if assessed with a longer follow-up. Indeed, many data are now available to think that the impact of post-operative myocardial infarction under patient's outcome is still important several months or even up to 2 years later.12–19 Consequently, a 30-day or an in-hospital follow-up may be too short to assess the benefits and risks of a perioperative care. To conclude, one of the corollaries of the POISE study that the new consensus guidelines should consider is that perioperative β-blockade should be initiated cautiously several weeks pre-operatively and that the effectiveness of such preventive policies should be assessed through a long-term follow-up.