To the Editor: Although preoperative administration of β-blockers has been acknowledged as important in patients with risk factors for coronary artery disease,1 recent results from the POISE trial2 suggest that caution should be used when using these medications immediately before anesthesia and surgery. A 68-yr-old 70 kg, 171 cm man was scheduled for lumbar discectomy. His medical history included asymptomatic coronary heart disease and aortoiliofemoral bypass, and current treatment included aspirin, a statin, and analgesics. His functional capacity could not be assessed because of back pain, but preoperative maximal dobutamine stress echocardiography performed 1 mo before surgery revealed septal akinesia that remained unchanged throughout the entire procedure. The ejection fraction was 65%. His cardiac disease was described as stable and in accordance with ACC/AHA guidelines,1 surgical risk was considered as intermediate and he received a β-blocker (5 mg of bisoprolol) the morning of surgery. Hydroxyzine (100 mg) was given on both the evening before and the morning of surgery and alprazolam (0.25 mg) was given the morning of surgery. Aspirin and the statin medication were maintained. The patient arrived in the operating room 1 h after the administration of bisoprolol and premedication. His arterial blood pressure was 125/80 mm Hg, his heart rate was 90 bpm, and hemoglobin oxygen saturation (Spo2) was 92%. While breathing oxygen, induction of anesthesia (90 min after oral administration of bisoprolol) using sufentanil 25 μg, thiopental 300 mg, and atracurium 35 mg was immediately followed by arterial hypotension (65/40 mm Hg), without tachycardia (65 bpm). Three bolus doses of 50 μg each of phenylephrine and four bolus doses of 6 mg each of ephedrine were given in a short period, along with 1000 mL of Ringer's lactate and 1000 mL of modified fluid gelatin. A continuous infusion of phenylephrine was immediately started at a rate of 500 μg/h and systolic arterial pressure remained around 85 mm Hg. Anesthesia was maintained with sevoflurane, with an alveolar concentration of 0.6 MAC. Fifty minutes after, induction of anesthesia surgery was cancelled because of the persistent hemodynamic instability, and transthoracic echocardiography showed a new dilated cardiomyopathy. Coronary angiography performed 1 day later revealed stenosis of the anterior interventricular artery and the ramus intermedius artery. Angioplasty with placement of a stent was performed on the latter stenosis. Anaphylaxis as a cause of the persistent arterial hypotension was excluded by the results of tryptase and histamin plasma levels. The timing of the onset of symptoms, i.e., immediately after induction of anesthesia, before surgical stress began, suggests that myocardial ischemia is not the cause but the consequence of the collapse. Takotsubo cardiomyopathy was ruled out by the patient's age and gender, the absence of typical ST elevation or T wave inversion, and the lack of echocardiographic features, such as transient apical and mild-ventricular regional wall motion abnormalities. Moreover, angiography showed two significant stenoses that could explain the symptoms.3 Although the ACC/AHA has recently recommended the use of preoperative β-blocker therapy in patients with cardiac risk,1 somewhat conflicting data concerning β-blockade and hypotension during anesthesia still exist. Although a recent retrospective database review did not identify preoperative β-blockade as a predictor of hemodynamic instability in the 10-min period after induction of general anesthesia,4 a recent meta-analysis on the influence of rigorous heart rate control as a measure of the efficacy of perioperative-blockade found that perioperative-blockade was associated with an increased incidence of bradycardia and congestive heart failure.5 Increased vagal tone is a well-described side effect of high-dose sufentanil anesthesia,6 and cases of sudden hypotension or bradycardia have been reported to follow induction of anesthesia with high dose of sufentanil in combination with benzodiazepine and nonvagolytic relaxant agents.7,8 Our dosage regimen for induction of anesthesia was in a normal range and especially sufentanil was used at low dose. The hypotension after induction of anesthesia, we report here, may have been caused by the cumulative effects of preoperative benzodiazepine, atracurium (a nonvagolytic relaxant agent), and sufentanil, precipitated by the introduction of bisoprolol aiming at decreasing the postoperative ischemic risk. Hence, the immediate preoperative period may be not the best time to start β-blockade therapy. Mathieu Boutonnet, MD Sylvain Ausset, MD Evelyne Lambert, MD Department of Anesthesiology Percy Military Hospital Paris, France [email protected] Guillaume de Saint Maurice, MD Yves Auroy, MD Cognitive Science Department IMASSA, Brétigny sur Orge France Helga Berbari, MD Department of Cardiology Percy Military Hospital Paris, France Bernard Lenoir, MD Department of Anesthesiology Percy Military Hospital Paris, France