We thank Drs. Avidan and Gozal for their interest in our article reporting on anesthetic-related cardiac arrest and its mortality. 1They are concerned that lack of detail about the fifteen reported cardiac arrests we determined to be attributable to anesthesia may invalidate the conclusion of the study, namely that our results more accurately reflect the risk of perioperative cardiac arrest and the real risk of anesthesia. Our study design was to prepare a case abstract from data obtained from the medical records using a standardized data collection form and submitting this abstract to a study commission. Using this information the study commission then judged whether anesthesia was attributable or contributory to the cardiac arrest. The fifteen cases exemplify the study commission's best judgment. Privacy and other considerations preclude provision of additional case details.

The question of who provided anesthesia care was raised. As noted in our article, anesthesia was provided by faculty, residents, and certified registered nurse anesthetists. It was our practice that an anesthesia faculty member was immediately available and responsible for every case.

To address concerns if a relative overdose of narcotic can lead to cardiac arrest in the PACU or the patient's room, the study commission judged that it could, based on the information available to them, and that the anesthesia provider was responsible or contributory. Neither event was related to nursing care.

In the case of the 65-yr-old patient, ASA physical status III who received 1 mg of midazolam as premedication and went on to cardiac arrest in the ambulatory surgical unit, these events did occur and the patient had not received any additional drugs prior to the arrest. The patient did have a complex medical history and had undergone several major operations in the past.

The two cases attributed to probable vagal reaction occurred with an anesthesia provider in attendance (even in the pediatric intensive care unit shortly after transport from the operating suite) and recovered uneventfully. No other cause could be determined.

Because general anesthesia is used in cases for AICD placement it is possible for dysrhythmias to occur related to anesthesia and before any placement of leads or elective induction of cardiac arrest.

Last, based on the information available to the study commission, it was their judgment that anesthesia was contributory to the perioperative myocardial infarction.

We would encourage other investigators and institutions to do similar studies of their patients receiving anesthesia and report the findings. In the future perhaps a national database, gathering information from each institution, will allow comparison between institutions as well as provide national statistics that are meaningful and accurate.

Newland MC, Ellis SJ, Lydiatt CA, Peters KR, Tinker JH, Romberger DJ, Ullrich FA, Anderson JR: Anesthetic-related cardiac arrest and its mortality: A report covering 72,959 anesthetics over 10 years from a US teaching hospital. A nesthesiology 2002; 97: 108–15