This Editorial View accompanies the following article: Rosenberg-Adamsen S, Lie C, Bernhard A, Kehlet H, Rosenberg J: Effect of oxygen treatment on heart rate after abdominal surgery. Anesthesiology 1999; 90:380-4.
MANY years ago it was said that it was no longer easy to conduct research because all the simple experiments had already been done. In the current issue of Anesthesiology, Rosenberg-Adamsen et al. demonstrate the fallacy of that statement. After abdominal surgery, in random order, each of 100 consecutive patients was given air or oxygen via nasal cannulae. Supplemental oxygen increased arterial oxyhemoglobin saturation as assessed by pulse oximetry (SpO2) and decreased heart rate. Oxygen decreased the heart rate when the initial SpO2was 90-98% but not when the SpO2was more or less (probably because this latter subgroup contained only six patients). Supplemental oxygen decreased the heart rate for all groups of patients categorized by basal heart rate. The importance of these observations lies in the putative link between tachycardia and postoperative myocardial ischemia. Increased heart rate increases myocardial oxygen consumption. In patients with coronary artery disease, this increase in myocardial work and oxygen consumption in the presence of limited ability to increase oxygen supply can result in myocardial ischemia. Postoperative myocardial ischemia in patients with or at high risk for coronary artery disease is associated with tachycardia, and decreasing heart rate in the postoperative period with a [small beta, Greek]-adrenergic antagonist or an [small alpha, Greek]2agonist decreases postoperative myocardial ischemia. Thus, it is possible that reducing heart rate by any means in the postoperative period might reduce the incidence of postoperative myocardial ischemia.
This simple study, as is true for many investigations, raises several questions. The result tantalize us with possibilities. In a previous study of 22 patients, this group of investigators found an association of tachycardia, electrocardiographic ST depression, and hypoxemia in several patients. Unfortunately, the current study does not present data regarding myocardial ischemia. Heart rate is a surrogate; it is not the variable of primary interest. Surrogates are useful only insofar as their sensitivity and specificity allow. Postoperative tachycardia has many possible causes, including pain, hypovolemia, fever, increased sympathetic activity, decreased parasympathetic activity, medications, anemia, hypercarbia, and hypoxia. Although oxygen decreased the surrogate measure (heart rate) in all heart rate groups, the effect of oxygen on the link (myocardial oxygen consumption) to the measure of interest (myocardial ischemia) likely would have varied substantially among groups. For patients with heart rates of 80-90 beats/min, the mean reduction of 4-5 beats/min by oxygen would have decreased myocardial oxygen consumption by only approximately 5%. However, for those patients with the highest heart rates, the reduction would have been more substantial, perhaps 20%.
Although 100 patients were studied, when the data are examined by subgroups (basal heart rate, basal SpO2, use of epidural analgesia, postoperative day), the numbers in each group become substantially smaller, and for some groups (for example, SpO2<90), the study lacked sufficient power. The investigators also found that patients treated with postoperative epidural analgesia had lower heart rates than those without epidural analgesia and that supplemental oxygen decreased heart rate equally in these two groups. A previous, randomized, prospective study in patients with or at high risk for coronary artery disease undergoing lower limb vascular surgery did not find a difference in myocardial ischemia between patients randomly allocated to general anesthesia or to epidural anesthesia plus 24 h of epidural analgesia. Unfortunately, in that study, the epidural analgesia was not continued through the entire high-risk period for myocardial ischemia. [4,9]Furthermore, the study would have required a sample size several fold greater for sufficient power to enable detection of a difference.
To what extent are the previous findings of association of postoperative myocardial ischemia with tachycardia the result of hypoxemia? Is oxygen as efficacious as the use of a [small beta, Greek]-adrenergic antagonist or an [small alpha, Greek]2agonist in preventing postoperative myocardial ischemia? Is epidural analgesia efficacious for reducing postoperative myocardial ischemia? Oxygen therapy is inexpensive, and its use may produce a salutary effect. The answer to these important questions and others raised by the study of Rosenberg-Adamsen et al. will require larger studies. Although the study did not provide the answers to these questions, its importance is that it raised them.
Richard B. Weiskopf, M.D.
Professor; Departments of Anesthesia and Physiology; Staff, Cardiovascular Research Institute; University of California; San Francisco, California 94143-0648;firstname.lastname@example.org