From the Department of Medicine, Long Island Jewish Medical Center, the Long Island Campus for the Albert Einstein College of Medicine, New Hyde Park, NY. The authors have no relevant financial interest in this article.
The American College of Physicians recommends perioperative use of β-blockers for certain patients to improve outcomes after surgery. Study of physician behavior with respect to guidelines and recommended practices have shown that β-blockers have been underutilized after myocardial infarction. We evaluated physician concordance with the perioperative use of β-blockers along with a specialty-related difference in the frequency of perioperative β-blocker use.
To determine perioperative use of β-blockers, we retrospectively analyzed the medical charts of adult patients who underwent open cholecystectomy at a tertiary care medical center from December 1997 through December 2001. Patients met criteria for perioperative β-blocker use if they had a history of coronary artery disease or if they had the presence of 2 or more of the following risk factors: 65 years or older; history of hypertension, diabetes mellitus, or hypercholesterolemia; or current smoking.
Among the 336 cases of cholecystectomy reviewed, criteria for β-blocker use were met in 146 patients (43%) who did not have emergency operations and/or contraindications to β-blocker use. Of these 146 patients, 123 (84%) had a documented preoperative medical evaluation by a physician in the medical chart. There were 44 patients (30%) receiving β-blockers prior to admission, and 102 patients (70%) were not receiving β-blockers. Of those 102 patients not receiving β-blockers at admission but who meet criteria for their use, 94 (92%) were not started on β-blocker therapy preoperatively. Of the 18 patients evaluated by a cardiologist, 4 (22%) were started on β-blocker therapy compared with 3 (6%) of 47 patients evaluated by a noncardiologist physician (P = .08).
Perioperative β-blocker therapy is underutilized in patients with risk factors for coronary artery disease despite evidence that its use in appropriate individuals may be lifesaving.
β-Blockers have been used for various clinical conditions. Several major multicenter trials have clearly demonstrated the benefit of β-blockade in patients with coronary artery disease (CAD).1- 4 More recently, perioperative use of β-blockers has been shown to reduced myocardial ischemia and improve long-term cardiac outcome in patients undergoing noncardiac surgery.5- 7 The study by Mangano et al5 has shown that administration of atenolol before and after noncardiac surgery among patients with or at high risk for CAD can significantly reduce long-term mortality and cardiovascular morbidity. Based on data from the study by Mangano et al, the American College of Physicians (ACP) has recommended the perioperative use of β-blockers in patients who have or are at risk for CAD.8
To assess physicians' concordance with the ACP guideline for the perioperative use of β-blockers, we performed a retrospective study at our institution. We evaluated the use of perioperative β-blockers in patients undergoing a noncardiac surgery and also compared specialty-related differences in the frequency of perioperative β-blocker use.
The study was performed at a tertiary care teaching hospital. The hospital serves a diverse community. Attending physicians supervise residents and fellows in all cases. In most cases, medical consultation or perioperative clearance on a surgical patient is performed by the patient's private physician or internist. In some cases, however, hospitalists perform the perioperative evaluation.
After obtaining approval from the institutional review board at our institution, medical charts of adult patients diagnosed as having cholecystectomy between December 1997 and December 2001 were procured. Using an International Classification of Diseases, Ninth Revision (ICD-9) codebook, we obtained the code for the diagnosis for open cholecystectomy. The charts were identified using TSI (Transitional Systems Inc), Atlanta, Ga, an administrational database containing procedural and diagnosis codes. Using this method, we reviewed 336 charts. No patient was contacted for purposes of our study at any time.
We prepared a questionnaire based on the recommendations from the ACP for perioperative β-blocker use. An abstractor sheet was designed to record the captured data. Three of us (A.K.S., S.A., and H.D.) then recorded the data. A κ statistic was calculated to measure the degree of interobserver agreement. In addition to basic demographic characteristics, we included clinical variables to be assessed for perioperative β-blocker candidacy according to the ACP recommendations (age ≥65 years, diabetes mellitus, hypertension, current smoking, hypercholesterolemia, and CAD). Age was obtained from the face sheet of the medical chart. A patient was considered to have diabetes if stated in the chart or was using any medication for diabetes (eg, α-glucosidase inhibitors, biguanides, thiazolidinediones, sulfonylureas, and insulin). For hypertension, only a chart history qualified for the purpose of data abstraction; an incidental finding of hypertension during the hospitalization did not qualify. Smoking history data (only current) was obtained by review of historical notes. Chart diagnosis of hypercholesterolemia or statin drug therapy established this condition for the purpose of our study. Coronary artery disease was defined as a chart diagnosis of history of angina (stable or unstable), myocardial infarction, coronary artery bypass, invasive coronary intervention, or atypical chest pain with a positive stress test result. Patients were considered to have an indication for perioperative β-blocker use if they had CAD as stated previously and/or 2 or more of the aforementioned risk factors (age ≥65 years, diabetes, hypertension, hypercholesterolemia, and current smoking).
Absolute contraindications to β-blocker use were also recorded: active bronchospasm; third-degree heart block; bradycardia (heart rate <55 beats/min); hypotension (systolic blood pressure <100 mm Hg); active heart failure; or allergy to β-blockers. The surgery was considered emergent if it was documented in the medical chart and/or if it was performed in patients within 24 hours of their presentation in emergency department with acute abdominal pain or within 24 hours of the development of abdominal pain during the hospitalization. Cases for which the evaluating physicians did not recommend perioperative β-blockers based on the criteria of Mangano et al5 but for which β-blockers were used postoperatively for various other medical indications such as hypertension, tachycardia, and myocardial ischemia were also recorded. In the final analysis, those cases were considered as not including perioperative β-blocker use.
Medical charts of patients who were started on β-blocker therapy were further analyzed. According to the ACP recommendations, we assessed the administration of β-blockers preoperatively and postoperatively, as recorded from medication sheets. Additionally, we tracked these patients until the day of discharge or postoperative day 7 (whichever came first) to determine the degree of β-blockade, as obtained from vital signs sheets and/or progress notes. Patients were considered to have optimal degree of β-blockade if the average heart rate was less than 65 beats/min. We further analyzed the charts to see whether the patients who did not have optimal degree of β-blockade had their β-blocker dosage titrated to achieve the heart rate goal. The number of physicians who performed preoperative surgical assessment and their subspecialties were recorded.
For statistical analysis, the Fisher exact test was used for samples proportions to perform statistical analysis. Statistical significance was defined as a 2-tailed P value less than .05.
The κ statistic did not reveal significant interobserver disagreement. Among the 336 cases of open cholecystectomy we reviewed, 157 (47%) patients met the criteria of Mangano et al5 for β-blocker use. From this group, 11 patients (7%) were excluded because their surgery was considered emergent (7 patients) and/or because they had contraindications to β-blocker use as defined by Mangano et al (5 patients [2 were bradycardic and 3 were hypotensive]). This left 146 patients (93%) for analysis. The characteristics of these patients are given in Table 1.
Our primary objective was to measure the perioperative use of β-blockers. Of those 146 patients who met the criteria for β-blocker use, 44 patients (30%) were taking β-blockers for various medical indications before their preoperative evaluation. Of the remaining 102 patients not using β-blockers at admission but who met the criteria for their use, 8 (8%) were started on preoperative β-blocker therapy by the evaluating physician and 94 (92%) were not started on β-blocker therapy preoperatively. Of 94 patients that were not started on β-blocker therapy, 77 had documented preoperative evaluations by 47 different physicians, and a physician did not evaluate more than 3 patients.
Of 94 patients who were not started on β-blocker therapy preoperatively, 2 (2%) did not receive β-blockers despite it being recommended by the evaluating physicians and 10 (11%) patients received β-blockers postoperatively for various other reasons. β-Blockers were given to 5 patients for tachycardia including new-onset atrial fibrillation, 4 patients for high blood pressure, and 1 patient for postoperative myocardial infarction. None of these patients were recommended for perioperative β-blocker therapy by the evaluating physician. Among 94 patients who did not receive β-blockers preoperatively, 22 (23%) had a history of CAD. Five (5.3%) of these 94 patients had a history of asthma, 2 (2.1%) had a history of congestive heart failure, and 6 (6.4%) had a history of chronic obstructive pulmonary disease, though these comorbidities were stable at the time of the preoperative evaluation. The characteristics of these patients are elaborated in Table 2.
Of those 8 patients who received β-blockers preoperatively, their preoperative and postoperative mean ± SD heart rate was 79 ± 13.3 beats/min and 86 ± 12.9 beats/min, respectively. Only 2 patients (25%) had optimal degree of β-blockade preoperatively, and 1 patient (13%) had optimal degree of β-blockade postoperatively. In 3 patients (38%), the β-blocker dosage was titrated by the evaluating physicians to achieve the heart rate goal.
Of 102 patients, 4 (22%) of the 18 patients evaluated by cardiologists were started on β-blocker therapy compared with 3 (6%) of the 47 patients evaluated by noncardiologist physicians (P = .08; odds ratio, 4.1; 95% confidence interval, 0.8-21.0). In 1 patient who received perioperative β-blockers, the specialty of the evaluating physician was not known. Four (50%) of 8 patients who received perioperative β-blockers had a history of CAD compared with 22 (31%) of 72 patients who did not receive β-blockers (P = .4; odds ratio, 2.2; 95% confidence interval, 0.5-9.9).
Approximately 30 million patients undergo noncardiac surgery every year in the United States. About 3 million of these patients are determined clinically to be at high risk of postoperative cardiac events, and more than 1 million patients experience cardiovascular morbidity and mortality.9,10 Over the past several decades, the cardiovascular assessment of patients undergoing noncardiac surgery has been intensely investigated, and several perioperative risk factors have been identified.11,12 Of those risk factors, postoperative myocardial ischemia is the single most important and potentially avoidable risk factor for cardiovascular complications and mortality after noncardiac surgery. Many therapies have been tried to modify this elevated coronary risk. Perioperative use of calcium channel blockers, nitroglycerin, and aspirin have not been shown to be of significant benefit.13- 15 On the other hand, perioperative use of β-blockers has clearly been shown to decrease cardiac morbidity and mortality in high-risk patients.5- 7
The earlier nonrandomized studies of perioperative β-blocker use have demonstrated a reduction in cardiac morbidity.16,17 The study by Mangano et al,5 however, was the first randomized placebo-controlled trial that compared the effectiveness of atenolol and placebo on overall survival and cardiovascular morbidity in patients with or at risk for CAD undergoing noncardiac surgery and clearly demonstrated a reduction in cardiovascular morbidity and mortality. Of the 200 patients studied, about 40% had definite CAD and 60% were at risk for CAD. Approximately 40% of patients underwent major vascular surgery and 60% had nonvascular surgery. Atenolol was administered intravenously before and immediately after surgery and orally thereafter for the duration of hospitalization. Adjustment in the atenolol dosing was performed according to patients' heart rates. Although not all patients had optimal β-blockade, during treatment the average heart rate was significantly lower in the atenolol group compared with the placebo group (75 beats/min vs 87 beats/min). There was a significant reduction in deaths and an increase in event-free survival in the atenolol group that was evident over 6 months. This survival benefit extended to 2 years after surgery, at which the mortality rate was 10% in the atenolol group vs 21% for placebo and the event-free survival was 83% vs 68%. Moreover, this was accomplished without any detectable increase in the frequency of adverse effects. Later, another randomized controlled trial performed in high-risk patients undergoing major vascular surgery also demonstrated a reduction in the incidence of cardiac death and nonfatal myocardial infarction with perioperative administration of bisoprolol.7
The biological rationale for β-blocker use includes its ability to decrease exaggerated sympathetic response during the perioperative period, associated with an increased myocardial oxygen demand and resultant postoperative ischemic events.18- 20 The beneficial effect of β-blockade may also be due to (1) the reduction in shear stress across atheromatous plaques, thereby reducing the incidence of plaque rupture and consequent acute coronary thrombosis, or (2) a reduction in the incidence of arrhythmias.21,22 Based on the data from the study by Mangano et al,5 the ACP suggests the perioperative use of β-blockers.8 In patients with intermediate risk for perioperative cardiac events, noninvasive cardiac tests are recommended for further risk stratification only in patients who are undergoing vascular surgery. For all patients with CAD or risk factors for CAD (according to the criteria of Mangano et al5) who are undergoing major surgery, however, perioperative use of atenolol is recommended by the ACP unless the patient has a significant contraindication.
We found underutilization of perioperative β-blockers in patients with risk factors for CAD despite evidence that their use in appropriate individuals may be lifesaving. More than two thirds of patients in our study were found to have or be at risk for CAD by the criteria of Mangano et al5 and met the criteria for β-blocker use yet did not receive perioperative β-blockers. Of notable importance, the lack of prescribing β-blockers was not limited to a few physicians but was universally present among different physicians. Of the 94 patients who met the criteria but did not receive β-blockers, we found that only 13 patients (14%) had a history of asthma, congestive heart failure, or chronic obstructive pulmonary disease, though these comorbidities were stable at the time the preoperative evaluation, and none of these 94 patients had any absolute contraindication to β-blocker use. Additionally, there were 22 patients (23%) who, despite having a history of CAD, did not received β-blockers. Of note, most of those patients who were given perioperative β-blockers did not demonstrate an optimal degree of β-blockade, and only in one third of cases did the evaluating physicians titrate the dosage to achieve an optimal heart rate. There was no significant difference in the perioperative use of β-blockers between patients who were evaluated by a cardiologist compared with patients who were evaluated by a noncardiologist physician.
Although our data are retrospective and from a single institution, they represent the practice behavior of a large group of physicians. Several factors may account for the low physician concordance with the guidelines for perioperative use of β-blockers. Doubt about the applicability of trial data to a particular patient may be an important factor in clinical decision making. For example, despite the convincing and strong evidence of benefit of perioperative use of β-blockers in several well-designed clinical trials, as a group, studies that support perioperative β-blocker use are relatively small, with a total enrollment of about 700 patients.23 Furthermore, the benefit of perioperative use of β-blockers is less clear in patients who are at intermediate risk for cardiac events compared with patients who are at high risk. Physicians, therefore, may perceive small benefit of perioperative β-blocker use in their patients and therefore may be reluctant to change their practice based on the limited number of trials. Other explanations for lack of agreement with the guidelines for perioperative use of β-blockers may include deficiency in knowledge, limited motivation to change practice, or concern of adverse drug effects, among others.
In conclusion, our study demonstrated underutilization of β-blockers despite convincing data regarding their clear benefit in patients at risk for cardiac events. Improved understanding of the reasons for the underutilization of β-blockers in the perioperative setting may be helpful for designing future interventions to diminish inappropriate variations in clinical practice, to control costs, and to improve patient outcomes.
Corresponding author and reprints: Joseph Mattana, MD, Department of Medicine, Long Island Jewish Medical Center, New Hyde Park, NY 11040 (e-mail: email@example.com).
Accepted for publication May 13, 2003.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Internal Medicine editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 36
Customize your page view by dragging & repositioning the boxes below.
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
Evidence From Guidelines
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.