Introduction
The resting 12-lead electrocardiogram (ECG) is the one of the most widely used diagnostic tests in medicine, and preoperative ECG is the most commonly obtained cardiovascular diagnostic test before surgery. Many epidemiologic studies have demonstrated an association between abnormal ECG findings and an increased risk of death from cardiovascular causes in the general population. Evidence to support the value of routine preoperative ECG to predict adverse perioperative cardiovascular events is conflicting, however, in part because of the wide variability in study design, population, and clinical endpoints.
The routine use of many screening tests has been called into question. An ideal preoperative screening test should be inexpensive, have high positive and negative predictive values, add to information obtained from the clinical history and physical examination, and change or modify perioperative decision making to prevent perioperative complications. Extensive preoperative testing can lead to false-positive results, additional expensive and invasive workups, and unnecessary delay or cancellation of necessary procedures. Sandler demonstrated, in a prospective study of medical patients, that more than 50% of clinical diagnoses and nearly 50% of management decisions were based on history alone, and routine studies contributed to less than 1% of all diagnoses. Several studies in the surgical population have found that routine preoperative screening evaluations rarely found abnormal test results not predicted by history alone, and when abnormalities were detected, management was not significantly altered. Wilson et al and Narr et al demonstrated that fitness for elective surgery can safely be predicted by a history and physical examination, and tests can be obtained intraoperatively or postoperatively, as indicated.
More than 100 million ECGs are obtained annually, at a cost of approximately $5 billion. With more than 45 million inpatient and 53.3 million ambulatory procedures performed annually in the United States, preoperative screening undoubtedly accounts for many of the ECGs obtained. The prevalence of abnormal preoperative screening ECG results has been estimated to be anywhere between 25% and 50%; the clinical implication of abnormal ECG findings is less clear, however, in that a change in management was observed in 0% to 2.2% of patients. Callaghan et al found that 18% of all preoperative ECGs are ordered without a clear indication, whereas Nash et al found that 30% of preoperative ECGs are never interpreted by an anesthesiologist.
Thus it is important from the standpoints of both patient risk stratification and public health to evaluate which patients will benefit from preoperative ECG screening. Evidence to support or refute the use of preoperative ECG screening is conflicting in the literature. As such, although guidelines exist from several medical societies, there is no consensus as to who may benefit from preoperative ECG. The purpose of this chapter is to summarize the available data in different populations as well as to review the current recommendations from different medical societies.
Options
An ECG could be obtained on all adult patients or could be required only in patients with specific risk factors. Patient factors that may merit further evaluation include a known history of or risk factors for cardiovascular disease, poor functional status, or new physical examination findings suggestive of cardiovascular disease. The type and invasiveness of surgical procedure may also be considered. Historically, age has been used as a criterion for preoperative cardiac evaluation, although more recently this practice has been called into question. Current approaches to obtain a preoperative ECG should consider three key questions: (1) What is the likelihood of cardiovascular disease in this patient, (2) What is the risk of this surgical procedure, and (3) Will the results of this test change perioperative management?
Evidence
It is difficult to compare the current literature because of the wide variability in patient populations, outcomes measured, and overall study design. Despite these limitations, several general patient populations tend to emerge in the literature. We will discuss the current literature in the following groups: asymptomatic patients, patients with known risk factors for cardiac disease, the elderly population, and patients undergoing “high” versus “low” risk surgery.
Asymptomatic Patients
Evidence to support or refute routine preoperative ECG in asymptomatic patients undergoing nonvascular, noncardiac surgery is perhaps the most widely variable, in large part because of the differences in patient groups and outcomes measured. Carliner and colleagues prospectively evaluated 200 patients undergoing elective major noncardiac surgery under general anesthesia. Using a multivariable model, they found that ST-T wave abnormalities, abnormal Q waves, and left ventricular hypertrophy (LVH) on preoperative ECG were the only statistically significant independent predictors of perioperative cardiac events. A smaller series by Younis et al examined 100 patients undergoing major noncardiovascular surgery. Although Q waves on resting ECG were predictive of adverse perioperative cardiac events on univariate analysis, they were not significant on multivariate analysis.
A prospective evaluation of 660 patients undergoing noncardiac, nonvascular surgery by Biteker et al found that 394 (59.7%) of patients had at least one abnormality on preoperative ECG, and 127 (19.2%) had a change in preoperative management. Thirty patients (4.5%) underwent additional preoperative testing, and a diagnosis of new or unstable cardiac disease was made in 21 cases (3.1%). Twelve of the 30 went on to surgery without delay. Patients with an abnormal preoperative ECG had a higher incidence of perioperative cardiovascular events. On multivariate analysis, only QTc prolongation was an independent predictor of perioperative cardiovascular events.
Several studies refute the claim that preoperative ECG results change perioperative management in a healthy population. A systematic review by Munro et al found that preoperative ECG results were abnormal in up to 32% of cases and led to a change in management in less than 2% of cases, and the effect on patient outcome was unknown. Rabkin and Horne corroborate this claim with their finding of new ECG abnormalities in 165 of 812 patients in a retrospective analysis but a delay or cancellation in only 13 cases. None of the documented reasons for delay or cancellation was related to the preoperative ECG abnormality. The choice of anesthesia was influenced in only two cases. Patient outcomes were not evaluated. Perez et al retrospectively evaluated 3131 patients of whom 2406 had a preoperative ECG. Only 5.6% had an unexpectedly abnormal ECG result, and a change in management occurred in only 0.5% of cases.
In a retrospective review, Turnbull and Buck found that of 101 abnormal preoperative ECG results, only four were significant by the criteria of Goldman et al, and no preoperative change in management occurred in any case. Four patients had a cardiac complication, and in two of these cases, the cardiac risk was apparent from the history and physical examination alone. Gold et al found similar results, in that less than 2% of patients with abnormal ECG results experienced an adverse perioperative cardiovascular event and preoperative ECG was useful in only half of the cases. On a review of the literature, Goldberger and O’Konski did not support routine preoperative ECG for all-comers but rather the selective use of screening for subsets of patients, including those with signs or symptoms of cardiac disease or those with risk for occult heart disease. Similarly, Barnard et al found preoperative screening ECG to be of limited value for relatively healthy patients.
Risk Factors
Over the last several decades, many studies have validated certain disease processes that are associated with adverse perioperative cardiovascular outcomes. Although they may be clinically asymptomatic, patients with ischemic heart disease (IHD), congestive heart failure, cerebrovascular disease, diabetes mellitus, and chronic renal insufficiency are at increased risk of cardiovascular morbidity and mortality. Hollenberg et al used continuous perioperative ECG monitoring to identify predictors of postoperative cardiac ischemia in patients at high risk of or with known coronary artery disease. They identified five major predictors for perioperative ischemia, including four factors ascertainable by clinical history (definitive history of coronary artery disease, hypertension, diabetes mellitus, or use of digoxin) and LVH by ECG. The clinical risk increased with the number of risk factors present.
Landesberg and colleagues investigated the association between preoperative ECG abnormalities and perioperative myocardial ischemia, infarction, and cardiac death in 405 patients undergoing major vascular surgery. They found that LVH by voltage criteria, ST segment depression, or both better predicted postoperative cardiac morbidity and mortality than clinical risk factors, including history of myocardial ischemia or infarction, angina pectoris, or diabetes mellitus.
Payne and colleagues performed a prospective observational cohort study of 345 patients undergoing major vascular surgery or laparotomy to evaluate the correlation between abnormal preoperative ECG and postoperative adverse cardiac events. They found that patients with an abnormal preoperative ECG had a significantly higher incidence of major adverse cardiac events. Multivariable analysis demonstrated that a clinical history of hypertension or prolongation of QTc or left ventricular strain by ECG were predictive of postoperative adverse cardiac events. More importantly, however, they examined the relationship between a history of known IHD and an abnormal result on preoperative ECG. They found that patients with a history of IHD and a normal result on preoperative ECG had the lowest rate of adverse postoperative cardiac events (2.4%) compared with no IHD and a normal result on ECG (8.6%), IHD and an abnormal result on ECG (24.2%), and no IHD and an abnormal result on ECG (20.3%) ( p = 0.001).
Jerger et al prospectively examined 172 patients with known coronary artery disease undergoing major noncardiac surgery to determine the association between preoperative ECG and long-term outcomes of all-cause mortality and major adverse cardiac events at 2 years. The overall prevalence of preoperative ECG abnormalities was between 38% and 53%, depending on the criteria used. After controlling for baseline clinical findings, the authors found ST depression and faster heart rate to be independent risk factors for all-cause mortality, as were renal failure and prior revascularization. Faster heart rate, advanced age, hypertension, peripheral arterial disease, and congestive heart failure were independent predictors of major adverse cardiac event.
Other studies, however, failed to find significant utility of routine preoperative ECG in this patient population. Tait and colleagues performed a retrospective chart review of 1000 American Society of Anesthesiologists (ASA) 1-2 patients undergoing low- to intermediate-risk surgery. Patients were allocated to cardiovascular risk or no risk as defined by a history of hypertension, hyperlipidemia, arrhythmia, diabetes mellitus, peripheral vascular disease, angina, or coronary artery disease. They found that patients with cardiovascular risk factors were more likely to have abnormal ECG results; however, there was no difference in the occurrence of adverse perioperative cardiac events.
In another study, Noordzij et al retrospectively studied 23,036 patients undergoing noncardiac surgeries with a primary endpoint of 30-day cardiovascular death. Cardiovascular death was observed in 199 patients (0.7%), and the incidence was higher in those with abnormal preoperative ECG results; however, the absolute difference in the incidence of cardiovascular death in patients undergoing low- or intermediate-risk surgery was only 0.5%, which casts doubt on its clinical usefulness in this population.
van Klei and colleagues evaluated 2967 patients undergoing noncardiac surgery and found that both left and right bundle branch blocks identified on the preoperative ECG were associated with an increase in postoperative myocardial infarction and death but failed to predict adverse perioperative cardiac events beyond clinical risk factors identified by history alone.
Preoperative Electrocardiogram and the Elderly
A wealth of epidemiologic data supports an increased prevalence of coronary artery disease with increasing age. The probability that a previously asymptomatic man at average risk will have myocardial ischemia, myocardial infarction, or cardiac death is less than 4 per 1000 at 40 years of age; this number increases to 18 per 1000 at 60 years of age. The prevalence of cardiovascular disease in patients 80 years and older is estimated to be greater than 30% in patients seen for noncardiac surgery. Furthermore, at least 25% of myocardial infarctions in the aging population are believed to be clinically silent, and the risk for recurrent cardiac ischemia is similar to those with recognized cardiac events. It is for this reason that some advocate routine preoperative ECG screening for the elderly. Nevertheless, data to support age alone as a valid reason for routine ECG screening are variable.
Several studies have demonstrated an increased incidence of abnormal ECG results in patients with advanced age. Seymour and colleauges suggest that, given the high prevalence of abnormal preoperative ECG results in the elderly population, preoperative screening should be performed routinely to ascertain “new” from “old” abnormalities, despite its poor ability to predict postoperative cardiovascular complications. Roizen suggests, on the basis of pooled data from multiple studies, routine preoperative ECG screening for men older than 40 years and women older than 50 years for all moderate- to high-risk procedures. Correll and colleagues found several risk factors, including history of heart failure, hyperlipidemia, angina, myocardial infarction, valvular heart disease, and age older than 65 years, to be predictive of a preoperative ECG result that would potentially affect perioperative management. In fact, in this study, age older than 65 was the most predictive risk factor of abnormal preoperative ECG results. Of note, there were no statistical differences in major postoperative cardiac complications between the two groups; this study was not powered, however, to detect differences in this endpoint.
Other studies refute the usefulness of preoperative ECG in the elderly population. Liu and colleagues prospectively observed 513 patients aged 70 years or older undergoing noncardiac surgery. Abnormal preoperative ECG results were found in 386 (75.2%) of patients, but the presence of abnormalities on preoperative ECG was not associated with an increased risk of postoperative cardiac complications. They also examined the possibility that patients with abnormal preoperative ECG results had changes in the preoperative or intraoperative period that might affect outcomes. None of the cases cancelled or postponed by the anesthesiologist was due to ECG abnormalities. Intraoperative care was the same in terms of use of beta- or calcium channel blockade, nitroglycerin, and invasive hemodynamic monitoring.
Schein and colleagues prospectively assigned 19,189 elderly patients scheduled to undergo cataract surgery to either routine preoperative testing or no preoperative testing. They found neither a difference in the overall rate of intraoperative or postoperative complications nor a difference in intraoperative or postoperative events.
Surgical Procedure
It has been widely demonstrated in the literature that the risk of cardiovascular morbidity and mortality is correlated with the type of surgery ; that is, “high-risk” procedures such as emergency or vascular surgery are associated with a higher rate of adverse perioperative events than “low-risk” procedures such as ambulatory or endoscopic procedures. Perhaps the mostly extensively studied group is patients undergoing major vascular surgery, who, by virtue of both high-risk surgery and underlying disease processes, are at increased risk of perioperative cardiac events. Patients undergoing lower risk procedures are at significantly lower cardiac risk. In the ambulatory surgery population, for example, preoperative ECG has not been shown to be predictive of adverse perioperative events, presumably because of the relatively low risk of the procedures performed as well as the relatively healthy patient population. As such, the decision to obtain a preoperative ECG should take into account the relative risk of the surgery itself in addition to the individual patient’s clinical risk factors and history.
Related posts:
Is a Preoperative Screening Clinic Cost-Effective?
Should Patients with Stable Coronary Artery Disease Undergo Prophylactic Revascularization before Noncardiac Surgery?
Is There a Best Approach for Patients with Difficult Airways: Regional versus General Anesthesia?
Does the Choice of Fluid Matter in Major Surgery?
Are There Special Techniques in Obese Patients?
Is Propofol Safe If Given by Nonanesthesia Providers?
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
