- Email: [email protected]
The safety and effectiveness of esmolol in the perioperative period in patients undergoing abdominal aortic surgery
The Safety and Effectiveness of Esmolol in the Perioperative Period in Patients Undergoing Abdominal Aortic Surgery Timothy N. Harwood, MD, John Butterworth, MD, Richard C. Prielipp, MD, Roger L. Royster, MD, Kimberley Hansen, MD, George Plonk, MD, and Richard Dean, MD Objectives: To determine (1) if perioperative use of esmoIol in major vascular surgery patients provides strict heart rate (HR) control, (2) what doses of esmolol are required to do this, and (3) does this control influence myocardial ischemia or result in adverse consequences. Design: Prospective study of 40 patients randomized to two groups: The HR was controlled to either less than 80 beats/min (group 80) or less than 110 beats/min (group 110) using esmolol. Patients were monitored continuously for electrocardiographic changes perioperatively. HR control began after induction of anesthesia and continued for 48 hours thereafter. Setting: Operating room and intensive care unit. Patients: Patients undergoing abdominal vascular surgery involving aortic cross-clamping. Interventions: Esmolol was titrated until the target HR was met. Measurements andResults: Only one patient demon-
strated an adverse effect. The median infusion rates were 100 and 12.5 pg/kg/min for groups 80 and 110. Target HR was met less in group 80 than in group 110, primarily in the postoperative period. Ischemia patterns were not significantly different between groups. Conclusion: Using esmolol for HR control in the intraoperarive period for abdominal vascular surgery patients is effective and safe. HR control was much less effective in the postoperative period, but esmolol is safe when used at recommended doses. Further study with a larger number of patients is necessary to determine whether strict HR control with esmolol affects the incidence of myocardial ischemia or infarction in this patient population.
p
Based on these studies, the authors designed a protocol that maintains a normal (or lower) heart rate (HR) with a [3-adrenergic receptor blocker to reduce the risk of perioperative myocardial ischemia and infarction. A short-acting agent was used so that dosage adjustments could be accomplished to match changes in autonomic activation and stress that occur during and after major surgery. The feasibility and safety of such prolonged infusion of esmolol in patients with comorbid illnesses seen in vascular surgery patients have not been established. The present research was designed to determine (1) whether strict HR control can be maintained perioperatively using esmolol, (2) what doses of esmolol are required to accomplish HR control, and (3) whether there might be any influence on myocardial ischemia or adverse consequences of this approach.
ATIENTS UNDERGOING major abdominal vascular sur-
gery are at higher risk for perioperative myocardial infarction and cardiac mortality than patients undergoing most other types of surgery. 1 These risks result from the greater prevalence of coronary atherosclerosis in this patient population, coupled with the superimposed metabolic and thrombotic changes of major surgery. Nevertheless, most patients undergo vascular surgery without adverse myocardial events. A subset of the population with perioperative myocardial ischemia appears to be at greater risk of adverse cardiac outcome, z3 Landesberg et ai4 determined that postoperative ischemia lasting longer than 2 hours, rather than the actual incidence of postoperative ischemia, is the factor most significantly associated with postoperative cardiac outcome after vascular surgery. They hypothesized that because postoperative cardiac complications are preceded by long-duration ST-segment depression, it is likely that the events leading to postoperative cardiac complications do not begin with acute coronary occlusion but with prolonged subendocardial ischemia. Efforts to prevent myocardial ischemia in these patients have included the use of [3-adrenergic blocking drugs given in the perioperative period. In a historical, case-matching study, Pastemack et ais studied the effects of a single dose of metoprolol given preoperatively to patients undergoing peripheral vascular surgery. Compared with historic controls, the patients given metoprolol had significantly less intraoperative myocardial ischemia incidence and duration. Stone et al6 concluded that mild hypertension, when untreated before the induction of anesthesia, resulted in myocardial ischemia in 28% of patients during endotracheal intubation. A single small oral dose of labetalol given preoperatively significantly reduced that incidence to less than 3%. Evidence suggests that the use of [3-adrenergic receptor blockers decreases cardiac death during the long-term postoperative period in patients at risk for coronary artery disease (CAD). 7,s It remains uncertain at this time whether [3-adrenergic receptor blockade reduces perioperative ischemia and, if so, whether it would reduce the incidence of myocardial ischemia and cardiac death in the surgical patient. Journal of Cardiothoracic and Vascular Anesthesia,
Copyright© 1999 by W.B. Saunders Company KEY WORDS: abdominal aortic surgery, ~-adrenergic blockers, myocardial ischemia, perioperative complications
MATERIALS AND METHODS This study was reviewed and approved by the Clinical Research Practices Committee of the Bowman Gray School of Medicine/North Carolina Baptist Hospital; written informed consent was obtained from every patient. Forty patients were enrolled at the North Carolina Baptist Hospital who were scheduled for elective abdominal aortic surgery or renal artery surgery. Only patients undergoing surgical procedures in which the abdominal aorta would be completely cross-clamped were enrolled. Patients with electrocardiogram (ECG) abnormalities that interfered with ST-segment analysis (eg, left bundle-branch block) and those with contraindications to ~3-adrenergic receptor blockade (a clinical history of bronchospastic pulmonary disease requiring treatment) were excluded from the study. To ensure adequate randomization, the patients were first stratified into those already taking [3-adrenergicreceptor blockers and those who
From the Departments of Anesthesiology and General Surgery, Wake Forest University School of Medicine, Winston-Salem, NC. Address reprint requests to Timothy N. Harwood, MD, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1009. Copyright © 1999 by W.B. Saunders Company 1053-0770/99/1305-0008510.00/0
Vo113,No 5 (October),1999:pp 555-561
555
556
HARWOOD ET AL
were not. Each patient was randomly assigned to have HR controlled by one of two treatment strategies: (1) controlled to less than 80 beats/rain (group 80) and (2) controlled to less than 110 beats/min (group 110), using esmolol as needed in either group. A strict placebo control was not used because the study was testing for safety and feasibility, and clinical practice was to avoid excessive degrees of tachycardia in the perioperative period. Each patient's HR was monitored continuously in the operating room and intensive care unit (ICU). If a subject's HR surpassed the target HR for more than 1 minute, the protocol called for initiating an esmolol infusion at 50 pg/kg/min and increasing by 25 gg/kg/min every 5 minutes until the HR decreased to the target. Likewise, if the subject's HR dropped below the target, the esmolol infusion rate was reduced by 25 ~tg/kg/min every 5 minutes until the HR rose to the target or the esmolol infusion was 0 (Fig 1). HR control protocol began at induction of anesthesia and was continued for 48 hours thereafter. Anesthetic technique consisted of an induction with thiopental, 3 to 5 mg/kg, fentanyl, 5 gg/kg, and pancuronium, 0.1 mg/kg. Maintenance of anesthesia consisted of isoflurane, fentanyl, and pancuronium. Pain control in both groups was achieved with patientcontrolled intravenous morphine or epidural morphine as selected by the surgeon and anesthesiologist. For detection of myocardial ischemia, patients were monitored continuously with a Monitor One TC Ambulatory Cardiac Monitor (qMed Inc, Clark, NJ) starting the day before surgery and continuing during the first 3 postoperative days using modified leads HI and V5. Analysis of the ECG was performed with the used of ECGraph software (qMed Inc., Clark, N J). The authors defined greater than 1 mm depression of the ST segment at the J-point plus 60 ms as diagnostic of significant myocardial ischemia. The analysis period was split into two groups. The time from placement of the monitor until anesthetic induction was termed the preoperative period. The period from induction to the end of all recording was termed the postinduction period. In expressing the HR for a specific time period, the HR recorded on the cardiac monitor at the time closest to the specified time interval (the number of hours after induction of anesthesia) was used. The authors expected periods when HR control per protocol could not be maintained and analyzed the amount of ischemia occurring both with protocol HR and when the HR increased above protocol values. To
diagnose myocardial infarction, cardiac enzymes (serum CK-MB) and 12-lead ECGs were obtained daily for the first 5 postoperative days. A CK-MB fraction of greater than 3% with an elevated total ereatine kinase (>280 IU/L) was defined as indicative of myocardial infarction. Likewise, a cardiologist (R.L.R.) interpreted all the ECGs to diagnose myocardial infarction (development of Q waves in II, HI, or aVF or in two of any leads in V1 through V6 using the criteria of Q waves > 1 mm in depth and >0.04 s in duration). Complications referable to the esmolol infusion were recorded as per institutional review board guidelines. Specifically, hypotension, bronchospasm, and mental status changes were looked for during daily patient examination and review of the patient records. Protocol orders specified that the nurse caring for the patient contact one of the study investigators if problems with the protocol or esmolol infusion arose. Preoperative subject characteristics and hemodynamics were analyzed using Student's t-test, Chi-square analysis, or Fisher's exact test as appropriate. Duration and average doses of esmolol were compared by Wilcoxon rank sum. Durations of ischemia and monitoring times in the preoperative and postinduction periods were calculated for each subject. HR groups were compared for differences in the incidence of ischemia by McNemar's test. Changes in individual occurrence and duration of ischemia were analyzed with Chi-square testing. Power analysis was performed to estimate the number of subjects needed to achieve statistical significance at a power of 80% and a significance level of 0.05. Influences of time or HR group on HRs measured at specific intervals during the study were measured with repeated measures analysis of variance. All testing was performed using SAS software (SAS Institute, Cary, NC). RESULTS
Patient Characteristics The demographics o f age, sex, severity o f CAD, use o f preoperative [3-adrenergic receptor blockers, surgical procedure, and preincision h e m o d y n a m i c s are listed in Table 1. Table 2 lists scoring parameters for severity o f CAD. N o significant differences existed b e t w e e n the groups in any category. Mean
Randomization
Group 80 or 110
I
I
HR < target
HR > target
[ Noesmolol I
Esmolol 50 pg/kg/min
I
I HR > target for 5"
HR < target
increase esmolol by 25 pg/kg/min (maximum 400 pg/kg/min)
esmolol off
l
I
HR < target
I
I~ t
HR > target for 5"
/ Fig 1. Algorithm for the use of esmolol beginning at induction of anesthesia to end of study period,
PERIOPERATIVE USE OF ESMOLOL IN AORTIC SURGERY
557
Table 1. Characteristics and Baseline Hemodynamics of Subjects HR 80 (n = 18)
Characteristic
HR 110 (n = 22)
Age (y) 64 _+ 9 66 _+ 8 Male(n) 14 15 Preoperative #-blocker therapy (n) 4 8 Coronary artery disease score (n)* 0 12 15 1 1 4 2 2 1 3 3 2 Procedures (n) AAA 10 11 RAB 5 4 ABF 2 6 AAA/RAB 1 1 Epidural analgesia 6 8 Baseline hemodynamicst HR (beats/min) 69 (3) 68 (3) MAP (mmHg) 85 (4) 92 (5) CI (cardiac output/m 2) 2.7 (0.2) 2.7 (0.2) PAOP (mmHg) 11 (1) 12(1)
Significance (p < 0.05)
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NOTE. Values are _+ standard deviation. Abbreviations: HR, heart rate; NS, not significant; AAA, abdominal aneurysmectomy; RAB, renal artery bypass; ABF, aortobifemoral bypass graft; MAP, mean arterial pressure; CI, cardiac index; PAOP, pulmonary artery occlusion pressure. *Coronary artery disease is graded from none apparent (0) to severe (3). See Table 2. tAll hemodynamic values were measured after induction of anesthesia and before incision.
systemic arterial blood pressure, cardiac indices, and pulmonary capillary wedge pressure were not significantly different between groups at the baseline period.
infusion rates are provided in Table 4 and differ significantly between the two groups (p = 0.0003). The median averaged doses, infusion rates, and duration of infusions were significantly lower in group 110. Heart Rate Control On average, chosen HR goals were achieved in the intraoperative period at a 90% success rate. Target HRs during the intraoperative period were achieved with greater success in group 110 (95% to 100% of recorded intervals) as opposed to group 80 (75% to 90% of recorded intervals). In group 80 during the postoperative period, however, these goals were less well achieved. Initial success in achieving target HRs diminished from 50% to 10% of recorded intervals over the postoperative course. In group 110, targets were achieved at an 87% to 100% success rate. As expected, a greater percentage of subjects' HRs increased above target values as the esmolol was withdrawn (Fig 3). In Fig 4, the deviation from the target HRs in group 80 is more closely examined. During postoperative days 2 and 3, the median HRs gradually increased above 80 beats/min despite continued use of esmolol. Myocardial Ischemia Incidence Slightly more patients in group 80 than in group 110 exhibited preoperative myocardial ischemia (29% v 14%, p = 0.27). Postoperatively, the incidence of ischemic events was similar in both groups (50% v 55%, p = 0.77). In terms of the association of preoperative and postoperative ischemia, a significant relationship existed in both HR groups (Table 5). In both groups, all patients with preoperative ischemia had ischemia in the postinduction period. A disproportionate amount of perioperative ischemia (98%) occurred during the postoperative period. This result held true for both HR groups.
Heart Rate Significant differences between the groups in terms of HRs were identified during the period from 3 hours to 48 hours postoperatively. Before and after that time, group HRs were similar (Fig 2, Table 3). Later in the protocol period, as the esmolol was withdrawn, ranges of HRs increased. Esmolol Requirements As expected, nearly all (94%) the patients in group 80 required esmolol infusion during the course of the study. Only 50% of the patients in group 110 required esmolol at any time. Values indicating durations of treatment, averaged doses, and
Changes in Ischemia Duration For group 80, the median (range) of time-averaged (minutes of ischemia/hour) ischemia in the preoperative and postinduction periods were 0 (0 to 35) and 0 (0 to 42). For group 110, these values were 0 (0 to 55) and 0.07 (0 to 43). All values reflect the low incidence and duration of iscbemic episodes. The study analyzed the change in the time-averaged and patient-averaged duration of ischemia (minutes of ischemia/hour/ patient) from the preoperative to postoperative periods. This average duration of myocardial ischemia decreased slightly in group 80 (from 6.4 to 4.1) and increased slightly in group 110
Table 2. Scoring of Coronary Artery Disease Score
Stenosis
Radiocontrast Ventriculography
Echocardiography
Nuclear (MUGA)
0 1 2 3
None <50% in any coronary artery 50-75% in 1-2 vessels >50% in 3 vessels
Normokinesis Mild hypokinesis Moderate hypokinesis Severe hypokinesis or presence of akinesis
Normokinesis Mild hypokinesis Moderate hypokinesis Severe hypokinesis or presence of akinesis
LVEF >60% LVEF = 45-60% LVEF = 30%-45% LVEF <30%
NOTE. Subjects were graded according to the results of cardiac testing. Severity of coronary artery disease or cardiomyopathy was graded using coronary artery stenosis and left ventricular dysfunction. Left ventricular dysfunction was judged by cardiac catheterization, echocardiogram, or gated pool nuclear scanning. All patients had one of the tests based on surgeon or cardiologist preference. Abbreviations: MUGA, multigated angiogram; LVEF, left ventricular ejection fraction.
558
HARWOOD ET AL
105 100 95 A
E
.z
90
Fig 2. Mean (+- standard error) of heart rate of subjects at specific time periods during study. &, group 80; II, group 110. Significant differences in HR between groups occurred from admission to the postanesthesia care unit/intensive care unit to hour 18 on postoperative day 2 (p < 0.05) as measured by repeated measures analysis of variance. Refer to Table 3 for explanation of time periods.
85 e,, ,1-
65
,!i"
. . . . . . . . . . . . . . . . . . . -, 0 1 2 3 4 5 6 7 8 91011121314151617181920212223242526
Study Periods
(2.5 to 3. l). Statistical analysis revealed no difference between time periods (preinduction versus postinduction) or between groups.
Complications of Therapy Withdrawal of esmolot led to an increase in HR in 78% of all patients receiving esmolol. One patient received the protocol's maximal dose for approximately 18 hours (400 pg/kg/min) and was rendered to a semicomatose state for several hours. This condition was thought to be due to accumulation of methanol, a Table 3. 13me Periods During Study* ]]me Period
Description
1 2 3 4 5 6 7 8 9 10 11
Arrival in operating room Immediately preinduction Immediately postinduction Incision Abdomen open Preaortic clamp(ng 5 minutes postclamping 5 minutes postunclamping Abdomen closed Leave operating room Admission to postanesthesia care unit/intensive care unit 3 hours later 6 hours later Postoperative day 1/hour 0 Postoperative day 1/hour 6 Postoperative day 1/hour 12 Postoperative day 1/hour 18 Postoperative day 2/hour 0 Postoperative day 2/hour 6 Postoperative day 2/hour 12 Postoperative day 2/hour 18 Postoperative day 3/hour 0 Postoperative day 3/hour 6 Postoperative day 3/hour 12 Postoperative day 3/hour 18 Postoperative day 4/hour 0
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 *See Figure 2.
metabolite of esmolol. Esmolol was subsequently tapered only after several oral doses of atenolol had been given. No patient developed wheezing or other signs of bronchospasm thought to be a result of the esmolol. One patient developed hypotension because of internal hemorrhage. The esmolol infusion in this patient was discontinued with stabilization of vital signs before surgical reexploration. No other patients experienced hypotensive episodes. DISCUSSION
This study confirms that many patients undergoing abdominal aortic surgical procedures have myocardial ischemia perioperatively, corroborating multiple previous investigations.1-3,9 Furthermore, this study demonstrates that aggressive use of esmolol appears safe in patients undergoing major vascular surgery. Although patients in the lower HR group tended to have fewer and shorter episodes of ischemia postoperatively, the study was not powered to show statistical significance. This study also indicates that esmolol is effective in lowering the HR of patients during and after major vascular surgical procedures without producing significant side effects at the doses used in most patients.
Postoperative Myocardial Ischemia Findings reflect other observations concerning the frequency and other characteristics of myocardial ischemia in patients Table 4. Duration of Esmolol Infusions and Total Doses of Esmolol According to Group
Group
HR 80 HR 110
No. Hours on TreatmentJ Media n (Range)
Average Dose*/ Median (Range)
Infusion Rate/ Median (Range)
42 (0-51) 85.35(0-238) IJg/kg/min 100 (0-400) gg/kg/min 9 (0-53) 11.45(0-123.8) !Jg/kg/min 12.5(0-200) IJg/kg/min
Abbreviation: HR, heart rate. *Average dose is calculated by the following formula: Total esmolol given during study (l~g) Body weight (kg) • Total study duration (72 h)
PERIOPERATIVEUSE OF ESMOLOL IN AORTIC SURGERY
559
100% 90%
80%
•
70%
I
60%
I
50%
•
40%
|
30%
D.
20%
~ i
Ill,
,o% 0% ,I I
Fig 3. Percent of time the subjects' heart rates remained at or below the target over the duration of the study. II, group 80; D , group 110.
• ii
Li
1 2 3456
0
,i
°
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Study Periods
undergoing abdominal aortic surgical procedures in the perioperative period. Although myocardial ischemia occurs in about half of these patients in the postoperative periods, most of the patients with ST-segment depression have it for short periods of time. Although it appears from previous studies that these patients are at higher risk for postoperative myocardial infarction than those without ischemia, only a small minority of the patients go on to develop myocardial infarction. The patients at risk in this study cannot be determined because of the small number of subjects and the lack of subjects who had a myocardial infarction.
[3-Adrenergic Blockade and Myocardial Ischernia Previous clinical studies show that 13-adrenergic receptor blockers given after myocardial infarction reduce the reinfarcdon rate. i0-14In addition, these drugs have been shown to reduce the size of experimental and naturally occurring infarcts. I5,16 Lange et all6 investigated the effects of esmolol on infarct size and segmental wall dynamics of reperfused myocardial infarcts in anesthetized dogs and showed reduced ischemic damage with the use of esmolol before injury. Neustein et a117 investigated the use of esmolol during myocardial revascularization in the prebypass period. No significant differences were observed in
ischemia. HRs were low in both groups. Clinical studies focusing on the use of [3-adrenergic receptor blockers in the perioperative period have been equivocal in their findings.18,19 This study confirms a low frequency of intraoperafive ischemic episodes. This low frequency could be a result of the consistently lower HRs observed during the anesthetic period. The phenomenon of lower HRs may occur because of the reduced stress response encountered during the anesthetic state or the tighter compliance with study protocol that occurred during the operative period. The increase in ischemia in the postoperative period may be influenced by the usual postoperative increase in coagulation that can encourage the formation of coronary thrombosis. Despite the causative factors, the data indicate that strategies aimed at reducing the incidence of ischemia should focus on the postoperative period, rather than the time during Which the patient is undergoing surgery.
Therapeutic Efficacy and Safety of Esmolol
E= 10
Esmolol has a documented history of providing efficacy (lowering HR) with minimal side effects. Known unwanted effects include congestive heart failure, hypotension, bradyarrhythmias, bronchoconstriction, and central nervous system depression. In the present study, esmolol proved to be a safe drug. Only one episode of hypotension (systolic blood pressure <90 mmHg) occurred. Likewise, probably because of the exclusion of subjects with active bronchospastic disease, episodes of clinically evident perioperative bronchospasm were absent.
E .=o 0
Table 5. Relationship of the Presence of Preinduction to Postinduction Ischemia in Group 110
15
& g
No Postinduction Ischemia
'~ Q -10 -15
!
I
I
I
I
I
I
I
!
:
I
:
I
~
~
I
:
I
I
I
I
:
I
I
2 3 4 5 6 7 8 91011 121314151617181920212223242526
Study Periods
Fig 4. Deviation of subjects" heart rates from 80 beats per minute over the duration of the study for group 80.
Group 80 (p = 0.003) No Yes Group 110 No Yes
Yes
Preinduction Ischemia
10 9 9 4
0.00 3, 0 5
560
HARWOOD ET AL
Limitations of Study Esmolol was chosen to achieve tight HR control, in part, because esmolol, in contrast to any other 13-adrenergic receptor blocker, is titratable and allows for a rapid change in drug effect. This characteristic is valuable because of the potential for rapid variability in hemodynamics in patients undergoing aortic surgery. Also, esmolol has relative 131 specificity and provides [3-adrenergic receptor blockade rapidly, z° Partly because of its short half-life, target HR was achieved in the low HR group. The data suggest that control of HR perioperatively with esmolol is feasible, but perfect control may not be accomplished in the ICU setting. By giving minute-to-minute control of the HR to a nursing staff that has many other duties to perform, the authors may have allowed for too much variability in the subjects' HRs. The incidence of myocardial ischemia in the preoperative period was not as high as reported elsewhere for asymptomatic individuals. 2~ The period of preoperative ambulatory ECG monitoring may not have been long enough to fully reflect the true incidence of myocardial ischemia. Also, measurements made in the hospital are done when the patient has a generally low activity level. Because of this, the difference in ischemia between the groups preoperatively may have been overestimated. Also, because of the low incidence of myocardial ischemia and death, an effect of HR treatment strategy on overall outcome could not be tested for. The lack of finding a statistically significant altered incidence or pattern of ischemia in the low HR group may be the result of three factors: (1) no difference in ischemia between the groups, (2) a nonnormal distribution of ischemia patterns requiring the use of less powerful nonparametric statistical testing, and (3) lack of enough subjects to find a difference in ischemia incidence between time periods. It is recognized that interpretation of these results is compli-
cared by the skewed distribution of patients with myocardial ischemia. Others have estimated that because of normal variation in daily myocardial ischemia as detected by ambulatory ECG monitoring, a true reduction of 15% or greater in ischemic episodes is required to have sufficient power to obtain statistical significance in a clinical study, z2 Further study with a larger population of subjects, particularly of patients with preoperative ischemia, is necessary to determine whether this HR treatment strategy would decrease cardiac morbidity or overall mortality. Moreover, any treatment-related improvement in outcome needs to be large enough to justify the direct and indirect costs and potential risks of strict HR control with esmolol. Although the number of patients with complications related to esmolol is low in this study, a larger number of subjects may reveal more complications, such as hypotension and bronchospasm. With a total number of subjects such as in the present study, combined with an expectedly low incidence of complications, this study was not powered to detect a significant difference in complication rates between groups. Tight perioperative control of HR is a possible alternative to the widespread use of preoperative myocardial ischemia testing. Testing all patients who are to undergo major vascular surgery is costly and has disadvantages. Moreover, testing has not been shown to save money or lives. Revascularization improves outcome for patients with defined indications (eg, unstable angina, left main CAD). Evidence that otherwise asymptomatic patients who demonstrate CAD on preoperative screening would benefit from percutaneous transluminal angioplasty or coronary artery bypass graft is not available. Further study should test whether pharmacologically protecting patients with 13-adrenergic blockers in the perioperative period would be the more cost-effective manner to prevent the major causes of morbidity and mortality in this patient population--myocardial ischemia and infarction.
REFERENCES
1. Mangano DT: Perioperative cardiac morbidity. Anesthesiology 72:153-184, 1990 2. Raby KE, Barry J, Creager MA, et al: Detection and significance of intraoperative and postoperative myocardial ischemia in peripheral vascular surgery. JAMA 268:222-227, 1992 3. Raby KE, Goldman L, Creager MA, et al: Correlation between preoperative ischemia and major cardiac events after peripheral vascular surgery. N Engl J Med 321:1296-1300, 1989 4. Landesberg G, Luria MH, Cotev S, et al: Importance of longduration postoperative ST-segment depression in cardiac morbidity after vascular surgery. Lancet 341:715-719, 1993 5. Pasternack PF, Grossi EA, Baumann FG, et al: Beta-blockade to decrease silent myocardial ischemia during peripheral vascular surgery. Am J Surg 158:113-116, 1989 6. Stone JG, Foex P, Sear JW, et al: Myocardial ischemia in untreated hypertensive patients: Effect of a single small oral dose of a betaadrenergic blocking agent. Anesthesiology 68:495-500, 1988 7. Wallace A, Layug B, Tateo I, et al: Prophylactic atenolol reduces postoperative myocardial ischemia. Anesthesiology 88:7-17, 1998 8. Mangano DT, Layug EL, Wallace A, Tateo I: Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 335:1713-1726, 1996 9. Mangano DT, Wong MG, London MJ, et al: Perioperative myocardial ischemia in patients undergoing noncardiac surgery: II. Incidence and severity during the 1st week after surgery. The Study of
Perioperative Ischemia (SPI) Research Group. J Am Coll Cardiol 17:851-857, 1991 10. Kjekshus JK: Importance of heart rate in determining betablocker efficacy in acute and long-term acute myocardial infarction intervention trials. Am J Cardio157:43F-49F, 1986 11. Randomised trial of intravenous atenolol among 16,027 cases of suspected acute myocardial infarction: ISIS-1. First International Study of Infarct Survival Collaborative Group. Lancet 2:57-66, 1986 12. Gundersen T, Gottum R Pedersen T, et al: Effect of timolol on mortality and reinfarction after acute myocardial infarction: Prognostic importance of heart rate at rest. Am J Cardiol 58:20-24, 1986 13. Autman EM, Lau J, Kupelnick B, et al: A comparison of results of meta-analyses of randomized control trials and recommendations of clinical experts: Treatments for myocardial infarction. JAMA 268:240248, 1992 14. Lau J, Antman EM, Jimenez-Silva J, et al: Cumulative metaanalysis of therapeutic trials for myocardial infarction. N Engl J Med 327:248-254, 1992 15. Mulcahy D, Keegan J, Cunningham D, et al: Circadian variation of total ischaemic burden and its alteration with anti-anginal agents. Lancet 2:755-759, 1988 16. Lange R, Kloner RA, Braunwald E: First ultra-short-acting beta-adrenergic blocking agent: Its effect on size and segmental wall dynamics of reperfused myocardial infarcts in dogs. Am J Cardiol 51:1759-1767, 1983
PERIOPERATIVE USE OF ESMOLOL IN AORTIC SURGERY
17. Neustein SM, Bronheim DS, Lasker S, et al: Esmolol and intraoperative myocardial ischemia: A double-blind study. J Cardiothorac Vasc Anesth 8:273-277, 1994 18. Slogoff S, Keats AS, Ott E: Preoperative propranolol therapy and aortocoronary bypass operation. JAMA 240:1487-1490, 1978 19. Magnusson J, Thulin T, Werner O, et al: Haemodynamic effects of pretreatment with metoprolol in hypertensive patients undergoing surgery. Br J Anaesth 58:251-260, 1986 20. Kirshenbaum JM, Kloner RA, Antman EM, et al: Use of an ultra
561
short-acting beta-blocker in patients with acute myocardial ischemia. Circulation 72:873-880, 1985 21. Coy KM, Imperi GA, Lambert CR, et al: Silent myocardial ischemia during daily activities in asymptomatic men with positive exercise test responses. Am J Cardio159:45-49, 1987 22. Celermajer DS, Spiegelhalter DJ, Deanfield M, et al: Variability of episodic ST segment depression in chronic stable angina: Implications for individual and group trials of therapeutic efficacy. J Am Coll Cardiol 23:66-73, 1994
strated an adverse effect. The median infusion rates were 100 and 12.5 pg/kg/min for groups 80 and 110. Target HR was met less in group 80 than in group 110, primarily in the postoperative period. Ischemia patterns were not significantly different between groups. Conclusion: Using esmolol for HR control in the intraoperarive period for abdominal vascular surgery patients is effective and safe. HR control was much less effective in the postoperative period, but esmolol is safe when used at recommended doses. Further study with a larger number of patients is necessary to determine whether strict HR control with esmolol affects the incidence of myocardial ischemia or infarction in this patient population.
p
Based on these studies, the authors designed a protocol that maintains a normal (or lower) heart rate (HR) with a [3-adrenergic receptor blocker to reduce the risk of perioperative myocardial ischemia and infarction. A short-acting agent was used so that dosage adjustments could be accomplished to match changes in autonomic activation and stress that occur during and after major surgery. The feasibility and safety of such prolonged infusion of esmolol in patients with comorbid illnesses seen in vascular surgery patients have not been established. The present research was designed to determine (1) whether strict HR control can be maintained perioperatively using esmolol, (2) what doses of esmolol are required to accomplish HR control, and (3) whether there might be any influence on myocardial ischemia or adverse consequences of this approach.
ATIENTS UNDERGOING major abdominal vascular sur-
gery are at higher risk for perioperative myocardial infarction and cardiac mortality than patients undergoing most other types of surgery. 1 These risks result from the greater prevalence of coronary atherosclerosis in this patient population, coupled with the superimposed metabolic and thrombotic changes of major surgery. Nevertheless, most patients undergo vascular surgery without adverse myocardial events. A subset of the population with perioperative myocardial ischemia appears to be at greater risk of adverse cardiac outcome, z3 Landesberg et ai4 determined that postoperative ischemia lasting longer than 2 hours, rather than the actual incidence of postoperative ischemia, is the factor most significantly associated with postoperative cardiac outcome after vascular surgery. They hypothesized that because postoperative cardiac complications are preceded by long-duration ST-segment depression, it is likely that the events leading to postoperative cardiac complications do not begin with acute coronary occlusion but with prolonged subendocardial ischemia. Efforts to prevent myocardial ischemia in these patients have included the use of [3-adrenergic blocking drugs given in the perioperative period. In a historical, case-matching study, Pastemack et ais studied the effects of a single dose of metoprolol given preoperatively to patients undergoing peripheral vascular surgery. Compared with historic controls, the patients given metoprolol had significantly less intraoperative myocardial ischemia incidence and duration. Stone et al6 concluded that mild hypertension, when untreated before the induction of anesthesia, resulted in myocardial ischemia in 28% of patients during endotracheal intubation. A single small oral dose of labetalol given preoperatively significantly reduced that incidence to less than 3%. Evidence suggests that the use of [3-adrenergic receptor blockers decreases cardiac death during the long-term postoperative period in patients at risk for coronary artery disease (CAD). 7,s It remains uncertain at this time whether [3-adrenergic receptor blockade reduces perioperative ischemia and, if so, whether it would reduce the incidence of myocardial ischemia and cardiac death in the surgical patient. Journal of Cardiothoracic and Vascular Anesthesia,
Copyright© 1999 by W.B. Saunders Company KEY WORDS: abdominal aortic surgery, ~-adrenergic blockers, myocardial ischemia, perioperative complications
MATERIALS AND METHODS This study was reviewed and approved by the Clinical Research Practices Committee of the Bowman Gray School of Medicine/North Carolina Baptist Hospital; written informed consent was obtained from every patient. Forty patients were enrolled at the North Carolina Baptist Hospital who were scheduled for elective abdominal aortic surgery or renal artery surgery. Only patients undergoing surgical procedures in which the abdominal aorta would be completely cross-clamped were enrolled. Patients with electrocardiogram (ECG) abnormalities that interfered with ST-segment analysis (eg, left bundle-branch block) and those with contraindications to ~3-adrenergic receptor blockade (a clinical history of bronchospastic pulmonary disease requiring treatment) were excluded from the study. To ensure adequate randomization, the patients were first stratified into those already taking [3-adrenergicreceptor blockers and those who
From the Departments of Anesthesiology and General Surgery, Wake Forest University School of Medicine, Winston-Salem, NC. Address reprint requests to Timothy N. Harwood, MD, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1009. Copyright © 1999 by W.B. Saunders Company 1053-0770/99/1305-0008510.00/0
Vo113,No 5 (October),1999:pp 555-561
555
556
HARWOOD ET AL
were not. Each patient was randomly assigned to have HR controlled by one of two treatment strategies: (1) controlled to less than 80 beats/rain (group 80) and (2) controlled to less than 110 beats/min (group 110), using esmolol as needed in either group. A strict placebo control was not used because the study was testing for safety and feasibility, and clinical practice was to avoid excessive degrees of tachycardia in the perioperative period. Each patient's HR was monitored continuously in the operating room and intensive care unit (ICU). If a subject's HR surpassed the target HR for more than 1 minute, the protocol called for initiating an esmolol infusion at 50 pg/kg/min and increasing by 25 gg/kg/min every 5 minutes until the HR decreased to the target. Likewise, if the subject's HR dropped below the target, the esmolol infusion rate was reduced by 25 ~tg/kg/min every 5 minutes until the HR rose to the target or the esmolol infusion was 0 (Fig 1). HR control protocol began at induction of anesthesia and was continued for 48 hours thereafter. Anesthetic technique consisted of an induction with thiopental, 3 to 5 mg/kg, fentanyl, 5 gg/kg, and pancuronium, 0.1 mg/kg. Maintenance of anesthesia consisted of isoflurane, fentanyl, and pancuronium. Pain control in both groups was achieved with patientcontrolled intravenous morphine or epidural morphine as selected by the surgeon and anesthesiologist. For detection of myocardial ischemia, patients were monitored continuously with a Monitor One TC Ambulatory Cardiac Monitor (qMed Inc, Clark, NJ) starting the day before surgery and continuing during the first 3 postoperative days using modified leads HI and V5. Analysis of the ECG was performed with the used of ECGraph software (qMed Inc., Clark, N J). The authors defined greater than 1 mm depression of the ST segment at the J-point plus 60 ms as diagnostic of significant myocardial ischemia. The analysis period was split into two groups. The time from placement of the monitor until anesthetic induction was termed the preoperative period. The period from induction to the end of all recording was termed the postinduction period. In expressing the HR for a specific time period, the HR recorded on the cardiac monitor at the time closest to the specified time interval (the number of hours after induction of anesthesia) was used. The authors expected periods when HR control per protocol could not be maintained and analyzed the amount of ischemia occurring both with protocol HR and when the HR increased above protocol values. To
diagnose myocardial infarction, cardiac enzymes (serum CK-MB) and 12-lead ECGs were obtained daily for the first 5 postoperative days. A CK-MB fraction of greater than 3% with an elevated total ereatine kinase (>280 IU/L) was defined as indicative of myocardial infarction. Likewise, a cardiologist (R.L.R.) interpreted all the ECGs to diagnose myocardial infarction (development of Q waves in II, HI, or aVF or in two of any leads in V1 through V6 using the criteria of Q waves > 1 mm in depth and >0.04 s in duration). Complications referable to the esmolol infusion were recorded as per institutional review board guidelines. Specifically, hypotension, bronchospasm, and mental status changes were looked for during daily patient examination and review of the patient records. Protocol orders specified that the nurse caring for the patient contact one of the study investigators if problems with the protocol or esmolol infusion arose. Preoperative subject characteristics and hemodynamics were analyzed using Student's t-test, Chi-square analysis, or Fisher's exact test as appropriate. Duration and average doses of esmolol were compared by Wilcoxon rank sum. Durations of ischemia and monitoring times in the preoperative and postinduction periods were calculated for each subject. HR groups were compared for differences in the incidence of ischemia by McNemar's test. Changes in individual occurrence and duration of ischemia were analyzed with Chi-square testing. Power analysis was performed to estimate the number of subjects needed to achieve statistical significance at a power of 80% and a significance level of 0.05. Influences of time or HR group on HRs measured at specific intervals during the study were measured with repeated measures analysis of variance. All testing was performed using SAS software (SAS Institute, Cary, NC). RESULTS
Patient Characteristics The demographics o f age, sex, severity o f CAD, use o f preoperative [3-adrenergic receptor blockers, surgical procedure, and preincision h e m o d y n a m i c s are listed in Table 1. Table 2 lists scoring parameters for severity o f CAD. N o significant differences existed b e t w e e n the groups in any category. Mean
Randomization
Group 80 or 110
I
I
HR < target
HR > target
[ Noesmolol I
Esmolol 50 pg/kg/min
I
I HR > target for 5"
HR < target
increase esmolol by 25 pg/kg/min (maximum 400 pg/kg/min)
esmolol off
l
I
HR < target
I
I~ t
HR > target for 5"
/ Fig 1. Algorithm for the use of esmolol beginning at induction of anesthesia to end of study period,
PERIOPERATIVE USE OF ESMOLOL IN AORTIC SURGERY
557
Table 1. Characteristics and Baseline Hemodynamics of Subjects HR 80 (n = 18)
Characteristic
HR 110 (n = 22)
Age (y) 64 _+ 9 66 _+ 8 Male(n) 14 15 Preoperative #-blocker therapy (n) 4 8 Coronary artery disease score (n)* 0 12 15 1 1 4 2 2 1 3 3 2 Procedures (n) AAA 10 11 RAB 5 4 ABF 2 6 AAA/RAB 1 1 Epidural analgesia 6 8 Baseline hemodynamicst HR (beats/min) 69 (3) 68 (3) MAP (mmHg) 85 (4) 92 (5) CI (cardiac output/m 2) 2.7 (0.2) 2.7 (0.2) PAOP (mmHg) 11 (1) 12(1)
Significance (p < 0.05)
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
NOTE. Values are _+ standard deviation. Abbreviations: HR, heart rate; NS, not significant; AAA, abdominal aneurysmectomy; RAB, renal artery bypass; ABF, aortobifemoral bypass graft; MAP, mean arterial pressure; CI, cardiac index; PAOP, pulmonary artery occlusion pressure. *Coronary artery disease is graded from none apparent (0) to severe (3). See Table 2. tAll hemodynamic values were measured after induction of anesthesia and before incision.
systemic arterial blood pressure, cardiac indices, and pulmonary capillary wedge pressure were not significantly different between groups at the baseline period.
infusion rates are provided in Table 4 and differ significantly between the two groups (p = 0.0003). The median averaged doses, infusion rates, and duration of infusions were significantly lower in group 110. Heart Rate Control On average, chosen HR goals were achieved in the intraoperative period at a 90% success rate. Target HRs during the intraoperative period were achieved with greater success in group 110 (95% to 100% of recorded intervals) as opposed to group 80 (75% to 90% of recorded intervals). In group 80 during the postoperative period, however, these goals were less well achieved. Initial success in achieving target HRs diminished from 50% to 10% of recorded intervals over the postoperative course. In group 110, targets were achieved at an 87% to 100% success rate. As expected, a greater percentage of subjects' HRs increased above target values as the esmolol was withdrawn (Fig 3). In Fig 4, the deviation from the target HRs in group 80 is more closely examined. During postoperative days 2 and 3, the median HRs gradually increased above 80 beats/min despite continued use of esmolol. Myocardial Ischemia Incidence Slightly more patients in group 80 than in group 110 exhibited preoperative myocardial ischemia (29% v 14%, p = 0.27). Postoperatively, the incidence of ischemic events was similar in both groups (50% v 55%, p = 0.77). In terms of the association of preoperative and postoperative ischemia, a significant relationship existed in both HR groups (Table 5). In both groups, all patients with preoperative ischemia had ischemia in the postinduction period. A disproportionate amount of perioperative ischemia (98%) occurred during the postoperative period. This result held true for both HR groups.
Heart Rate Significant differences between the groups in terms of HRs were identified during the period from 3 hours to 48 hours postoperatively. Before and after that time, group HRs were similar (Fig 2, Table 3). Later in the protocol period, as the esmolol was withdrawn, ranges of HRs increased. Esmolol Requirements As expected, nearly all (94%) the patients in group 80 required esmolol infusion during the course of the study. Only 50% of the patients in group 110 required esmolol at any time. Values indicating durations of treatment, averaged doses, and
Changes in Ischemia Duration For group 80, the median (range) of time-averaged (minutes of ischemia/hour) ischemia in the preoperative and postinduction periods were 0 (0 to 35) and 0 (0 to 42). For group 110, these values were 0 (0 to 55) and 0.07 (0 to 43). All values reflect the low incidence and duration of iscbemic episodes. The study analyzed the change in the time-averaged and patient-averaged duration of ischemia (minutes of ischemia/hour/ patient) from the preoperative to postoperative periods. This average duration of myocardial ischemia decreased slightly in group 80 (from 6.4 to 4.1) and increased slightly in group 110
Table 2. Scoring of Coronary Artery Disease Score
Stenosis
Radiocontrast Ventriculography
Echocardiography
Nuclear (MUGA)
0 1 2 3
None <50% in any coronary artery 50-75% in 1-2 vessels >50% in 3 vessels
Normokinesis Mild hypokinesis Moderate hypokinesis Severe hypokinesis or presence of akinesis
Normokinesis Mild hypokinesis Moderate hypokinesis Severe hypokinesis or presence of akinesis
LVEF >60% LVEF = 45-60% LVEF = 30%-45% LVEF <30%
NOTE. Subjects were graded according to the results of cardiac testing. Severity of coronary artery disease or cardiomyopathy was graded using coronary artery stenosis and left ventricular dysfunction. Left ventricular dysfunction was judged by cardiac catheterization, echocardiogram, or gated pool nuclear scanning. All patients had one of the tests based on surgeon or cardiologist preference. Abbreviations: MUGA, multigated angiogram; LVEF, left ventricular ejection fraction.
558
HARWOOD ET AL
105 100 95 A
E
.z
90
Fig 2. Mean (+- standard error) of heart rate of subjects at specific time periods during study. &, group 80; II, group 110. Significant differences in HR between groups occurred from admission to the postanesthesia care unit/intensive care unit to hour 18 on postoperative day 2 (p < 0.05) as measured by repeated measures analysis of variance. Refer to Table 3 for explanation of time periods.
85 e,, ,1-
65
,!i"
. . . . . . . . . . . . . . . . . . . -, 0 1 2 3 4 5 6 7 8 91011121314151617181920212223242526
Study Periods
(2.5 to 3. l). Statistical analysis revealed no difference between time periods (preinduction versus postinduction) or between groups.
Complications of Therapy Withdrawal of esmolot led to an increase in HR in 78% of all patients receiving esmolol. One patient received the protocol's maximal dose for approximately 18 hours (400 pg/kg/min) and was rendered to a semicomatose state for several hours. This condition was thought to be due to accumulation of methanol, a Table 3. 13me Periods During Study* ]]me Period
Description
1 2 3 4 5 6 7 8 9 10 11
Arrival in operating room Immediately preinduction Immediately postinduction Incision Abdomen open Preaortic clamp(ng 5 minutes postclamping 5 minutes postunclamping Abdomen closed Leave operating room Admission to postanesthesia care unit/intensive care unit 3 hours later 6 hours later Postoperative day 1/hour 0 Postoperative day 1/hour 6 Postoperative day 1/hour 12 Postoperative day 1/hour 18 Postoperative day 2/hour 0 Postoperative day 2/hour 6 Postoperative day 2/hour 12 Postoperative day 2/hour 18 Postoperative day 3/hour 0 Postoperative day 3/hour 6 Postoperative day 3/hour 12 Postoperative day 3/hour 18 Postoperative day 4/hour 0
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 *See Figure 2.
metabolite of esmolol. Esmolol was subsequently tapered only after several oral doses of atenolol had been given. No patient developed wheezing or other signs of bronchospasm thought to be a result of the esmolol. One patient developed hypotension because of internal hemorrhage. The esmolol infusion in this patient was discontinued with stabilization of vital signs before surgical reexploration. No other patients experienced hypotensive episodes. DISCUSSION
This study confirms that many patients undergoing abdominal aortic surgical procedures have myocardial ischemia perioperatively, corroborating multiple previous investigations.1-3,9 Furthermore, this study demonstrates that aggressive use of esmolol appears safe in patients undergoing major vascular surgery. Although patients in the lower HR group tended to have fewer and shorter episodes of ischemia postoperatively, the study was not powered to show statistical significance. This study also indicates that esmolol is effective in lowering the HR of patients during and after major vascular surgical procedures without producing significant side effects at the doses used in most patients.
Postoperative Myocardial Ischemia Findings reflect other observations concerning the frequency and other characteristics of myocardial ischemia in patients Table 4. Duration of Esmolol Infusions and Total Doses of Esmolol According to Group
Group
HR 80 HR 110
No. Hours on TreatmentJ Media n (Range)
Average Dose*/ Median (Range)
Infusion Rate/ Median (Range)
42 (0-51) 85.35(0-238) IJg/kg/min 100 (0-400) gg/kg/min 9 (0-53) 11.45(0-123.8) !Jg/kg/min 12.5(0-200) IJg/kg/min
Abbreviation: HR, heart rate. *Average dose is calculated by the following formula: Total esmolol given during study (l~g) Body weight (kg) • Total study duration (72 h)
PERIOPERATIVEUSE OF ESMOLOL IN AORTIC SURGERY
559
100% 90%
80%
•
70%
I
60%
I
50%
•
40%
|
30%
D.
20%
~ i
Ill,
,o% 0% ,I I
Fig 3. Percent of time the subjects' heart rates remained at or below the target over the duration of the study. II, group 80; D , group 110.
• ii
Li
1 2 3456
0
,i
°
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Study Periods
undergoing abdominal aortic surgical procedures in the perioperative period. Although myocardial ischemia occurs in about half of these patients in the postoperative periods, most of the patients with ST-segment depression have it for short periods of time. Although it appears from previous studies that these patients are at higher risk for postoperative myocardial infarction than those without ischemia, only a small minority of the patients go on to develop myocardial infarction. The patients at risk in this study cannot be determined because of the small number of subjects and the lack of subjects who had a myocardial infarction.
[3-Adrenergic Blockade and Myocardial Ischernia Previous clinical studies show that 13-adrenergic receptor blockers given after myocardial infarction reduce the reinfarcdon rate. i0-14In addition, these drugs have been shown to reduce the size of experimental and naturally occurring infarcts. I5,16 Lange et all6 investigated the effects of esmolol on infarct size and segmental wall dynamics of reperfused myocardial infarcts in anesthetized dogs and showed reduced ischemic damage with the use of esmolol before injury. Neustein et a117 investigated the use of esmolol during myocardial revascularization in the prebypass period. No significant differences were observed in
ischemia. HRs were low in both groups. Clinical studies focusing on the use of [3-adrenergic receptor blockers in the perioperative period have been equivocal in their findings.18,19 This study confirms a low frequency of intraoperafive ischemic episodes. This low frequency could be a result of the consistently lower HRs observed during the anesthetic period. The phenomenon of lower HRs may occur because of the reduced stress response encountered during the anesthetic state or the tighter compliance with study protocol that occurred during the operative period. The increase in ischemia in the postoperative period may be influenced by the usual postoperative increase in coagulation that can encourage the formation of coronary thrombosis. Despite the causative factors, the data indicate that strategies aimed at reducing the incidence of ischemia should focus on the postoperative period, rather than the time during Which the patient is undergoing surgery.
Therapeutic Efficacy and Safety of Esmolol
E= 10
Esmolol has a documented history of providing efficacy (lowering HR) with minimal side effects. Known unwanted effects include congestive heart failure, hypotension, bradyarrhythmias, bronchoconstriction, and central nervous system depression. In the present study, esmolol proved to be a safe drug. Only one episode of hypotension (systolic blood pressure <90 mmHg) occurred. Likewise, probably because of the exclusion of subjects with active bronchospastic disease, episodes of clinically evident perioperative bronchospasm were absent.
E .=o 0
Table 5. Relationship of the Presence of Preinduction to Postinduction Ischemia in Group 110
15
& g
No Postinduction Ischemia
'~ Q -10 -15
!
I
I
I
I
I
I
I
!
:
I
:
I
~
~
I
:
I
I
I
I
:
I
I
2 3 4 5 6 7 8 91011 121314151617181920212223242526
Study Periods
Fig 4. Deviation of subjects" heart rates from 80 beats per minute over the duration of the study for group 80.
Group 80 (p = 0.003) No Yes Group 110 No Yes
Yes
Preinduction Ischemia
10 9 9 4
0.00 3, 0 5
560
HARWOOD ET AL
Limitations of Study Esmolol was chosen to achieve tight HR control, in part, because esmolol, in contrast to any other 13-adrenergic receptor blocker, is titratable and allows for a rapid change in drug effect. This characteristic is valuable because of the potential for rapid variability in hemodynamics in patients undergoing aortic surgery. Also, esmolol has relative 131 specificity and provides [3-adrenergic receptor blockade rapidly, z° Partly because of its short half-life, target HR was achieved in the low HR group. The data suggest that control of HR perioperatively with esmolol is feasible, but perfect control may not be accomplished in the ICU setting. By giving minute-to-minute control of the HR to a nursing staff that has many other duties to perform, the authors may have allowed for too much variability in the subjects' HRs. The incidence of myocardial ischemia in the preoperative period was not as high as reported elsewhere for asymptomatic individuals. 2~ The period of preoperative ambulatory ECG monitoring may not have been long enough to fully reflect the true incidence of myocardial ischemia. Also, measurements made in the hospital are done when the patient has a generally low activity level. Because of this, the difference in ischemia between the groups preoperatively may have been overestimated. Also, because of the low incidence of myocardial ischemia and death, an effect of HR treatment strategy on overall outcome could not be tested for. The lack of finding a statistically significant altered incidence or pattern of ischemia in the low HR group may be the result of three factors: (1) no difference in ischemia between the groups, (2) a nonnormal distribution of ischemia patterns requiring the use of less powerful nonparametric statistical testing, and (3) lack of enough subjects to find a difference in ischemia incidence between time periods. It is recognized that interpretation of these results is compli-
cared by the skewed distribution of patients with myocardial ischemia. Others have estimated that because of normal variation in daily myocardial ischemia as detected by ambulatory ECG monitoring, a true reduction of 15% or greater in ischemic episodes is required to have sufficient power to obtain statistical significance in a clinical study, z2 Further study with a larger population of subjects, particularly of patients with preoperative ischemia, is necessary to determine whether this HR treatment strategy would decrease cardiac morbidity or overall mortality. Moreover, any treatment-related improvement in outcome needs to be large enough to justify the direct and indirect costs and potential risks of strict HR control with esmolol. Although the number of patients with complications related to esmolol is low in this study, a larger number of subjects may reveal more complications, such as hypotension and bronchospasm. With a total number of subjects such as in the present study, combined with an expectedly low incidence of complications, this study was not powered to detect a significant difference in complication rates between groups. Tight perioperative control of HR is a possible alternative to the widespread use of preoperative myocardial ischemia testing. Testing all patients who are to undergo major vascular surgery is costly and has disadvantages. Moreover, testing has not been shown to save money or lives. Revascularization improves outcome for patients with defined indications (eg, unstable angina, left main CAD). Evidence that otherwise asymptomatic patients who demonstrate CAD on preoperative screening would benefit from percutaneous transluminal angioplasty or coronary artery bypass graft is not available. Further study should test whether pharmacologically protecting patients with 13-adrenergic blockers in the perioperative period would be the more cost-effective manner to prevent the major causes of morbidity and mortality in this patient population--myocardial ischemia and infarction.
REFERENCES
1. Mangano DT: Perioperative cardiac morbidity. Anesthesiology 72:153-184, 1990 2. Raby KE, Barry J, Creager MA, et al: Detection and significance of intraoperative and postoperative myocardial ischemia in peripheral vascular surgery. JAMA 268:222-227, 1992 3. Raby KE, Goldman L, Creager MA, et al: Correlation between preoperative ischemia and major cardiac events after peripheral vascular surgery. N Engl J Med 321:1296-1300, 1989 4. Landesberg G, Luria MH, Cotev S, et al: Importance of longduration postoperative ST-segment depression in cardiac morbidity after vascular surgery. Lancet 341:715-719, 1993 5. Pasternack PF, Grossi EA, Baumann FG, et al: Beta-blockade to decrease silent myocardial ischemia during peripheral vascular surgery. Am J Surg 158:113-116, 1989 6. Stone JG, Foex P, Sear JW, et al: Myocardial ischemia in untreated hypertensive patients: Effect of a single small oral dose of a betaadrenergic blocking agent. Anesthesiology 68:495-500, 1988 7. Wallace A, Layug B, Tateo I, et al: Prophylactic atenolol reduces postoperative myocardial ischemia. Anesthesiology 88:7-17, 1998 8. Mangano DT, Layug EL, Wallace A, Tateo I: Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 335:1713-1726, 1996 9. Mangano DT, Wong MG, London MJ, et al: Perioperative myocardial ischemia in patients undergoing noncardiac surgery: II. Incidence and severity during the 1st week after surgery. The Study of
Perioperative Ischemia (SPI) Research Group. J Am Coll Cardiol 17:851-857, 1991 10. Kjekshus JK: Importance of heart rate in determining betablocker efficacy in acute and long-term acute myocardial infarction intervention trials. Am J Cardio157:43F-49F, 1986 11. Randomised trial of intravenous atenolol among 16,027 cases of suspected acute myocardial infarction: ISIS-1. First International Study of Infarct Survival Collaborative Group. Lancet 2:57-66, 1986 12. Gundersen T, Gottum R Pedersen T, et al: Effect of timolol on mortality and reinfarction after acute myocardial infarction: Prognostic importance of heart rate at rest. Am J Cardiol 58:20-24, 1986 13. Autman EM, Lau J, Kupelnick B, et al: A comparison of results of meta-analyses of randomized control trials and recommendations of clinical experts: Treatments for myocardial infarction. JAMA 268:240248, 1992 14. Lau J, Antman EM, Jimenez-Silva J, et al: Cumulative metaanalysis of therapeutic trials for myocardial infarction. N Engl J Med 327:248-254, 1992 15. Mulcahy D, Keegan J, Cunningham D, et al: Circadian variation of total ischaemic burden and its alteration with anti-anginal agents. Lancet 2:755-759, 1988 16. Lange R, Kloner RA, Braunwald E: First ultra-short-acting beta-adrenergic blocking agent: Its effect on size and segmental wall dynamics of reperfused myocardial infarcts in dogs. Am J Cardiol 51:1759-1767, 1983
PERIOPERATIVE USE OF ESMOLOL IN AORTIC SURGERY
17. Neustein SM, Bronheim DS, Lasker S, et al: Esmolol and intraoperative myocardial ischemia: A double-blind study. J Cardiothorac Vasc Anesth 8:273-277, 1994 18. Slogoff S, Keats AS, Ott E: Preoperative propranolol therapy and aortocoronary bypass operation. JAMA 240:1487-1490, 1978 19. Magnusson J, Thulin T, Werner O, et al: Haemodynamic effects of pretreatment with metoprolol in hypertensive patients undergoing surgery. Br J Anaesth 58:251-260, 1986 20. Kirshenbaum JM, Kloner RA, Antman EM, et al: Use of an ultra
561
short-acting beta-blocker in patients with acute myocardial ischemia. Circulation 72:873-880, 1985 21. Coy KM, Imperi GA, Lambert CR, et al: Silent myocardial ischemia during daily activities in asymptomatic men with positive exercise test responses. Am J Cardio159:45-49, 1987 22. Celermajer DS, Spiegelhalter DJ, Deanfield M, et al: Variability of episodic ST segment depression in chronic stable angina: Implications for individual and group trials of therapeutic efficacy. J Am Coll Cardiol 23:66-73, 1994