| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology, Rabin Medical Center, Beilinson Campus, Petah Tikva, Israel;
Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel;
Department of Anesthesiology, Wolfson Medical Center, Holon, Israel;
Department of Anesthesiology, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; and
||Intensive Care Unit and Departments of
¶Cardiology,
#Anesthesiology, and
**Plastic Surgery, Kaplan Medical Center, Rehovot, Israel (affiliated with The Hebrew University School of Medicine, Jerusalem, Israel)
Address correspondence and reprint requests to Lucio Glantz, MD, Department of Anesthesiology, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. Address e-mail to glantzl{at}hotmail.com
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
IMPLICATIONS: This study demonstrates an increased incidence of myocardial ischemia when sternectomy for mediastinitis is performed within one week of coronary artery bypass graft surgery, and this ischemia is associated with a 25% incidence of myocardial infarction.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Previous reports have recommended postponing noncardiac surgery for at least 2 wk after percutaneous transluminal coronary angioplasty with stenting, because operations performed earlier have been associated with an alarming 32% mortality rate (9). However, little information is available regarding the appropriate time to perform noncardiac surgery after CABG surgery.
A frequent incidence of ischemic events has been reported in the early postoperative period of coronary surgery itself, without associated noncardiac surgery. One study (10) found myocardial ischemia in 38% of the patients within the first 2 days of CABG surgery; this decreased to 24% at the end of the first postoperative week. In an earlier study (11), we reported ischemia on the first postoperative day in 45% of patients after CABG surgery. Because 98% of these ischemic events are asymptomatic, the diagnosis is frequently missed.
The incidence of ischemic events during noncardiac surgery performed shortly after CABG surgery has not been characterized. We hypothesized that patients undergoing noncardiac surgery in the first week after CABG surgery are probably at increased risk of perioperative myocardial ischemia compared with patients undergoing this procedure later. To test this assumption, we prospectively compared the incidence of electrocardiographic (ECG) changes consistent with myocardial ischemia in patients who underwent sternectomy during the first and second week after CABG surgery.
Because mortality dramatically decreases with early surgical intervention (12,13), only patients with mediastinitis were selected, as they need to undergo sternectomy as soon as possible after CABG surgery to achieve optimal benefit. Such patients constitute an excellent group for the observation of cardiac events occurring during surgery performed shortly after coronary surgery.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Because the outcome of patients with sternitis improves with early surgical intervention (12,13), sternectomy was performed soon after the diagnosis was established. Only patients undergoing sternectomy during the first (Group 1) or second (Group 2) week after CABG were included in the study.
The study protocol was approved by the IRB, and patients gave their informed consent. Patients whose preoperative ECG did not allow for ST segment analysis because of bundle-branch block or pacing were excluded from the study. All operations were performed by the same plastic surgeon (AL).
The patients continued to receive all cardiovascular medications (nitrates, ß-blockers, and calcium channel blockers) until the operation. Anesthesia and postoperative care were provided by anesthesiologists who were blinded to the study goals.
Immediate postoperative care was provided in the intensive care unit (ICU) or in the postanesthesia care unit (PACU) if there were no beds available in the ICU. IV morphine provided via patient-controlled analgesia devices was used for postoperative pain. Patients were monitored by ECG, pulse oximetry, temperature, and central venous pressure. Blood pressure was measured for at least 24 perioperative hours via a radial arterial catheter inserted 30 min before the induction of anesthesia. A pulmonary artery catheter was inserted only when clinically indicated. An APACHE III (updated Acute Physiology and Chronic Health Evaluation) score was calculated 1 day before sternectomy in both groups of patients to assess disease severity.
ECG changes were assessed with a two-channel AM Holter ECG recorder (MR4000 series; Oxford Instruments, Abingdon, Oxon, UK), beginning 60 min before the induction of anesthesia and continuing for 48 h. Two bipolar leads, V5 and II, were monitored. Because the electrodes were positioned close to the surgical field (particularly V5), they were protected with sterile waterproof tape. The baseline recordings were made with the subject in the supine, sitting, and left and right lateral positions for 5-min periods to determine the effects of positional changes.
Episodes of ischemia were defined as reversible ST segment changes, horizontal or downsloping, lasting for at least 1 min and involving a shift from baseline (adjusted for positional changes) of 
0.1 mV of ST depression (slope 
0) or a 0.2-mV (2-mm) ST increase at the J point (10,14,15). ST segment depression was measured 60 ms after the J point, unless that point fell within the T wave, in which case it was measured at least 40 ms after the J point. The baseline level of the ST segment was defined as its position during a stable period of at least 15 min preceding each episode of change. All episodes of ST-trend segment change were printed at 25 mm/s with a calibrated amplitude signal of 1 mV, equivalent to 1 cm, and were compared with the traces obtained before anesthesia in the same patient’s position. Two investigators blinded to the identity of the patients and their clinical course analyzed the Holter recordings.
Symptoms suggestive of ischemia (chest pain/pressure, shortness of breath, or diaphoresis) were recorded. Because the patients were on bed rest on the first postoperative day, angina was defined as chest pain with typical radiation to the shoulder, jaw, or inner aspect of the left arm, relieved by nitroglycerin in <10 min. Incisional (sternal) pain was defined as pain in the surgical wound, exacerbated by chest movement, without radiation, and unresponsive to nitroglycerin.
A 12-lead ECG was obtained before surgery, on admission to the ICU or PACU, during the first postoperative day, and before hospital discharge. At the same time points, creatinine phosphokinase (CPK) and its myocardial band isoenzyme MB were measured. Myocardial infarct was defined as ECG evidence of a new pathologic Q wave and an increase in the CPK-MB fraction to >5% of the total CPK value (16).
Blood pressure data were obtained from the monitor recordings. The heart rate was evaluated from the Holter trends register. To eliminate artifacts, full disclosure prints were recorded at the time of changes in heart rate trends. The average values recorded for 30 min in the preanesthetic period were considered basal. Because the duration of surgery was a nonuniform variable, we compared only periods of changes 20% of basal values in both groups during surgery.
In previous studies, a 48% incidence of myocardial ischemia was reported after CABG itself during the first postoperative days (10,11), with a significant decrease to 24% at the end of the first week (11). A similarly reduced incidence of ischemia (20%) was reported by Kennedy et al. (15) 1 mo after uneventful CABG surgery.
We expected at least a similar rate of ischemia in patients undergoing sternectomy soon after CABG surgery and a similar 60% change (from 48% to 20%) in patients who were operated on during the second week after CABG. With a significance level (
) of 0.05 and a power of 0.8, a minimum of 100 patients (50 in each group) would be required to reach statistical significance. On the basis of recent 5-yr sternal surgery records in our hospital, we programmed the study for 3 yr. After this period, we opened the study and evaluated the results. At this point, 82 patients were included. Because the difference in the incidence of ischemia between groups was eightfold more frequent than previously hypothesized, the results were statistically significant.
The data are presented as mean ± SD, unless otherwise stated. A statistical comparison of continuous variables was made by using the nonparametric Wilcoxon two-sided statistic. Parametric analyses were performed with the two-sample Student’s t-test, whose results were consistent with those of the nonparametric analyses. P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
|
ECG changes consistent with myocardial ischemia were ST depression in both groups (Table 3). All intraoperative ST changes were associated with increased heart rate and hypotension (defined as a 20% change in baseline values). The postoperative events were independent of hemodynamic changes in 62.5% of the patients.
|
Of the eight ischemic patients in Group 1, two (25%) experienced an acute myocardial infarction, representing a 5.7% infarct incidence in this group. In one patient, with incomplete coronary revascularization (performed in an off-pump setting because of severe aortic root calcification), the infarct was complicated by heart failure, and the patient died of multisystem organ failure 2 wk later. This was the only death in Group 1 (2.8%). There were no infarcts or deaths in Group 2.
In five patients, perioperative myocardial infarct was documented during CABG surgery (Table 1). Ischemic events during sternectomy were not recorded in any of these patients.
Short episodes of self-limited ventricular tachycardia occurred in two patients (Group 1), and paroxysmal atrial fibrillation was observed in five patients (three in Group 1 and two in Group 2). All episodes of dysrhythmia were recorded within 16 h after sternectomy.
The sole complication among patients with preoperative reduced left ventricular function was pulmonary congestion (in one patient during head-down positioning for pulmonary artery catheter insertion before anesthesia), which was promptly resolved when the patient was returned to a semisitting position. Surgery was not canceled.
In total, 95% of the patients were discharged from the ICU or PACU 24 h after sternectomy (P = not significant between groups). A follow-up 3 mo after sternectomy showed no late deaths.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The incidence of myocardial ischemia in Group 1 was similar to that previously reported during the first week after CABG surgery alone (10). The increased incidence in this group was probably related to the unstable period after coronary revascularization itself, which has been attributed to graft occlusion, graft spasm, or incomplete revascularization (17–19).
The less frequent incidence of perioperative ischemia in Group 2 (4.25%) was even less frequent than that previously reported during the first month after CABG surgery alone (15). The incidence of perioperative ischemia in Group 2 in our study was considerably less than the rates reported for noncardiac surgery performed without previous coronary surgery in high-risk patients (20), in patients undergoing high-risk surgery (21), and even in patients undergoing low-risk procedures, such as cataract surgery and fiberoptic bronchoscopy (22,23).
Sternal wound infection is a life-threatening condition that is superimposed on the unstable period close to myocardial revascularization. Differences in the severity of illness before sternectomy due to sepsis or multisystem organ failure may be an important factor in outcome. To evaluate the severity of systemic disease, we calculated the APACHE III score, a reliable predictor of mortality and morbidity in patients after coronary surgery (24,25), on the first preoperative day. No between-group difference in score was found in our study, suggesting a similar health status.
Changes in the ST segment may be due to factors other than coronary ischemia, such as electrolyte concentration, positional changes, drugs, or pericarditis. To increase the specificity of ST changes in our study, we evaluated its reversibility and compared the ST changes with the patient’s basal recording in the same position.
Another limitation is the small sample size, precluding the use of a multivariate analysis approach. After three years, we decided to terminate the study because changes in surgical techniques and postoperative care would make the group less homogeneous during a longer period.
The frequency of myocardial infarct and cardiac death was more in Group 1 than in Group 2. However, because troponin levels were not obtained in all patients, the incidence of perioperative infarct may have been underestimated in both groups. Because the study was not designed to evaluate infarct or death, the sample was too small to draw conclusions.
A control group (patients who did not undergo sternectomy after CABG surgery) was not available for comparison. Compared with historic controls (10,15), the surgical stress of sternotomy does not appear to influence the incidence of perioperative ischemia after CABG surgery.
Another limitation of our study was the lack of patients operated on during the first 72 hours after CABG surgery. Because sternal infections take many days to develop, the minimum interval after cardiac surgery in our sample was four days. Considering that the most frequent incidence of cardiac complications has been reported in the first 48 hours after CABG (11,12), we evaluated only the survivors of this critical period.
Close to 40% of the patients in both groups received ß-blockers. Those included in Group 2 probably had a different pharmacokinetic profile because of longer oral intake after CABG surgery. However, no significant between-group difference in perioperative heart rate changes was observed.
This study shows that the incidence of perioperative myocardial ischemia associated with sternectomy is more frequent when the interval between sternectomy and CABG surgery is less than one week. However, even the increased rate did not differ from the rates previously reported in the literature after CABG surgery alone. Furthermore, myocardial ischemia in sternectomy performed during the first week after CABG was associated with a 25% incidence of acute myocardial infarction. However, patients undergoing sternectomy are a unique, very ill population and probably cannot be compared with those scheduled for noncardiac surgery after uncomplicated CABG surgery. Therefore, further study is required to determine whether the results of this study are generalizable to patients undergoing elective noncardiac surgery soon after uneventful surgical coronary revascularization.
|
Acknowledgments |
|---|
The authors gratefully acknowledge the editorial assistance of Gloria Ginzach.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Lema, J. Urzua, R. Canessa, and L. Glantz Sternectomy After Cardiac Surgery: Noncardiac Surgery? * Response Anesth. Analg., December 1, 2003; 97(6): 1856 - 1857. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
