New segmental wall motion abnormalities (SWMAs) detected by transesophageal echocardiography (TEE) are the most sensitive indicator of intraoperative myocardial ischemia (1–3). Thus, the use of TEE during off-pump coronary artery bypass (OPCAB) surgery might be useful for guiding changes to limit the duration of ischemia in patients who develop marked new SWMAs, e.g., by inserting an adequately sized intracoronary perfusion catheter (4,5) or by repositioning the epicardial stabilizer. Moreover, new SWMAs persisting at the end of surgery may predict a complicated postoperative course (6). However, it has been questioned whether TEE reliably allows for monitoring left ventricular segmental wall motion during OPCAB surgery (7,8): the vicinity between TEE probe and heart is reduced by pericardial retraction, lap pad placement below the heart, and vertical displacement of the heart.
No study has systematically evaluated this question. Therefore, our study was designed to analyze whether TEE reliably allows for monitoring left ventricular segmental wall motion during cardiac displacement for off-pump revascularization of all three main coronary arteries. Using the 16-segment model, we considered availability of ≥14 segments as adequate for reliably monitoring segmental wall motion. We hypothesized that the percentage of patients in whom ≥14 segments were readable would remain >80% after displacement of the heart. In addition, we hypothesized that positioning of the epicardial stabilizer alone may occasionally provoke ischemia before any surgical action to the target coronary artery. Finally, we performed a post hoc analysis to compare the characteristics of patients who developed new SWMAs that persisted at the end of surgery with those of patients without SWMAs.
With approval from the Basel committee on human research, we prospectively studied 60 consecutive patients scheduled for elective OPCAB surgery at the Kantonsspital, University of Basel. Informed patient consent was waived because all patients undergoing OPCAB surgery at our institution are routinely monitored by TEE, standard clinical care was not changed, and surgery was not delayed. Preoperative patient characteristics are presented separately for patients who developed new SWMAs that persisted at the end of surgery and for those who did not have persistent SWMAs (Table 1).
Routine patient monitoring and anesthetic procedures were performed as previously described (2,9). In short, anesthesia was induced with thiopentone (3–5 mg/kg) or etomidate (0.2–0.3 mg/kg), midazolam (50–80 μg/kg), fentanyl (2–4 μg/kg), and pancuronium (0.1 mg/kg). Anesthesia was maintained with isoflurane (0.4%–0.6% end-expiratory concentration) and repeated boluses of fentanyl (1–4 μg/kg each) as needed. Controlled mechanical ventilation with 50% oxygen in air was provided to achieve normocapnia as indicated by continuous end-tidal CO2 monitoring and intermittent arterial blood gas studies. A central venous catheter was inserted after the induction of anesthesia in each of the 60 patients, and a pulmonary artery catheter was inserted in 43 patients. Mean pulmonary artery pressure was documented at each study time point, whereas pulmonary artery occlusion pressure and cardiac index were systematically measured only at baseline, after insertion of an intracoronary perfusion catheter, and at the end of surgery.
To prevent hypotension during surgery, each patient received 500 mL of 6% hetastarch in a 0.9% sodium chloride injection (B. Braun, Melsungen, Germany) after the induction of anesthesia. If hypotension occurred (defined as mean arterial blood pressure <60 mm Hg), the following was sequentially done: 1) the legs were raised; 2) up to 500 mL of additional hetastarch was administered; 3) a bolus of ephedrine 2.5–5 mg or phenylephrine 50–100 μg, as judged appropriate by the anesthesiologist in charge on the basis of the heart rate, was administered. If hypertension (defined as systolic arterial blood pressure ≥160 mm Hg) occurred despite adequate depth of anesthesia, either nitroglycerin 25–50 μg (Schwarz Pharma, Monheim, Germany) or, in the presence of tachycardia, esmolol 0.5 mg/kg (Baxter Healthcare Corp., Deerfield, IL) was administered.
TEE studies were performed with a multiplane 4- to 7-MHz TEE probe and a Sonos 5500 imaging system (Philips, Best, The Netherlands). During OPCAB surgery, the midesophageal four-chamber view, two-chamber view, and long-axis view were obtained for clinical monitoring and for subsequent scientific evaluation at the following study time points: after median sternotomy (baseline), immediately after the epicardial stabilizer was placed, while the distal anastomosis was performed, and at the end of surgery. Three cardiac cycles of each view were obtained at each time point and digitally stored on reusable 2.3-GB magneto-optical discs (Hewlett-Packard, Andover, MA). Marked new SWMAs detected by the study echocardiographer were communicated to the anesthesiologist and surgeon in charge of the patient, but study results were based on offline analysis of the stored cycles. For electrocardiographic (ECG) detection of ischemia, leads I, II, III, aVL, aVF, V5, and V6 were monitored and printed on paper after patient arrival in the operating room and at each study time point, simultaneously with TEE data acquisition (Schiller AT 10; Schiller, Baar, Switzerland).
Surgical procedures were performed as described by Sergeant et al. (10). In short, midline sternotomy was performed in all patients, followed by preparation of the left internal mammary artery for subsequent grafting in most patients. The pericardium was suspended to the left side. In contrast, the right-sided pericardium was never suspended to achieve mobility of the heart while distal anastomoses were performed. A V-shaped sponge was fixed down on the right posterior mediastinum halfway on the line joining the inferior vena cava to the left inferior pulmonary vein. Closing and retracting the V-shaped sponge enucleated the left atrium and the left ventricle. The epicardium in the area of the distal anastomosis was immobilized by an Octopus III epicardial stabilizer (Medtronic, Minneapolis, MN). The Starfish heart positioner (Medtronic) was not used in any of the study patients. To decrease the duration of ischemia (4,5) and to achieve a bloodless surgical field, intracoronary perfusion catheters were used in coronary vessels with a diameter >1.5 mm. The left internal mammary artery was used as the only arterial graft and was routinely anastomosed as the first graft to the left anterior descending coronary artery (LAD) or a diagonal branch. Venous grafts were always attached to the aorta. Shed blood was collected at the operating field and retransfused (Cobe BRAT 2; Cobe Cardiovascular, Inc., Arvada, CO). The amount of blood loss was estimated by the anesthesiologist in charge of the patient. The decision to switch to cardiopulmonary bypass during the procedure was based on significant hemodynamic instability and/or ventricular arrhythmia.
All scientific analyses were performed offline by using the digital recordings. Segmental wall motion of the left ventricle was analyzed by using the 16-segment model and the 5-grade scale according to current guidelines (11). By considering both endocardial motion and myocardial thickening, the grading system defines score 1 = normokinesis, score 2 = mild hypokinesis, score 3 = severe hypokinesis, score 4 = akinesis, and score 5 = dyskinesis (11). As previously described (9), at least 50% of the endocardial and epicardial border had to be visible in a segment graded as normal to be considered adequate for the analysis of wall motion, and approximately 33% had to be visible in a segment graded as abnormal. If the epicardium was not visible, a segment was still considered adequate for analysis if the endocardial border was almost completely visible (approximately ≥90%) throughout the cardiac cycle. Segments not fulfilling these criteria were graded as 0 = no view. Two experienced readers independently performed wall motion analyses by using split screens. The baseline finding obtained after opening of the sternum was always presented on the left side of the screen, whereas the readers were blinded to the time point of the loop on the right-hand side. Full blinding was achievable for most revascularizations in the area of the LAD. In contrast, the angle of the heart in loops obtained after vertical displacement for revascularizations in the areas of the left circumflex coronary artery (LCX) and right coronary artery (RCA) was different from that seen in the baseline loops, preventing full blinding of the readers. Therefore, the readers could only be fully blinded to the steps “epicardial stabilizer placed” and “distal anastomosis” (target coronary artery occluded/perfusion catheter inserted). Consensus readings were performed when the initial scores of the two readers were not in agreement (as defined below) but were used for analysis only if a third reader independently confirmed the consensus reading. Otherwise, or if no agreement could be obtained at the consensus readings, the segment was regarded as unreadable. The wall motion score index at each step of the study was calculated by dividing the sum of scores of all readable segments by the number of readable segments.
The number of left ventricular segments available for monitoring by TEE during OPCAB surgery was analyzed for each individual patient on the basis of the recordings obtained after placement of the epicardial stabilizer. The number of readable segments after cardiac displacement and placement of the epicardial stabilizer on each of the three main coronary arteries was calculated separately, and the findings were compared with the number of segments readable at baseline. We also calculated how many of the segments in the territory of the coronary artery being anastomosed were readable. Ten segments were completely or partially attributed to the LAD: the basal and middle anterior, anteroseptal, and septal segments and all four apical segments. Five segments were attributed completely or partially to the LCX: the basal, middle, and apical lateral segments and the basal and middle posterior segments. Three segments were attributed to the RCA: the basal, middle, and apical inferior segments. Finally, we calculated the number and percentage of patients in whom ≥14 segments were readable at each study period. A segment was regarded as readable if 1) both readers assigned a grade other than 0, and 2) the score assigned by the two readers differed by less than one grade. Consensus readings were performed for all segments with initial readings that differed by more than one grade.
The incidence of new ischemia during OPCAB surgery was analyzed with digital TEE recordings and paper printouts of the seven-lead ECG. TEE analysis for detection of ischemia was performed by comparing the three midesophageal views obtained at each subsequent study time point with the corresponding baseline views obtained after sternotomy. Evidence of ischemia was defined as worsening of segmental wall motion by two or more grades in two or more segments in the territory vascularized by the target coronary artery. These marked changes in wall motion were required to maintain a high specificity of TEE for diagnosing ischemia in a situation when displacement of the heart and placement of an epicardial stabilizer can complicate the analysis of wall motion. Consensus readings were performed if one reader diagnosed these changes but the second reader did not.
ECG evidence of ischemia in the seven-lead ECG was defined as ≥1-mm horizontal or down-sloping ST segment depression or ≥1-mm horizontal ST segment elevation at 60 ms after the J point; if baseline ST segment changes were present, an additional ≥2-mm ST segment shift was required for diagnosis of ischemia. If the amplitude of the QRS complex was <5 mm, the ECG was excluded from analysis.
Diagnosis of postoperative myocardial infarction was based on 12-lead ECGs performed in all patients upon arrival in the intensive care unit, on the first and second postoperative mornings, and before hospital discharge. Myocardial infarction was diagnosed if Q waves of ≥0.1 mV and ≥0.04 s developed in ≥2 leads in the 12-lead ECG. A single experienced reader who was blinded to all other patient data analyzed the intraoperative and postoperative ECGs offline.
Analyses of patient characteristics and intraoperative cardiovascular findings were performed separately for patients with new SWMAs persisting at the end of surgery versus those without. Patients with new SWMAs persisting at the end of surgery were invited for an echocardiographic follow-up examination 6–12 mo after surgery. Transthoracic echocardiography was performed in the left lateral decubitus position by using a Sonos 5500 imaging system and a 1.8- to 3.6-MHz probe (Philips). The three apical views by transthoracic echocardiography were obtained to compare segmental wall motion at the follow-up with segmental wall motion obtained by the three midesophageal TEE views at baseline and at the end of surgery. Wall motion analysis of transthoracic images was performed identically as described for TEE images.
Continuous variables are presented as mean ± sd, and dichotomous variables are presented as numbers and percentages. For all statistical analyses, a P value <0.05 was considered statistically significant. The χ2 test or Fisher's exact test was used for analysis of dichotomous variables. Baseline hemodynamic findings of the two groups—those with persisting new SWMAs versus those without—were analyzed with the Mann-Whitney U-test, as were the numbers of readable segments at each study period. Repeated intergroup and intragroup comparisons of continuous variables at the different study steps were performed by analysis of variance for repeated measures followed by the Scheffé post hoc test. All statistical analyses were performed with StatView 5.0 (SAS Institute Inc., Cary, NC). McNemar's test for paired proportions was used to compare the incidence of ischemia at baseline and when distal anastomoses were performed. Reproducibility of wall motion readings is indicated by the κ coefficient for interobserver agreement regarding the presence or absence of new SWMAs according to our study definitions and by the κ coefficient for intraobserver agreement after repeated readings. These two analyses were performed with GraphPad QuickCalcs (GraphPad Software, San Diego, CA).
OPCAB was completed as planned in 55 of the 60 study patients. In three patients, one of the planned grafts was not performed after the epicardial stabilizer had been placed, because the LCX in one patient and the RCA in a second patient were judged too small for revascularization at the inspection after placement of the stabilizer. In the third patient, the epicardial stabilizer could not be positioned on the RCA without inducing severe ischemia and hemodynamic instability. In another two patients, conversion to cardiopulmonary bypass was performed because of severe ischemia and marked hemodynamic instability after placement of the epicardial stabilizer on the RCA. In both patients, a left internal mammary artery graft to the LAD had been performed off-pump, and two additional venous grafts were performed after conversion to cardiopulmonary bypass. In summary, 148 bypass grafts were performed in the 60 study patients; 144 were performed off-pump. Intraoperative patient characteristics are described in Table 2. The number of grafts performed in the group of patients without persisting SWMA was larger, but the use of perfusion catheters was less frequent than in the group of patients with SWMAs persisting at the end of surgery.
A total of 76 off-pump grafts were performed in the LAD territory: in 43 patients, grafts were made to the LAD; in 16 patients, both to the main LAD and to a diagonal branch of the LAD; and in 1 patient, to a diagonal branch only. A total of 40 off-pump grafts were performed in the LCX territory: in 24 patients, grafts were made to the main LCX, and in 16 patients they were made to a marginal artery. A total of 28 off-pump grafts were performed in the RCA territory; Table 3 shows the findings after 31 placements of an epicardial stabilizer on the RCA territory, including 3 additional patients in whom the epicardial stabilizer was placed but no anastomosis could be performed off-pump. Thus, Table 3 shows findings after placement of the stabilizer on the main RCA in 25 patients and on distal branches of the RCA in 6 patients. A perfusion catheter was inserted during 56 distal anastomoses to the LAD, during 25 anastomoses to the LCX, and during 14 anastomoses to the RCA.
Cardiac displacement and placement of the epicardial stabilizer caused a decrease in the number of readable left ventricular segments (Table 3) that reached statistical significance after placement of the stabilizer on the LAD and LCX territories. The percentage of patients with ≥14 readable segments decreased significantly after cardiac displacement and stabilizer placement on the LAD and LCX territory; ≥14 segments were still readable in approximately 4 of 5 patients. An example of the midesophageal view at baseline and after cardiac displacement and placement of an epicardial stabilizer on the transiently occluded RCA is given in Figure 1.
The overall regional wall motion at baseline and its subsequent changes are illustrated by the wall motion score index in Figure 2. This index was significantly increased during surgery both in the group of patients who did not develop persisting new SWMAs and in those who did, and in the latter group it did not normalize to baseline values at the end of surgery. The differences between groups reached statistical significance while the distal anastomosis was performed and at the end of surgery. Table 4 shows that new SWMAs fulfilling the predefined criteria for the diagnosis of ischemia were frequently detectable during surgery on all coronary arteries. Placement of an epicardial stabilizer alone induced the predefined TEE and/or ECG criteria of ischemia in up to 38%. New SWMAs fulfilling the predefined criteria for ischemia were detected during 57 distal anastomoses in 41 patients.
Persisting new SWMAs were found at the end of OPCAB surgery in 22 patients. In each of two patients, the persisting new SWMAs were partially or totally in the supply area of a stenotic coronary artery that had not been revascularized. Four of the 22 patients also fulfilled ECG criteria for ischemia. Myocardial infarction was subsequently confirmed in 2 of these 4 patients by new pathologic Q waves in the 12-lead ECG; the ECG in the 2 other patients normalized within 2 days. Twenty patients with persisting SWMAs returned for follow-up echocardiography ≥6 mo after surgery. The SWMAs had recovered in 11 patients and persisted in 9 patients. Persistent SWMAs had already been detected in 6 of these 9 patients immediately after placement of the epicardial stabilizer, i.e., before any surgical action to the supplying coronary artery.
Cardiovascular medication and fluids administered during OPCAB surgery are described in Table 2, and hemodynamic findings are listed in Table 5. Hemodynamic differences between the groups of patients who did or did not develop new SWMAs that persisted at the end of surgery were small, as were hemodynamic changes during surgery. The only statistically significant intergroup differences were between the baseline values of central venous pressure, mean pulmonary artery pressure, and pulmonary artery occlusion pressure. An increase in heart rate was found in both groups during surgery, but in the group without persisting SWMAs, it reached statistical significance only at the end of surgery. A small but statistically significant decrease in mean arterial blood pressure was found in the group with persisting SWMAs at the time when the distal anastomoses were performed and at the end of surgery.
The percentage of agreement of the two independent readers before and after consensus reading is described in Table 3, which also describes why some segments were considered unreadable: “no view” means that one or both readers thought that the segment was not sufficiently visible (grade 0). “Disagreement between readers” means that both readers regarded the segment as sufficiently visible, although they could not agree on scores that were within one grade, or that their consensus reading was not confirmed by a third independent reader. The κ coefficient for interobserver agreement regarding presence or absence of a new SWMA fulfilling the study criteria for the diagnosis of ischemia was 0.81; κ for intraobserver agreement of repeated readings was 0.91 and 0.93.
Our study showed that TEE midesophageal views allow for monitoring left ventricular segmental wall motion in most patients during cardiac displacement and placement of an epicardial stabilizer for off-pump revascularization of all three main coronary arteries. Although ≥14 segments were readable less frequently after stabilizer placement on the LAD and LCX territories, this number of readable segments was still found in approximately 4 of 5 patients. Moreover, the number of readable segments in the supply area of the coronary artery being anastomosed was also large in most patients. These findings, in conjunction with a high reproducibility of wall motion readings, show that high-quality visualization of left ventricular segmental wall motion was obtained in most patients—a prerequisite for echocardiographic monitoring for ischemia. Our finding that TEE allows for monitoring left ventricular wall motion is in agreement with previous studies that included off-pump revascularizations of the LAD (6,12,13) and a small number of revascularizations of the RCA (6). Our findings allow for extending and generalizing the conclusion that left ventricular segments can be imaged and monitored reliably by TEE during OPCAB surgery.
Left ventricular segments could be imaged by relying on midesophageal views. The transgastric views were not analyzed in this study because a pilot study had shown that they are not obtainable in most patients after lap pad placement below the heart and cardiac displacement. The loss of transgastric views is not a limitation of TEE during OPCAB surgery because the 3 midesophageal views allow for comprehensive echocardiographic monitoring of all 16 segments of the left ventricle (11).
Reliable visualization of segmental wall motion at a quality that allows for reproducible wall motion analysis is the basis for echocardiographic monitoring for ischemia. Myocardial ischemia during off-pump revascularization, as based on the occurrence of a new SWMA, was diagnosed less frequently in our study than in previous OPCAB studies. Previous OPCAB studies found severe new SWMAs during coronary occlusion in nearly all patients who did not have a previous myocardial infarction in the corresponding region of the left ventricle (6,12,13). Potential reasons for the decreased incidence of diagnosed ischemia in our study is that we required a larger degree and extent of wall motion abnormality for the diagnosis of ischemia (worsening of segmental wall motion by two or more grades in two or more segments in the territory vascularized by the target coronary artery) than previous studies (6,12,13); the frequent use of intracoronary perfusion catheters, which may prevent ischemic impairment of ventricular function (5); and a potentially different degree of coronary stenosis. We were aware that the strict TEE criteria might reduce the sensitivity of TEE for detection of ischemia, but our goal was to maintain its specificity in a situation in which the placement of an epicardial stabilizer and changes in preload can complicate the analysis of segmental wall motion. An epicardial stabilizer may complicate wall motion reading by markedly reducing epicardial movement, although it will not eliminate systolic myocardial thickening in the absence of ischemia. Acutely reduced preload may cause a localized SWMA in the absence of ischemia in patients with preexisting wall motion abnormalities (9). These observations strongly suggest that small reductions in segmental wall motion in the area influenced by an epicardial stabilizer or a marked reduction in an isolated segment may not always be indicative of ischemia during OPCAB surgery.
It might be postulated that the concept of OPCAB surgery implies that transient myocardial ischemia is elicited and that only major changes in wall motion (indicating impending hemodynamic instability) are of clinical importance. However, our findings obtained after placement of an epicardial stabilizer suggest that more accurate analysis of segmental wall motion might be of clinical importance during OPCAB surgery. The question is whether the stabilizer just locally reduces segmental wall motion or whether placement itself induces myocardial ischemia in some patients, thus prolonging the total duration of ischemia. In our hemodynamically stable patients, predefined TEE criteria of myocardial ischemia were met after placement of an epicardial stabilizer in 23% of patients (Table 4). Several aspects of our findings strongly suggest that our predefined strict TEE criteria were useful for diagnosing ischemia, i.e., that placement of a stabilizer occasionally induced ischemia. First, established ECG criteria for ischemia were found after 9 (6%) of 147 stabilizer placements (Table 4) despite frequent loss of an ECG of sufficient QRS size after cardiac displacement. Second, OPCAB surgery was abandoned because of severe ischemia and hemodynamic instability after placement of an epicardial stabilizer in three patients, and this was complicated by ventricular fibrillation in one of them. Third, in 6 of 9 patients with a new SWMA that persisted 6 or more months after surgery, the new SWMA was detected immediately after placement of the epicardial stabilizer, i.e., before any surgical action to the supplying coronary artery. We hypothesize that compression and/or suction applied by the epicardial stabilizer may cause myocardial ischemia by functional occlusion of the side branches or the target coronary artery itself. Cardiac displacement can further contribute to the development of ischemia if decreases in arterial blood pressure and, subsequently, in coronary flow occur. Our observation of ischemia caused by cardiac displacement and placement of an epicardial stabilizer is consistent with the observation that cardiac displacement and use of an apical suction device also can induce ischemia in some patients (14).
Our frequent use of intracoronary perfusion catheters is in contrast with previously published studies (6,12,13) and might have contributed to the decreased incidence of ischemia found during distal anastomoses (5) in our study. Conversely, an animal study has found impairment of endothelial function by adequately sized perfusion catheters (15). Our post hoc comparison of intraoperative patient characteristics, surprisingly, showed that perfusion catheters had been used more frequently in patients with persistent SWMAs (Table 2). However, our study was not designed to analyze the use of perfusion catheters, and multiple confounders may have influenced this result; e.g., one important indication for using a perfusion catheter is an increased risk of ischemia and hemodynamic instability (5,16).
The finding of new SWMAs persisting at the end of surgery in 22 of 60 patients, and several months after surgery in 9 of them, raises important questions. This includes the issues of whether persistent SWMAs are more frequent after OPCAB surgery or after on-pump coronary artery bypass surgery and what factors contribute to their development. Future studies are needed to analyze these clinically important issues.
A limitation of our study was that its results are based on offline evaluation of segmental wall motion, whereas online evaluation must be performed in clinical practice. Online evaluation is less sensitive for detecting SWMAs, especially if changes in wall motion are mild (17). Although side-by-side comparison of digital loops acquired at different time points has become available in the operating room, online evaluation of segmental wall motion is still complicated by the need for immediate diagnosis in a busy setting. Another limitation is that our TEE study protocol did not include a systematic evaluation of mitral valvular function, although new or worsened mitral regurgitation may be useful as an indicator of acute ischemia. A further limitation is that our study did not analyze the ability to obtain TEE images during use of the Starfish heart positioner (Medtronic). Our anecdotal experience of maintained ability to obtain TEE images with the Starfish device is in agreement with the anatomical consideration that the midesophageal views are obtained from the base of the heart that is not lifted from the neighboring esophagus, whether or not the Starfish device is used. Our limited experience and the lack of a scientific analysis do not allow for definitive conclusions on this point. Finally, it must be noted that our study analyzed whether left ventricular segments can be monitored during OPCAB surgery but not whether such monitoring influences patient outcome.
In conclusion, our study shows two findings. First, TEE reliably allows for the visualization of left ventricular segmental wall motion during OPCAB surgery in most patients—a prerequisite for echocardiographic monitoring for ischemia. Second, placement of an epicardial stabilizer without surgical intervention on the target coronary artery can induce ischemic ST segment changes and new SWMAs, and some of these SWMAs might persist for six or more months after surgery.
The authors thank Claudia Werner, RNA, and Esther Seeberger, RNIC, for help with data collection and analysis and Joan Etlinger, BA, for editorial assistance.
1. Smith MD, MacPhail B, Harrison MR, et al. Value and limitations of transesophageal echocardiography in determination of left ventricular volumes and ejection fraction. J Am Coll Cardiol 1992;19:1213–22.
2. Seeberger MD, Skarvan K, Buser P, et al. Dobutamine stress echocardiography to detect inducible demand ischemia in anesthetized patients with coronary artery disease. Anesthesiology 1998;88:1233–9.
3. Comunale ME, Body SC, Ley C, et al. The concordance of intraoperative left ventricular wall-motion abnormalities and electrocardiographic S-T segment changes: association with outcome after coronary revascularization—Multicenter Study of Perioperative Ischemia (McSPI) Research Group. Anesthesiology 1998;88:945–54.
4. Coulson A, Bakhshay S, Quarnstrom J, Spohn P. Temporary coronary artery perfusion catheter during minimally invasive coronary surgery. Chest 1998;113:514–20.
5. Lucchetti V, Capasso F, Caputo M, et al. Intracoronary shunt prevents left ventricular function impairment during beating heart coronary revascularization. Eur J Cardiothorac Surg 1999;15:255–9.
6. Moises VA, Mesquita CB, Campos O, et al. Importance of intraoperative transesophageal echocardiography during coronary artery surgery without cardiopulmonary bypass. J Am Soc Echocardiogr 1998;11:1139–44.
7. Gayes JM. The minimally invasive cardiac surgery voyage. J Cardiothorac Vasc Anesth 1999;13:119–22.
8. Platt MJ, Davies S, Riedel BJ, et al. Case 4–2002: near-fatal pulmonary embolism in the immediate postoperative period after off-pump coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 2002;16:502–7.
9. Seeberger M, Cahalan M, Foster E, et al. Acute hypovolemia may cause segmental wall motion abnormalities in the absence of myocardial ischemia. Anesth Analg 1997;85:1252–7.
10. Sergeant P, de Worm E, Meyns B, Wouters P. The challenge of departmental quality control in the reengineering towards off-pump coronary artery bypass grafting. Eur J Cardiothorac Surg 2001;20:538–43.
11. Shanewise JS, Cheung AT, Aronson S, et al. ASE/SCA guidelines for performing a comprehensive intraoperative multiplane transesophageal echocardiography examination: recommendations of the American Society of Echocardiography Council for Intraoperative Echocardiography and the Society of Cardiovascular Anesthesiologists Task Force for Certification in Perioperative Transesophageal Echocardiography. Anesth Analg 1999;89:870–84.
12. Brown PM Jr, Kim VB, Boyer BJ, et al. Regional left ventricular systolic function in humans during off-pump coronary bypass surgery. Circulation 1999;100(suppl 19):II125–7.
13. Malkowski MJ, Kramer CM, Parvizi ST, et al. Transient ischemia does not limit subsequent ischemic regional dysfunction in humans: a transesophageal echocardiographic study during minimally invasive coronary artery bypass surgery. J Am Coll Cardiol 1998;31:1035–9.
14. D'Ancona G, Karamanoukian H, Kawaguchi A, Bergsland J. Coronary artery exposure in off-pump CABG: a word of caution. Heart Surg Forum 2001;4:243–5; discussion 245–6.
15. Demaria RG, Fortier S, Malo O, et al. Influence of intracoronary shunt size on coronary endothelial function during off-pump coronary artery bypass. Heart Surg Forum 2003;6:160–8.
16. Angelini GD, Taylor FC, Reeves BC, Ascione R. Early and midterm outcome after off-pump and on-pump surgery in Beating Heart Against Cardioplegic Arrest Studies (BHACAS 1 and 2): a pooled analysis of two randomised controlled trials. Lancet 2002;359:1194–9.
17. Bergquist BD, Leung JM, Bellows WH. Transesophageal echocardiography in myocardial revascularization. I. Accuracy of intraoperative real-time interpretation. Anesth Analg 1996;82:1132–8.