Total spinal anesthesia for cardiac surgery: does it make a difference in patient outcomes?
The normal physiological response to stress occurs as a protective and adaptive mechanism that is designed to maintain homeostasis within the body (McEwen & Mendelson, 1993). Selye (1946) described the first two phases of the general adaptation system (GAS) as alarm and resistance. These two phases of the stress response are protective and adaptive. However, the third phase of the GAS is exhaustion, and results in pathophysiologic decompensation (Mitchell, Gallucci, & Fought, 1991). Prolonged or excessive stress has the potential to induce physiologic exhaustion.
Cardiac surgery is a potent physiological stressor that initiates the release of a cascade of catecholamines and other stress hormones (Hoar, Stone, Falatas, Bendixen, Head, & Berkowitz, 1980; Lee, Grocott, Schwinn, & Jacobsohn 2003). A sustained or poorly managed surgical stress response has the potential to cause adverse consequences in the perioperative period including pain, cardiac ischemia and hemodynamic instability, renal decompensation, pulmonary decompensation, increased catabolism, impaired immunity, and hypercoagulability syndromes (Lubenow, Ivankovich, & McCarthy, 2001). To manage and reduce the surgical stress response, regional anesthesia has become an integral component of cardiac anesthesia in many tertiary centres (Lee & Jacobsohn, 2008).
With the goals of attenuating the surgical stress response, reducing morbidity, and improving patient outcomes, one cardiac anesthesiologist at a Canadian tertiary care centre initiated the use of a novel anesthetic approach that combines total spinal anesthesia (TSA) with intrathecal (IT) morphine and general anesthesia (GA), hereafter referred to as TSA. To date, this unique method of anesthesia has not been explored from a nursing perspective. Therefore, the goal of this research was to add to the knowledge of patient outcomes following TSA and thereby enhance the ability of critical care nurses to anticipate the needs of this unique group of cardiac surgery patients.
Review of the literature
Ideally, perioperative anesthetic and analgesic interventions should focus on preventing the surgical stress response rather than controlling its adverse sequelae (Lubenow et al., 2001). Accordingly, the following review of the literature provides a brief overview of the evolution of the use of TSA in cardiac surgery.
The ongoing quest for a superior cardiac anesthetic technique has been the impetus for the use of various adjuncts to GA. Researchers have explored the use of perioperative, high-dose, intravenous opioids (Liem, Booij, Hasenbos, & Gilien, 1992a; Mangano et al., 1992), epidural anesthesia (Liem, Williams, Hensens, & Singh, 1998; Loick et al., 1999), and high-dose IT morphine (Aun, Thomas, St. John-Jones, Colvin, Savege, & Lewis, 1985; Boulanger, Perreault, Choiniere, Prieto, Lavoie, & Laflamme, 2002; Chaney, Furry, Fluder, & Slogoff, 1997; Vanstrum, Bjornson, & Ilko, 1988). To date, however, the ideal cardiac anesthetic technique has not been established.
In a recent clinical trial of coronary artery bypass graft (CABG) surgery patients (n=38), Lee and associates (2003) compared the use of GA in combination with high-dose IT bupivicaine (ITB-GA) spinal anesthesia to GA alone. Results revealed that serum epinephrine, norepinephrine, and cortisol levels were significantly lower in the ITB-GA group. In the pre-cardiopulmonary bypass (CPB) period, the ITB-GA group had significantly lower mean arterial pressures and systemic vascular resistance (SVR) indexes. The ITB-GA group also had improved regional left ventricular wall motion, as measured by echocardiography in the post-intubation and the pre-CPB periods. Post-CPB, the ITB patients had a higher cardiac index and lower pulmonary vascular resistance, as well as a trend towards improved left ventricular wall motion. Lee and associates (2003) concluded that the ITB-GA technique reduces the surgical stress response and [beta]-receptor dysfunction in CABG surgery patients. Moreover, these findings are consistent with previous research that suggested numerous benefits of cardiac sympathetic blockade, including relief of angina, and improvement in ST segment depression, oxygen supply and demand ratio in ischemic myocardium, diameter of stenotic epicardial coronary arteries, and left ventricular function (Blomberg et al., 1990; Blomberg, Emanuelsson, & Ricksten, 1989; Klassen, Bramwell, Bromage, & Zborowska-Sluis, 1980; Kock, Blomberg, Emanuelsson, Lomsky, Stomblad, & Ricksten, 1990).
Based on the evidence from Lee and associates' (2003) research and a subsequent ITB study (Jacobsohn, Lee, Amadeo, Syslak, Debrouwere, & Bell, 2005), the ITB-GA technique has been modified as described in Lee and Jacobsohn (2008) to combine total spinal anesthesia with IT morphine and GA for cardiac surgery patients.
It is postulated that this novel TSA technique attenuates the surgical stress response, and consequently improves perioperative outcomes of cardiac surgery. With this perspective, the purpose of our pilot study was to describe and explore the outcomes of the TSA technique and to compare these outcomes with a matched control sample that received GA.
The use of theory offers structure and organization to nursing knowledge. In research, nursing theory should provide a systematic framework for data collection and analysis in order to describe, explain, and predict nursing practice. Accordingly, the Human Response to Health and Illness Model (HRM) (Heitkemper & Shaver, 1989) was adapted and used to guide this research. This biopsychosocial model clarifies nursing's domain as the patient's response to the health problem or stressor and not to the stressor itself. The multidimensional approach of the HRM includes physiologic, pathophysiologic, behavioural, and experiential perspectives. As well, modifiable and non-modifiable person factors, and environmental risk factors influence the response to clinical therapeutics, and individual adaptation. The HRM provided organizational structure to the study development, implementation, and data analysis and, thus, was the driving force for this research.
We used a retrospective, descriptive correlational design in this research. The study setting was an urban tertiary care institution, where more than 700 cardiac surgeries were performed in 2005. The study population included the 50 TSA cases and all GA cases completed between September 2003 and May 2005. Based on previous research related to extubation following CABG surgery (Liem et al., 1992b), we determined that a sample size of 25 TSA patients and a matched sample of 25 CGA patients would provide an 80% power in detecting a 27% reduction in extubation time in the TSA group (p=0.05). The matched pair sample was randomly selected, controlling for pertinent factors, including age (within 10 years), gender, previous cardiac surgery, type of CABG versus valve replacements, surgeon, no contraindications to TSA, and approximate date of surgery (within three months).
The research literature provided insight into the selection of variables to operationalize the complex surgical stress response. The HRM provided the organizational structure to the numerous relevant variables identified (see Tables One to Five). Following ethics approval, patient consents were obtained and chart review data collection was completed. Statistical consultation was provided by the University of Manitoba Biostatistical Consulting Unit, using the Statistical Analysis System version 9.0 (ASA institute Inc., Cary, NC).
The study sample (n=70) included 35 TSA and 35 GA patients. According to the HRM, person factors are modifiable or non-modifiable patient characteristics or demographic factors (Mitchell et al., 1991). When these characteristics were compared, there were no significant differences between the two groups. Thus, the typical patient in the study cohort was male, in his sixth or seventh decade of life, with hypertension, a history of a previous MI with a LVEF >40%, undergoing surgery for two to three CABGs. The majority of patients in both groups had co-morbidities of hypertension, moderate obesity (BMI 29.8 to 33.3 m2), and a history of smoking. Most frequent medical therapies included [beta]-blockers, lipid-lowering agents, and ACE inhibitors (see Tables One to Three).
The pathophysiological responses were operationalized with variables that would indicate decompensation. Although multivariate analysis was not feasible due to the small sample size, the overall trend was that TSA patients were less likely to experience decompensation. They had fewer complications in 11 of the 16 pathophysiological responses (see Table Four).
The behavioural response component of the HRM includes the observable and measurable components of assessment, which can be used to identify, diagnose, and evaluate the specific patient response (Mitchell et al., 1991). The behavioural perspective was operationalized by measurements of patient comfort, length of endotracheal intubation, and ICU and hospital length of stay (LOS).
During the initial 24-hour postoperative period, morphine administration was significantly different between the two groups, with the TSA group receiving less morphine than the GA group (see Table Five). There were no significant differences between the two groups in total use of other analgesics. For example, average postoperative acetaminophen with 30 milligrams of codeine tablet use in the TSA group and the GA group was essentially the same ([bar.x] = 13 [+ or -] 7).
Of particular note, TSA patients were significantly more likely to be extubated in the operating room (82.4% versus 29.4%; [X.sup.2] = 18.65; df = 2; p <0.0001). Overall, the duration of postoperative endotracheal intubation was significantly shorter in the TSA group compared to the GA group (p<0.0008). On average, the TSA group was extubated 4.9 hours earlier than the GA group. As well, the median difference in intubation time between the two groups was 1.5 hours, indicating that, on average, the TSA group was extubated 1.5 hours prior to their matched pair.
The mean differences in surgical intensive care unit (SICU) LOS and in-hospital LOS for the TSA versus the GA group were not statistically significant. However, the mean total number of hospital days until discharge was three days less in the TSA group than the GA group (7.5 versus 10.4 days). As well, 50% of the TSA sample (n=17) was discharged < 5 days postoperatively compared to 35% of the GA group (n=11).
The HRM provided an appropriate framework to explore the outcomes of a medical intervention from a nursing perspective. Matched pair analysis of person factors was an effective strategy to strengthen the comparisons of the pathophysiological responses by effectively eliminating statistically significant variable differences. Although not statistically significant, the trends in the data suggested that the TSA group had fewer perioperative complications.
To compare the different anesthetic approaches, pathophysiological outcome variables were selected for measurement. The occurrence of postoperative myocardial infarctions, new onset atrial fibrillation, re-operation for bleeding, vasopressor support in SICU, inotrope support in SICU, and postoperative requirement for intra-aortic balloon pump are indicators of hemodynamic instability. For example, the increased likelihood of re-operation for bleeding and greater use of vasopressors in the GA group lend support for the hypothesis that surgical stress response leads to decompensation. A sustained surgical stress response, which burdens the cardiovascular system by increasing HR, myocardial contractility, and systemic vascular resistance may result in hypertension, tachycardia, and dysrhythmias and, in turn, may lead to myocardial ischemia in susceptible patients (Lubenow et al., 2001).
The differences in complication rates between the two groups may have clinical significance. For example, while differences in the re-operation for bleeding variable were not statistically significant, adjusting for sample size revealed and odds ratio predicted that the GA group was 7.3 times more likely to have a re-operation for postoperative bleeding than the TSA group. The physiologic basis for this difference is unclear; however, based on the study hypothesis, TSA may decrease surgical stress response, resulting in hemodynamic stability, less stress on new surgical sites, and consequently, less bleeding post-operatively.
There was also a trend for the TSA group to have a lower occurrence of postoperative peak increases of serum creatinine levels, which suggests a positive effect of the TSA technique. This finding is consistent with previous research. In a randomized trial of the potential benefits of thoracic epidural anesthesia (TEA) and analgesia in patients undergoing CABG, Scott and associates (2001) reported a significant reduction in postoperative renal failure, as defined by a twofold increase in serum creatinine in the TEA group. While the exact mechanism of this favourable response remains unclear, a variety of underlying causes of acute renal failure following cardiac surgery have been identified, including low cardiac output leading to acute tubular necrosis, multiple blood transfusions, renovascular emboli, preoperative contrast agents, nephrotoxic drugs, and prolonged CPB (Bahar et al., 2005; Finkelmeier, 1995). TSA may reduce the pathophysiological surgical stress response and the consequent release of intrarenal angiotensin II and subsequent afferent arteriolar constriction. This effect would improve renal blood flow and result in renal preservation. It is also possible that the more stable hemodynamic course of TSA is protective and provides improved blood flow to organs, including the kidneys.
Previous research has identified the risk of neuraxial hematoma as one of the barriers to epidural or intrathecal injection for cardiac surgery anesthesia and analgesia (Kowalewski, MacAdams, Eagle, Archer, & Bharadwaj, 1994; Lee et al., 2003; Scott et al., 2001). In the current study, none of the TSA patients were specifically investigated for, or diagnosed with the complication of a neuraxial hematoma. This is an important finding, as it lends support for the safety of the TSA technique and subsequent larger scale studies on carefully selected patients.
The findings of significantly reduced postoperative morphine requirements and reduced duration of endotracheal intubation in the TSA group are consistent with previous research. For example, in a recent study, Jacobsohn and associates (2005) found that IT morphine (which is a component of TSA) provided pain management and deep breathing benefits to cardiac surgery patients. Based on their research, Cheng et al. (1996) concluded that early extubation reduced ICU costs by 53% and total costs per CABG patient by 25%. Furthermore, elective surgery cancellations were significantly reduced. This evidence implies that TSA may benefit the patient, as well as the over-taxed Canadian health care system.
SICU LOS was not statistically different between the two groups. It has been postulated that postoperative care of surgical patients has not kept pace with emerging intraoperative techniques that enhance perioperative physiological stability (Lee & Jacobsohn, 2000). We speculate that the reason stable cardiac patients were not transferred out of the SICU earlier was because this was not the current hospital unit practice, as opposed to them not being ready for transfer. Delayed transfer to the ward may have been confounded by factors that are not related to the patient's readiness for transfer, such as a shortage of beds, nursing staff shortages, and delayed discharges and subsequent housekeeping and room preparation. Knowledge of the human response to TSA anesthesia would allow critical care nurses to anticipate patient care needs, enhance patient care, and to optimize patient outcomes.
Although not statistically significant, the mean postoperative hospital LOS was three days less in the TSA group than the GA group. This trend may have been buffered by the routine practices of the unit, such as postoperative monitoring in the step-down unit for 24 hours or more. Hence, the TSA patients may have been discharged based on unit routines rather than individual assessments regarding readiness for discharge. Based on these findings, a power analysis was conducted; the calculations indicated that a sample size of 161 pairs would be required to establish the statistical significance of 2.8 days difference. Therefore, further research with a larger sample is necessary to provide more convincing evidence in this regard.
In summary, the small sample size was a limitation of this study. However, matched sampling was effective in controlling for numerous person factors and strengthened the ability to make group comparisons related to pathophysiological and behavioural responses. The retrospective research design necessitated reliance on data that were previously recorded, and was, therefore, subject to incomplete data sets and recording error. However, this pilot study provides substantive evidence to support the hypothesis that TSA has the potential to enhance recovery from cardiac surgery.
This pilot study has implications for future directions in nursing education, practice, and research. Ongoing learning is paramount to improving management of critical care patients. The findings of this study provide evidence that can be used by the bedside nurse to plan nursing care for the TSA patient. Thus, critical care nurses can use these results to optimize patient outcomes through the prevention, early detection, and management of the surgical stress response. Nurse managers must collaborate with practice leaders to make changes necessary to maximize the financial benefits of new technologies such as TSA. Nurses also need to take a leadership role in the evaluation of technologies that affect nursing practice and their ability to influence patient outcomes. Qualitative research methods could provide insight into the "lived experience" of the TSA patient. Prospective, randomized, controlled trials should use larger samples, as well as evaluate how TSA enhances outcomes.
This pilot study provides initial evidence to substantiate the hypothesis that there are significant positive differences in outcomes between patients receiving TSA and GA for cardiac surgery. These findings have important implications for critical care nurses. In order to ensure evidence-based care of cardiac surgery patients, nurses must continue to play an active role in research related to the rapidly changing face of perioperative management. Moreover, through comprehensive evaluation, dissemination, and incorporation of research findings, nurses can influence individualized patient care and the broader health care system and, thus, move health care forward into the future.
The authors acknowledge funding support for this project from the Saint Boniface General Hospital and Research Foundation.
Aun, C., Thomas, D., St. John-Jones, L., Colvin, M.P., Savege, T.M., & Lewis, C.T. (1985). Intrathecal morphine in cardiac surgery. European Journal of Anesthesiology, 2, 419-426.
Bahar, I.,Akgul,A., Ozatik,A., Vural, K., Demirbag, A.E., Boran, M., et al. (2005). Acute renal failure following open-heart surgery: Risk factors and prognosis. Perfusion, 20, 317-322.
Blomberg, S., Emanuelsson, H., Kvist, H., Lamm, C., Ponten, J., Waagstein, F., et al. (1990). Effects of thoracic epidural anesthesia on coronary arteries in patients with coronary artery disease. Anesthesiology, 73, 840-847.
Blomberg, S., Emanuelsson, H., & Ricksten, S.E. (1989). Thoracic epidural anesthesia and central hemodynamics in patients with unstable angina pectoris. Anesthesia and Analgesia, 69, 558-562.
Boulanger, A., Perreault, S., Choiniere, M., Prieto, I., Lavoie, C., & Laflamme, C. (2002). Intrathecal morphine after cardiac surgery. The Annals of Pharmacotherapy, 36, 1337-1343.
Chaney, M.A., Furry, P.A., Fluder, E.M., & Slogoff, S. (1997). Intrathecal morphine for coronary artery bypass grafting and early extubation. Cardiovascular Anesthesia, 84, 241-248.
Cheng, D.C.H., Karski, J., Peniston, C., Raveendran, G., Asokumar, B., Carroll, J., et al. (1996). Early tracheal extubation after coronary artery bypass graft surgery reduces costs and improves resource use: A prospective, randomized, controlled trial. Anesthesiology, 85, 1300-1310.
Finkelmeier, B. (1995). Complications of cardiac operations. In B. Finkelmeir (Ed.), Cardiothoracic surgical nursing, (pp. 289-302). Chicago: Lippincott.
Heitkemper, M., & Shaver, J. (1989). Nursing research opportunities in enteral nutrition. Nursing Clinics of North America, 24, 415-426.
Hoar, P.F., Stone, J.G., Falatas, A.N., Bendixen, H.H., Head, R.J., & Berkowitz, B.A. (1980). Hemodynamic and adrenergic responses to anesthesia and operation for myocardial revascularization. Journal of Thoracic and Cardiovascular Surgery, 80, 242-248.
Jacobsohn, E., Lee, T.W.R., Amadeo, R.J., Syslak, P.H., Debrouwere, R.G., & Bell, D. (2005). Low-dose intrathecal morphine does not delay early extubation after cardiac surgery. Canadian Journal of Anesthesia, 52, 848-857.
Klassen, G.A., Bramwell, R.S., Bromage, P.R., & ZborowskaSluis, D.T. (1980). Effect of acute sympathectomy by epidural anesthesia on canine coronary circulation. Anesthesiology, 52, 8-15.
Kock, M., Blomberg, S., Emmanuelsson, H., Lomsky, M, Stromblad, S.O., & Ricksten, S.E. (1990). Thoracic epidural anesthesia improves global and regional left ventricular function during stress-induced myocardial ischemia inpatients with coronary artery disease. Anesthesia and Analgesia, 71, 625-630.
Kowalewski, R.J., MacAdams, C.L., Eagle, C.J., Archer, D.P., & Bharadwaj, B. (1994). Anesthesia for coronary artery bypass surgery supplemented with subarachnoid bupivicaine and morphine: A report of 18 cases. Canadian Journal of Anesthesiology, 41, 1189-1195.
Lee, T.W.R., Grocott, H.P., Schwinn, D., & Jacobsohn, E. (2003). High spinal anesthesia for cardiac surgery. Anesthesiology, 98, 499-510.
Lee, T.W.R., & Jacobsohn, E. (2008). Spinal anesthesia in cardiac surgery. Techniques in Regional Anesthesia and Pain Management, 12, 54-56.
Lee, T.W.R., & Jacobsohn, E. (2000). Pro: Tracheal extubation should occur routinely in the operating room following CABG surgery. Journal of Cardiothoracic Vascular Anesthesia, 14, 603-610.
Liem, T.H., Booij, L.H.D.J., Hasenbos, M.A.W.M., & Gilien, M.J.M. (1992a). Coronary artery bypass grafting using two different techniques: Part 1: Hemodynamic results. Journal of Cardiothoracic Vascular Anesthesia, 6, 148-155.
Liem, T.H., Booij, L.H.D.J., Hasenbos, M.A.W.M., & Gilien, M.J.M. (1992 b). Coronary artery bypass grafting using two different techniques: Part 2: Postoperative outcome. Journal of Cardiothoracic Vascular Anesthesia, 6, 156-161.
Liem, T.H., Williams, J.P., Hensens, A.G., & Singh, S.K. (1998). Minimally invasive direct coronary artery bypass procedure using a high thoracic epidural plus general anesthetic technique. Journal of Cardiothoracic Vascular Anesthesia, 12, 668-672.
Loick, H.M., Schmidt, C., Van Aken, H., Junker, R., Erren, M., Berendes, E., et al. (1999). High thoracic epidural anesthesia, but not clonidine, attenuates the perioperative stress response via sympatholysis and reduces troponin T in patients undergoing coronary artery bypass grafting. Anesthesia and Analgesia, 88, 701-709.
Lubenow, T.R., Ivankovich, A.D., & McCarthy, R.J. (2001). Management of acute postoperative pain. In P.G. Barash, B.F. Cullen, & R.K. Stoeiting (Eds.), Clinical Anesthesia (4th ed., pp. 1403-1430). Philadelphia: Lippincott Williams & Wilkins.
Mangano, D.T., Siliciano, D., Hollenberg, M., Leung, J.M., Browner, W.S., Gochner, P., et al. (1992). Postoperative myocardial ischemia: Therapeutic trials using intensive analgesia following surgery. Anesthesiology, 76, 342-353.
McEwen, B., & Mendelson, S. (Eds.). (1993). Effects of stress on the neurochemistry and morphology of the brain: Counterregulation versus damage (2nd ed.). New York: Free Press.
Mitchell, P.H., Gallucci, B., & Fought, S.G. (1991). Perspectives on human response to health and illness. Nursing Outlook, 39(4), 154-157.
Scott, N.B., Turfrey, D.J., Ray, D.A., Nzewi, O., Sucliffe, N.P., Lal, A.B., et al. (2001). A prospective randomized study of the potential benefits of thoracic epidural anesthesia and analgesia in patients undergoing coronary artery bypass grafting. Anesthesia and Analgesia, 93, 528-535.
Selye, H. (1946). The general adaptation syndrome and the diseases of adaptation. Journal of Clinical Endocrinology, 6(2), 117-185.
Vanstrum, G.S., Bjornson, K.M., & Ilko, R. (1988). Postoperative effects of intrathecal morphine in coronary artery bypass surgery. Anesthesia and Analgesia, 67, 261-272.
By Susan Mertin, RN, MN, Jo-Ann V. Sawatzky, RN, PhD, William L. Diehl-Jones, BSc, MSc, PhD, RN, and Dr. TrevorW.R. Lee, BSc, BScMed, MD, MRCA, FRCPC, FASE, FRSM
Susan Mertin, RN, MN, Nurse Clinician, Cardiac Sciences Program of Manitoba, St. Boniface General Hospital, Winnipeg, MB, Nurse Practitioner Student, Athabasca University.
Jo-Ann V. Sawatzky, RN, PhD, Associate Professor, Faculty of Nursing, University of Manitoba.
William L. Diehl-Jones, BSc, MSc, PhD, RN, Associate Professor, Faculty of Nursing & Faculty of Science, University of Manitoba.
Dr. Trevor W.R. Lee, BSc, BScMed, MD, MRCA, FRCPC, FASE, FRSM, Assistant Professor of Anesthesia at the University of Manitoba and Head of the Department of Anesthesia and Perioperative Medicine at St. Boniface General Hospital in Winnipeg, Manitoba.
Table One. Matched baseline person factors (categorical descriptors) (n=70) TSA GA Characteristic n=35 n=35 (%) n (%) n (%) Males 31 (88.6) 30 (85.7) Females 4 (11.4) 5 (14.3) # of grafts [less than or equal to] 2 12 (34.3) 12 (34.3) 3 14 (40.0) 11 (31.4) 4 9 (25.7) 8 (22.8) [greater than or equal to] 5 0 (0.0) 4 (11.4) Radial artery harvest 4 (11.4) 5 (14.7) Aortic valve replace 4 (11.4) 4 (11.4) Mural valve replace 3 (8.6) 3 (8.6) Mural annuloplasy 1 (2.9) 1 (2.9) Redo surgery 1 (2.9) 1 (2.9) Characteristic Probability (%) p-value Sig. Males .72 N/S Females .73 N/S # of grafts [less than or equal to] 2 .53 * N/S 3 4 [greater than or equal to] 5 Radial artery harvest .73 N/S Aortic valve replace 1.00 N/S Mural valve replace 1.00 N/S Mural annuloplasy 1.00 N/S Redo surgery .50 N/S Note. Fisher's Exact Test; * Chi-Square value for group= 4.419 df= 0.22 Table Two. Comparison of baseline person factors (continuous variables) (n=70) TSA GA Characteristic ([bar.x]) SD ([bar.x]) SD Age (years) 64.4 10.8 64.4 10.3 Weight (kg) 85.5 13.9 87.3 14.6 Height (cm) 167.5 19.3 171.1 7.3 Body mass index ([m.sup.2]) 33.3 2.3 29.8 4.9 Cross clamp (minutes) 78.7 40.1 79.0 33.9 Cardiopulmonary bypass (minutes) 114.9 50.6 118.9 44.9 Characteristic f-value df p value Age (years) 0.0 34 .947 Weight (kg) 0.28 34 .601 Height (cm) 1.21 34 .279 Body mass index ([m.sup.2]) 0.77 34 .387 Cross clamp (minutes) 0.0 34 .966 Cardiopulmonary bypass (minutes) 0.27 34 .069 Note. ANOVA Table Three. Comparison of baseline person factors (comorbidities and outcome determinants) (n=70) * TSA GA n=35 n=35 Characteristic (%) n (%) n (%) Smoker Never 15 (42.8) 11 (33.3) Previous 16 (45.7) 18 (54.5) Current 4 (11.4) 3 (9.0) Hypertension 24 (68.5) 25 (71.4) Diabetes 11 (31.4) 7 (20.0) PVD 4 (11.8) 2 (5.7) Asthma 1 (2.9) 0 (0.0) COPD 3 (8.6) 3 (8.6) Other lung disease 2 (5.7) 0 (0.0) CVA/TIA 0 (0.0) 3 (8.6) Renal disease 8 (22.9) 7 (20.0) CHF history 8 (22.9) 7 (20.0) CHF current 3 (8.5) 2 (5.7) Preop MI [less than or equal to] 3 months 10 (28.6) 12 (34.3) 3-6 months 0 (0.0) 3 (8.6) 6-12 months 2 (5.7) 1 (2.9) Class IV angina 7 (20.0) 8 (27.6) Ejection fraction >60% 17 (48.6) 18 (51.4) 40-60% 10 (28.6) 13 (37.1) 21-39% 7 (20.0) 4 (11.4) <20% 1 (2.9) 0 (0.0) Chronic atrial fib. 3 (8.6) 3 (8.6) Medications [beta] Blocker 26 (74.3) 23 (67.6) Digoxin 2 (5.7) 0 (0.0) Ca Channel blocker 7 (20.0) 7 (20.0) Lipid lowering agents 20 (57.1) 24 (68.6) Oral Hypoglycemic 9 (25.7) 4 (11.4) Insulin 1 (2.9) 0 (0.0) ACE inhibitors 18 (51.4) 20 (57.1) Angiotensin II blocker 2 (5.7) 2 (5.7) Diuretics 13 (37.1) 11 (31.4) Characteristic (%) p-value Sig. Smoker (1) .69 * N/S Never Previous Current Hypertension 1.00 N/S Diabetes 0.41 N/S PVD .42 N/S Asthma 1.00 N/S COPD 1.00 N/S Other lung disease .49 N/S CVA/TIA .23 N/S Renal disease 1.00 N/S CHF history 1.00 N/S CHF current .64 N/S Preop MI (2) .34 * N/S [less than or equal to] 3 months 3-6 months 6-12 months Class IV angina Ejection fraction (3) .53 * N/S >60% 40-60% 21-39% <20% Chronic atrial fib. Medications [beta] Blocker .60 N/S Digoxin .49 N/S Ca Channel blocker 1.00 N/S Lipid lowering agents .45 N/S Oral Hypoglycemic .22 N/S Insulin 1.00 N/S ACE inhibitors .81 N/S Angiotensin II blocker 1.00 N/S Diuretics .17 N/S Note. Fisher's Exact Test * Chi-square value for group (1) Smoker, df=2, Chi-square= .743, (2) Pre op MI, df=3, Chi-square= 3.896, (3) Ejection fraction, df= 3, Chi-square= 2.238 Table Four. Pathophysiological responses: TSA versus GA (n=70) TSA (n) % GA (n) Postoperative MI 0 (0.0) 1 New onset atrial fibrillation 11 (31.4) 13 Re-operation for bleeding 1 (2.8) 6 Required blood product transfusion 11 (31.4) 8 Vasopressor support SICU 17 (48.6) 23 Inotrope support SICU 3 (8.6) 7 Sternal dehiscence 0 (0.0) 0 Renal failure requiring dialysis 1 (2.8) 0 Serum creatinine > 50% [up arrow]in preop value 2 (5.7) 6 Respiratory failure requiring reintubation 1 (2.8) 2 Significant pleural effusion 5 (14.3) 5 Cerebral vascular accident/TIA 0 (0.0) 0 Postoperative confusion 2 (5.7) 3 Incisional leg infection 0 (0.0) 1 In-hospital death 0 (0.0) 1 Neuraxial hematoma 0 (0.0) 0 % p-value Postoperative MI (2.8) 1.0 New onset atrial fibrillation (37.1) .801 Re-operation for bleeding (17.1) .055 Required blood product transfusion (22.8) .592 Vasopressor support SICU (65.7) .227 Inotrope support SICU (20.0) .306 Sternal dehiscence (0.0) 1.0 Renal failure requiring dialysis (0.0) 1.0 Serum creatinine > 50% [up arrow]in preop value (17.1) .151 Respiratory failure requiring reintubation (5.7) 1.0 Significant pleural effusion (14.3) 1.0 Cerebral vascular accident/TIA (0.0) 1.0 Postoperative confusion (8.6) 1.0 Incisional leg infection (2.8) 1.0 In-hospital death (2.8) 1.0 Neuraxial hematoma (0.0) 1.0 Note. Fisher's exact test Table Five. Behavioural responses: TSA versus GA IV morphine use (n = 60) * Total IV Morphine use (total mg) Time (hours TSA GA postop) ([bar.x]) (SD) ([bar.x]) SD 0-6 3.10 3.94 10.91 4.70 6-12 0.83 1.68 5.21 3.90 12-18 1.60 2.59 3.70 2.05 18-24 1.16 1.83 2.90 3.36 24-30 1.43 2.13 2.36 2.39 30-36 0.96 1.51 1.33 2.12 36-42 0.41 0.85 1.13 1.79 42-48 0.40 1.00 1.16 1.66 Total 9.91 9.31 28.73 11.32 Time (hours postop) f-value df p value 0-6 55.65 29 < .0001 6-12 30.68 29 < .0001 12-18 12.76 29 .001 18-24 8.22 29 .008 24-30 2.37 29 .134 30-36 0.57 29 .458 36-42 3.64 29 .066 42-48 4.76 29 .037 Total 50.74 29 < .0001 Note. * ANOVA