The median sternotomy incision and the mediastinal tube insertion sites are major contributing factors for moderate-to-severe pain during the early recovery phase following cardiac surgery [1,2]. Effective postoperative pain control is an essential component of the care of the cardiac surgical patient, as greater pain intensity after surgery has been associated with a reduced capacity to cough, reduced mobility, increased frequency of atelectasis and prolonged recovery [2,3]. Additionally, adequate pain control after cardiac surgery has an important role in maintaining stable haemodynamics, low myocardial oxygen consumption and freedom from ischaemic events [4,5].
A wide variety of techniques including intravenous (i.v.) patient-controlled or nurse-controlled opioid analgesia, local or regional analgesia and multimodal analgesia have been used to obtain adequate pain relief after cardiac surgery. Among these techniques, intrathecal and epidural analgesia reliably induce enhanced postoperative pain relief, yet may be associated with the risk of epidural haematoma formation in patients who are anticoagulated perioperatively . Currently, parenteral opioids remain the mainstay of postoperative pain management in many institutions for cardiac surgery . Although the use of opioids can provide intense analgesia, the doses necessary to provide effective pain relief may lead to undesirable side-effects such as sedation, respiratory depression, nausea and vomiting . Therefore, instead of placing full reliance on a single method, it may be more convenient to use other analgesic methods in combination with parenteral opioids after cardiac surgery .
Infiltration of the surgical wounds with local anaesthetics has been shown to improve pain scores or reduce opioid consumption after various types of minor and major surgery [8-13]. Anecdotal reports at our institution have indicated that infiltration of the median sternotomy incision and the mediastinal tube sites with local anaesthetics in combination with parenteral opioids is associated with good-quality analgesia after cardiac surgery. Therefore, this prospective, randomized and double-blind study was planned to investigate the effect of 0.25% levobupivacaine infiltration of the median sternotomy wound and the mediastinal tube sites in addition to morphine patient-controlled analgesia (PCA) on postoperative pain, morphine consumption, sedation and nausea and vomiting when compared with morphine PCA alone in patients undergoing elective coronary artery bypass grafting (CABG) surgery.
Fifty patients aged 18-65 yr scheduled for elective CABG with cardiopulmonary bypass (CPB) were included in this study, which was prospectively carried out with permission of the Human Ethics Board of the hospital. Informed consent was obtained from all patients after the nature of the study was explained. The exclusion criteria from the study were as follows: emergency surgery, previous CABG or heart valve surgery, combined valve and CABG surgery, poor left ventricular function (ejection fraction < 40%) or congestive heart failure, preoperative use of inotropic drugs or intra-aortic balloon pump, pre-existing pulmonary or neuromuscular disease, coagulopathy, allergy to local anaesthetics or morphine, opioid or benzodiazepine tolerance and any condition that creates inability to use the PCA device. The patients who required prolonged CPB times (>120 min), continued inotropic support (dopamine >5 μg kg−1 min−1 or epinephrine infusion) after weaning from CPB and patients with uncorrected coagulopathy were excluded from the study before randomization into the study groups. Postoperative exclusion criteria included haemodynamic instability including the occurrence of serious arrhythmias, bleeding, surgical re-exploration and any condition that created inability of the patient to properly use the PCA pump. Haemodynamic instability was defined as systolic blood pressure less than 90 mmHg with dopamine (>5 μg kg−1 min−1) or epinephrine infusion. The patients were randomized into the study groups based on a computer-generated randomization scheme. The study drugs were prepared by a nurse according to the computer-generated randomization scheme; the investigators and surgeons remained blinded.
All patients included in this study were informed about the visual analogue scale (VAS) and the use of a PCA device during the pre-anaesthetic examination. All routine cardiac medications were continued until the morning of surgery. All patients were premedicated with peroral 5-10 mg diazepam and intramuscular 0.1 mg kg−1 morphine sulphate. At the operating theatre, one 16-G i.v. catheter and a 20-G radial arterial cannula were inserted. The patients were continuously monitored with DII and V5 leads of electrocardiogram (ECG) and ST segment analysis (Horizon XL, Mennen Medical Inc.). After monitoring ECG and invasive arterial pressure, anaesthesia was induced with 5 μg kg−1 fentanyl, 0.3 mg kg−1 etomidate, 1 mg kg−1 lidocaine and 0.1 mg kg−1 vecuronium, followed by ventilation via a face mask with 2-4% sevoflurane in 50% oxygen-50% air. The patients were intubated and mechanically ventilated targeting an end-tidal CO2 partial pressure of 32-35 kPa (Ventilator 710; Siemens, Sweden). Anaesthesia was maintained with 5-10 μg kg−1 h−1 fentanyl infusion and 1-2% sevoflurane in 50% oxygen-50% air. A triple-lumen central venous catheter (Certofix Duo, B. Braun Melsungen A.G.) was introduced into the right internal jugular vein. Central venous pressures, central and rectal body temperatures and urine output were also continuously monitored. Fluid replacement was established using crystalloids in line with the haemodynamic data. Additional vecuronium was given at a dose of 0.05 mg kg−1 for muscular relaxation as required.
Surgery was performed in a standard fashion through a median sternotomy incision with saphenous veins and internal thoracic arteries harvested as conduits. A standard crystalloid prime was used in the CPB circuit. CPB was established via standard aortic and single venous cannulation using a Sarns modified roller pump (Sarns, Ann Arbor, MI, USA). During CPB, oxygenation was achieved with a membrane oxygenator (D 708 Simplex, Dideco, Mirandola, Italy) and sevoflurane was administered via the oxygenator at a minimum of 1.0 MAC (minimum alveolar concentration). Myocardial protection was achieved with intermittent cold blood cardioplegia. During CPB, the hematocrit was maintained between 20% and 25%, the non-pulsatile pump flow between 2.0 and 2.5 L min−1 m−2, and mean arterial pressures (MAP) between 50 and 65 mmHg. Body temperature was maintained between 28°C and 30°C during CPB and active rewarming to 37°C was established before aortic cross-clamp removal. The patients were randomized into two groups just before sternal wire placement at the end of surgery: sternotomy and mediastinal tube sites were infiltrated with either 60 mL of 0.25% levobupivacaine (infiltration group, n = 25) or 60 mL of saline placebo (control group, n = 25). The study solution was infiltrated as follows: a total of 30 mL of the sternum: 15 mL on each side, and a total of 30 mL of the study solution was infiltrated deeply around the mediastinal tubes: 15 mL on each side. At the end of surgery, sevoflurane was discontinued and a propofol infusion started at 0.5-1 mg kg−1 h−1 and all patients were given morphine 2 mg i.v.
The patients were transferred to the ICU and mechanically ventilated (Servo 900 D, Siemens, Uppsala, Sweden). Heart rate (HR), arterial pressure, oxygen saturation, central venous pressure and ventilatory frequency were monitored continuously. The intensivist working in the ICU was blinded to the study protocol and made decisions regarding the administration of postoperative inotropes, vasodilators and/or vasoconstrictors, as well as chest tube removal and extubation. The patients were initially managed with volume-controlled mechanical ventilation until rewarming and recovery from anaesthesia was complete. As soon as the patients were able to breath spontaneously at a respiratory rate of 10-25 breaths min−1, the ventilator was set to synchronized intermittent mandatory ventilation combined with inspiratory pressure support (respiratory rate 12 breaths min−1 and pressure support of 20 mmHg). The propofol infusion was stopped once the patient was warm (core temperature > 36°C), awake, co-operative, haemodynamically stable and achieved good clinical neuromuscular recovery. If the initial arterial blood gas values were normal, the intermittent mandatory ventilation rate was reduced by 2min−1 every half hour until 4 min−1. Simultaneously, pressure support was reduced to 10 mmHg. The patients were extubated if they met the following criteria: patient awake and responsive to command, fully warmed with core temperature >36°C, haemodynamically stable without significant dysrhythmias, well-perfused with adequate urine output (>1.0 mL kg−1 h−1), mediastinal bleeding < 100 mL h−1 for 2 h and able to maintain PaO2/FiO2 > 200 at peak end-expiratory pressure levels < 5 cm H2O, PaCO2 between 35 and 45 mmHg and pH > 7.35 on pressure support ventilation of 10 mmHg with fractional inspired oxygen concentration ≤0.5, for half an hour.
All patients were supplied with a PCA pump set to deliver i.v. morphine boluses in 2 mg increments (bolus dose: 2 mg, lock-out time: 15 min, 4 h limit: 20 mg) during the initial 24 h after extubation. An anaesthesiologist blinded to group assignment recorded pain and sedation scores. Postoperative pain was assessed using a VAS (0 = no pain, 10 = the worst pain imaginable) both at rest and on coughing at postoperative 1, 2, 3, 4, 8, 12, 18 and 24 h. Morphine (mg) delivered via the PCA pump, MAP, HR, respiratory rate, oxygen saturation and arterial blood gas (PaO2 and PaCO2) values were also simultaneously recorded. All patients were given additional i.v. morphine in 2 mg boluses when they required rescue analgesia beyond their PCA lockout. The sedation level was evaluated at postoperative 1, 2, 4, 8, 12 and 24 h using a six-point scale described by Ramsay and colleagues (1 = anxious and agitated or restless or both; 2 = co-operative, oriented, tranquil; 3 = responds to commands only; 4 = brisk response to a light glabellar tap or loud auditory stimulus; 5 = sluggish response to a light glabellar tap or loud auditory stimulus; 6 = no response to light glabellar tap or loud auditory stimulus) . Supplementary morphine and any opioid-related side-effects (such as nausea, vomiting, cough suppression, constipation, urinary retention and pruritus) were recorded. Postoperative nausea and vomiting were treated with ondansetron, 4 mg i.v., when needed. All patients were interviewed at 24 h after discharge from the ICU and were asked to classify the quality of their initial postoperative analgesia as excellent, good, satisfactory or poor.
The primary end-point of the study was defined as a reduction in postoperative pain scores, a decrease in VAS score of 2 being regarded as clinically significant . The sample size was determined by power analysis; with a power of 0.9 and significance level of 0.05, 22 subjects per study group were required. To allow a 20% dropout, 28 patients per group were included. Statistical analysis of the data from the study was performed with the SPSS (SPSS for Windows Release 14.0) statistical package. A one-sample Kolmogorov-Smirnov test was used to determine the normal distribution of numerical data. Normally distributed numerical data were statistically analysed using the t-test on independent samples. The χ2-test was used to analyse gender, patient satisfaction with pain relief and number of patients requiring additional postoperative analgesic. The U-test was used to analyse VAS scores, sedation scores and cumulative morphine requirements between groups. The statistical analysis of VAS scores and sedation scores within groups was performed using the Friedman test and Wilcoxon signed rank sum test. Haemodynamic and respiratory data were analysed using analysis of variance (ANOVA) for repeated measurements. The results are presented as mean ± SD and P < 0.05 was considered statistically significant.
Fifty-six patients were initially enrolled in the study. Three patients in each group were excluded because of haemodynamic instability or the need for postoperative revision for haemostasis. The patient groups were comparable in terms of age, body weight, gender distribution, ejection fraction and duration of surgery, cross-clamp and CPB (Table 1).
VAS scores at rest and with cough decreased within each group over time during the assessment period, but there were no differences between the two study groups (Figs 1 and 2). Cumulative morphine consumption via the PCA pump increased within each group over time and morphine use was significantly lower in the infiltration group compared with the control group at all assessment times during the 24 h following extubation (P < 0.05) (Table 2). The number of patients requiring rescue morphine and total amount of rescue morphine consumption were significantly lower in the infiltration group compared with the control group (22 vs. 10 patients and 4.2 ± 2.7 vs. 1.2 ± 1.7 mg, respectively) (P < 0.05). The total consumption of morphine - including morphine use via the PCA pump and rescue morphine use during the first 24 h following extubation - was significantly lower in the infiltration group compared with the control group (29.5 ± 5.1 vs. 42.8 ± 4.7 mg, respectively) (P < 0.05).
The mean sedation scores at 1, 2 and 4 h after extubation were significantly higher in the control group when compared with the infiltration group (2.7 ± 0.5 vs. 2.2 ± 0.5, P = 0.000; 2.6 ± 0.5 vs. 2.1 ± 0.3, P = 0.001 and 2.5 ± 0.5 vs. 2.1 ± 0.3, P = 0.002, respectively), but mean sedation scores at 8, 12 and 24 h after extubation were similar between groups (Table 3). The sedation scores within the infiltration group were similar over time during the assessment period. The sedation scores within the control group were significantly lower at 8, 12 and 24 h after extubation compared with sedation scores at the end of the first postoperative hour (P < 0.05) (Table 3). Postoperative respiratory rates, arterial blood gas and SPO2 values were similar between groups; none of the patients required reintubation and neither excessive sedation nor significant respiratory depression was encountered.
The postoperative course was uneventful for all of the patients included in the study; no myocardial infarctions, arrhythmias, re-explorations for bleeding or infections occurred. There were no significant differences in MAP and HR values between groups, all values remaining within the normal range throughout the study period. The durations of postoperative propofol infusion were comparable between groups: 119.2 ± 34.0 vs. 114 ± 23.5 min in the control and infiltration groups, respectively. The times between the end of propofol infusion and extubation were > 120 min in four of the patients in the control group (maximum: 195 min) and five of the patients in the infiltration group (maximum: 150 min). The times to extubation were similar between groups: 211.4 ± 55.0 vs. 207.6 ± 55.4 min in the control and infiltration groups, respectively. There were no significant differences between groups in the incidence of postoperative nausea and vomiting; seven patients (three in the infiltration group, and four in the control group) experienced nausea and received a single 4 mg i.v. dose of ondansetron. There were no episodes of vomiting, pruritus or other opioid-related side-effects. Patient satisfaction with postoperative analgesia was found to be similar in both groups; the number of patients classifying the quality of their initial postoperative analgesia as excellent, good, satisfactory or poor were similar in the infiltration and control groups: 15/9/1/0 and 16/8/1/0, respectively.
Pain after cardiac surgery is less than that after upper abdominal surgery and lateral thoracotomy, since median sternotomy does not entail dividing muscles . However, studies on the subject show that moderate pain affects 43% and severe pain 34% of patients undergoing cardiac surgery depending on postoperative pain management [17,18]. The location and the intensity of postoperative pain were evaluated in a previous study including 30 adult patients submitted to cardiac surgery using sternotomy . The level of pain was greatest on the first postoperative day, and moderate pain was sited in the region of the sternotomy until the fifth postoperative day, passing to the leg associated with the saphenectomy thereafter.
The present study aimed to investigate any possible beneficial effects of local anaesthetic infiltration of the sternotomy wound and the mediastinal tube sites as an adjunct to postoperative opioid analgesia in selected patients undergoing CABG surgery. The choice of levobupivacaine, the pure S-enantiomer of bupivacaine administered via infiltration, was made on the basis of its long duration of effect and less cardiovascular toxicity when compared with the R-enantiomer or racemic mixture . Infiltration of the median sternotomy incision and the mediastinal tube insertion sites with 0.25% levobupivacaine in addition to morphine PCA provided effective postoperative analgesia when compared with morphine PCA alone, as indicated by a significant decrease (31%) in postoperative morphine consumption at 24 h in the infiltration group for comparable pain scores between groups (29.5 ± 5.1 vs. 42.8 ± 4.7 mg, respectively, P < 0.05). The reporting of similar pain scores by patients in both groups indicates that patients have titrated PCA morphine to bring their pain to tolerable levels. The median VAS-rest scores of 4.0 in both infiltration and control groups at extubation compare favourably with a previously reported median pain score of 4.0 after cardiac surgery, when i.v. morphine analgesia was administered by nursing staff . The median VAS-rest scores of 3.0 in both study groups at postoperative 2 h and thereafter show that good-quality analgesia was achieved after cardiac surgery. PCA handsets were successfully used by all of the patients included in the study, provided that the PCA instructions were repeated at intervals by the nursing staff. The mean sedation scores at postoperative 1, 2 and 4 h after extubation were significantly higher in the control group when compared with the infiltration group, but neither excessive sedation nor significant respiratory depression was encountered. It is interesting that sedation scores after postoperative 4 h were comparable between control and infiltration groups, despite continued use of much greater amounts of morphine by the control group. Perhaps, this study was not powered enough to analyse postoperative sedation score differences between the two groups, as investigating postoperative VAS scores and morphine consumption were the primary aims of this study. The incidences of opioid-related postoperative nausea and vomiting were found to be small and similar between groups; this may be attributed to the antiemetic effect of propofol which was used for sedation in both groups during the early postoperative period. Postoperative interviews with our patients at 24 h after extubation showed that overall satisfaction with pain relief was high and similar in both groups; 64%/32% of patients in the infiltration group and 60%/36% of patients in the control group classified the quality of their initial postoperative analgesia as excellent/good.
Infiltration of surgical wounds has been reported to enhance postoperative analgesia after various procedures such as upper abdominal surgery , open cholecystectomy , abdominal hysterectomy , caesarean section  and inguinal herniotomy . There are also reports of negative findings on the analgesic or opioid-sparing effect of wound infiltration after different surgical procedures . The conflicting results in the various pre- and postoperative infiltration studies might be due to heterogeneous infiltration techniques and different doses and volumes of local anaesthetic leading to variable degrees of afferent block. Prospective studies investigating the effect of wound infiltration in cardiac surgical patients are scarce and give conflicting results [22,23]. The efficacy of intraoperative bupivacaine 0.5% infiltration (10 mL) of the median sternotomy wound, followed by continuous subcutaneous infusion (120 mL 24 h−1) was investigated in 47 patients undergoing cardiac surgery . Bupivacaine infiltration, followed by continuous subcutaneous infusion for 36 h, did not seem to reduce postoperative pain or to shorten assisted ventilation time. The lower portion of the wound was found to be uncovered by the local anaesthetic due to the characteristics of the catheter used for infusion, which had a few holes only at the tip . Another reason for inadequate pain relief in the study described above could be that pain relief was provided only at the level of the wound. This may have been only partly effective in cardiac surgical patients, as mediastinal drains are a cause for major discomfort after cardiac surgery . Another study including 17 patients undergoing cardiac surgery evaluated the effects of a parasternal block combined with local anaesthetic infiltration of the sternotomy wound and tube insertion sites using 54 mL of 0.25% levobupivacaine on postoperative pain and respiratory function over 24 h . The results of the study demonstrated reduced morphine consumption in the infiltration group during the first 4 h after surgery (20.8 ± 6.2 vs. 33.2 ± 10.9 mg, respectively), but when analysed across the 24-h time period by repeated measures ANOVA, morphine use was found to be significantly less in the infiltration group compared with the placebo group for 24 h after surgery (P = 0.02). The need for rescue medication was also reduced in the infiltration group when compared with placebo. Postoperative pain and sedation scores were not significantly different between groups, despite a trend towards better sedation scores in the treatment group. The investigators reported that the small number of patients in the study, and thus large SDs, did not allow significance of sedation scores to be revealed . The arterial blood gases indicated oxygenation parameters were better in the infiltration group at the time of extubation, but bedside spirometry values were similar between the infiltration and the control groups throughout the study period. Additionally, peak serum levobupivacaine concentrations were found to be below potentially toxic levels for all patients included in the study .
In this study, the median sternotomy incision and the mediastinal tube insertion sites were infiltrated using a standard technique and the maximum recommended dose of local anaesthetic (60 mL of 0.25% levobupivacaine) for all patients. The local anaesthetic solution was infiltrated before sternal wire placement at the end of surgery to provide as profound an afferent block as possible. The surgeons performing the infiltration anaesthesia were careful to verify negative aspiration during local anaesthetic injections, as the safety of infiltration anaesthesia relies on avoiding inadvertent intravascular injections. In a previous study comparing 110 rehospitalized CABG patients to 224 CABG patients not rehospitalized, the majority (19%) of readmissions within 30 days were found to be related to wound infection . Other studies investigating complications after cardiac surgery reported that heart failure and wound infections were the two most frequently occurring postoperative problems . Potential complications of infiltration anaesthesia including compromised healing or wound infection were not encountered in the present study, but we believe that larger sample sizes and longer follow-up periods are needed to identify postoperative complications related to the wound infiltration technique.
The primary end-points of this study were pain scores and cumulative morphine consumption after 0.25% levobupivacaine infiltration of the median sternotomy incision and the mediastinal tube sites when compared to placebo after cardiac surgery. Secondary end-points were the incidences of side-effects in both study groups. We had considered it impractical to include respiratory function tests in the study protocol, as we aimed to keep patients in a calm yet co-operative state. All patients included in the study were closely monitored using arterial blood gases, respiratory rates and oxygen saturation values throughout the study period. It may be argued that the evaluation of VAS scores after cardiac surgery is influenced by the cerebral side-effects of the surgical procedure, as a variety of central nervous system abnormalities ranging from subtle cognitive decline to stroke have been reported after CABG surgery . Cerebral microemboli, global cerebral hypoperfusion, systemic inflammatory response and genetic susceptibility have been thought to play a role in the aetiology of cognitive dysfunction after cardiac and non-cardiac surgery . Postoperative cognitive dysfunction, which may complicate early recovery after CABG in significant numbers of cardiac surgical patients is often associated with abnormal neurological signs . However, it was confirmed that all of the patients included in the present study had good co-operation and orientation during the early postoperative period, without any abnormal neurological signs. In addition, patient characteristics such as age, ejection fraction, CPB and cross-clamp times - which may be important risk factors for postoperative cognitive impairment - were comparable between both study groups. Another limitation to the present study may be that only a select group of cardiac surgical patients - only those with good ventricular function undergoing CABG for the first time and no serious comorbidity - were included. Therefore, it may be argued that our study group does not resemble the general cardiac surgical population encountered in routine practice. However, the results of the present study show importance, as statistical significance was reached for one of our primary end-points, cumulative morphine use.
In conclusion, infiltration of the median sternotomy incision and the mediastinal tube insertion sites with 0.25% levobupivacaine in addition to morphine PCA was found to be effective in reducing morphine consumption when compared with placebo during the first 24 h after cardiac surgery. The described infiltration anaesthesia technique also resulted in improved sedation scores during the early postoperative period after extubation. Further studies may be needed to investigate the benefits and the side-effects associated with this technique in cardiac surgical patients.
1. Mueller XM, Tinguely F, Tevaearai HT, Revelly JP, Chiolero R, von Segesser LK. Pain location, distribution, and intensity after cardiac surgery. Chest
2000; 118: 391-396.
2. Lahtinen P, Kokki H, Hynynen M. Pain after cardiac surgery. Anesthesiology
2006; 105: 794-800.
3. O'Connor CJ. Pain relief and pulmonary morbidity after cardiac surgery. Crit Care Med
1999; 27: 2314-2316.
4. Chaney MA. How important is postoperative pain after cardiac surgery? J Cardiothorac Vasc Anesth
2005; 19: 705-707.
5. Mangano DT, Siliciano D, Hollenberg M et al.
Postoperative myocardial ischemia. Therapeutic trials using intensive analgesia following surgery. Anesthesiology
1992; 76: 342-353.
6. Power I. Recent advances in postoperative pain therapy. Br J Anaesth
2005; 95: 43-51.
7. Reimer-Kent J. From theory to practice: preventing pain after cardiac surgery. Am J Crit Care
2003; 12: 136-143.
8. Bartholdy J, Sperling K, Ibsen M, Eliasen K, Mogensen T. Preoperative infiltration of the surgical area enhances postoperative analgesia of a combined low-dose epidural bupivacaine and morphine regimen after upper abdominal surgery. Acta Anaesthesiol Scand
1994; 38: 262-265.
9. Johansson B, Glise H, Hallerbäck B et al.
Preoperative local infiltration with ropivacaine for postoperative pain relief after cholecystectomy. Anesth Analg
1994; 78: 210-214.
10. Hannibal K, Galatius H, Hansen A, Obel E, Ejlersen E. Preoperative wound infiltration with bupivacaine reduces early and late opioid requirement after hysterectomy. Anesth Analg
1996; 83: 376-381.
11. Giacomazzi CM, Lagni VB, Monteiro MB. Postoperative pain as a contributor to pulmonary function impairment in patients submitted to heart surgery. Braz J Cardiovasc Surg
2006; 21: 386-392.
12. Dierking GW, Ostergaard E, Ostergard HT, Dahl JB. The effects of wound infiltration with bupivacaine versus saline on postoperative pain and opioid requirements after herniorrhaphy. Acta Anaesthesiol Scand
1994; 38: 289-292.
13. Kakagia D, Fotiadis S, Tripsiannis G, Tsoutsos D. Postoperative analgesic effect of locally infiltrated levobupivacaine in Fleur-de-Lys abdominoplasty. Aesthetic Plast Surg
2007; 31: 128-132.
14. Ramsay ME, Savege TM, Simpson BJ, Goodwin R. Controlled sedation with alphaxalone-alphadolone. Br Med J
1974; 2: 656-659.
15. Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Defining the clinically important difference in pain outcome measures. Pain
2000; 88: 287-294.
16. Weissman C. Pulmonary function after cardiac and thoracic surgery. Anesth Analg
1999; 88: 1272-1279.
17. Watt-Watson J, Stevens B, Katz J, Costello J, Reid GJ, David T. Impact of preoperative education on pain outcomes after coronary artery bypass graft surgery. Pain
2004; 109: 73-85.
18. Walther T, Falk V, Metz S et al.
Pain and quality of life after minimally invasive versus conventional cardiac surgery. Ann Thorac Surg
1999; 67: 1643-1647.
19. McLeod GA, Burke D. Levobupivacaine. Anaesthesia
2001; 56: 331-341.
20. Nay PG, Elliott SM, Harrop-Griffiths AW. Postoperative pain. Expectation and experience after coronary artery bypass grafting. Anaesthesia
1996; 51: 741-743.
21. Klein JR, Heaton JP, Thompson JP, Cotton BR, Davidson AC, Smith G. Infiltration of the abdominal wall with local anaesthetic after total abdominal hysterectomy has no opioid-sparing effect. Br J Anaesth
2000; 84: 248-249.
22. Magnano D, Montalbano R, Lamarra M et al.
Ineffectiveness of local wound anesthesia to reduce postoperative pain after median sternotomy. J Card Surg
2005; 20: 314-318.
23. McDonald SB, Jacobsohn E, Kopacz DJ et al.
Parasternal block and local anesthetic infiltration with levobupivacaine after cardiac surgery with desflurane: the effect on postoperative pain, pulmonary function, and tracheal extubation times. Anesth Analg
2005; 100: 25-32.
24. Mueller XM, Tinguely F, Tevaearai HT, Ravussin P, Stumpe F, von Segesser LK. Impact of duration of chest tube drainage on pain after cardiac surgery. Eur J Cardiothorac Surg
2000; 18: 570-574.
25. Beggs VL, Birkemeyer NJ, Nugent WC, Dacey LJ, O'Connor GT. Factors related to rehospitalization within thirty days of discharge after coronary artery bypass grafting. Best Pract Benchmarking Healthc
1996; 1: 180-186.
26. Guadagnoli E, Ayanian JZ, Cleary PD. Comparison of patient-reported outcomes after elective coronary artery bypass grafting in patients aged greater than or equal to and less than 65 years. Am J Cardiol
1992; 70: 60-64.
27. Gao L, Taha R, Gauvin D, Othmen LB, Wang Y, Blaise G. Postoperative cognitive dysfunction after cardiac surgery. Chest
2005; 128: 3664-3670.
28. Murkin J. Prognosis: cognitive function at hospital discharge predicts long-term cognitive function after coronary artery bypass surgery. Can J Anesth
2002; 49: 497-499.