PAIN MEDICINE

SECTION EDITOR SPENCER S. LIU

The Use of Intrathecal Morphine for Postoperative Pain Relief After Liver Resection: A Comparison with Epidural Analgesia
Lesley De Pietri, MD*, Antonio Siniscalchi, MD*, Alexia Reggiani, MD*, Michele Masetti, Bruno Begliomini, MD*, Matteo Gazzi, MD*, Giorgio E. Gerunda, MD, and Alberto Pasetto, MD*
MD,

*Division of Anesthesiology and Liver and Multivisceral Transplant Center, University of Modena and Reggio Emilia, Italy

An epidural catheter is used in some institutions for postoperative analgesia after liver surgery. However, anesthesiologists may not feel comfortable leaving a catheter in the epidural space because of concern about coagulation disturbances and possible bleeding complications caused by impaired liver function. In this study, we tested a single-shot intrathecal morphine technique and compared it to a continuous epidural naropine infusion for postoperative analgesia in liver surgery. Fifty patients were randomly assigned to an epidural analgesia group (EP group; n 25) and an intrathecal analgesia group (IN group; n 25). The quality of analgesia assessed by a visual analog scale (VAS), the side effects, and the additional IV analgesic requirements were recorded. We did not observe any signs of cord compression. Time to first pain drug requirement was longer in the EP group compared to the IN group (25 18.5 h versus 12 10.3 h; P 0.05). In

both groups, the VAS remained less than 30 mm throughout the 48-h follow-up period. Consumption of IV morphine with a patient-controlled analgesia device in the IN group was larger (mostly from 24 to 48 h after surgery) than the EP group (12.0 5.54 mg versus 3.1 2.6 mg, respectively; P 0.01). The incidence of vomiting was 4% in both groups, whereas the incidence of pruritus (16% versus 0%) and nausea (16% versus 4%) was more frequent in the IN group. No postdural puncture headache and no spinal hematoma occurred. After liver resection, a single dose of intrathecal morphine followed by patient-controlled morphine analgesia can provide satisfactory postoperative pain relief. The quality of this treatment, according to the VAS, is not inferior to continuous epidural analgesia up to 48 h after surgery. (Anesth Analg 2006;102:115763)

iver resection is performed with increasing frequency as a treatment for primary neoplasms, which are frequently related to liver cirrhosis, or to remove metastatic lesions (1). Early patient mobilization is very important in the short-term outcome, but it can be painful, and a good level of analgesia is therefore required. Continuous epidural analgesia is an accepted technique in major abdominal surgery because it provides the required level of analgesia to support mobilization and significant reduction in pulmonary and cardiovascular morbidity in the early postoperative period (2).
Accepted for publication November 9, 2005. Address correspondence and reprint requests to Lesley De Pietri, c/o Division of Anesthesiology, University of Modena and Reggio Emilia, # 71 via del Pozzo, 41100 Modena, Italy. Address e-mail to lesley.depietri@tin.it. DOI: 10.1213/01.ane.0000198567.85040.ce
2006 by the International Anesthesia Research Society 0003-2999/06

Postoperative coagulation disturbances related to liver surgery, even in patients with normal preoperative coagulation undergoing uncomplicated hepatectomy (3), raise concerns about the safety of postoperative analgesia administered through an epidural catheter. Coagulation changes after liver surgery and the possible increased risk of bleeding complications, including spinal hematoma (4,5), have limited the use of epidural analgesia. The extent of the liver resection may affect the magnitude and duration of the postoperative coagulation disturbances and makes the proper timing of epidural catheter removal important (6,7). A single dose of intrathecal morphine, administered immediately before surgery, can be a useful alternative method to obtain safe and prolonged postoperative analgesia when an epidural catheter is contraindicated. This prospective, randomized study compared, after liver resection and over a 48-h follow-up period, analgesia and side effects of a single
Anesth Analg 2006;102:115763

1157

1158

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

ANESTH ANALG 2006;102:115763

dose of intrathecal morphine combined with morphine patient-controlled analgesia (PCA), with a continuous epidural naropine infusion supported by morphine PCA. The aim of this study was to evaluate the noninferiority of analgesia (monitored by a visual analog scale [VAS]) of the intrathecal technique compared with the continuous epidural infusion.

Methods
We enrolled 50 adult patients of both sexes, ASA I and II, scheduled for liver surgery, in a prospective, randomized fashion. Approval by the local hospital ethics committee and informed written patient consent was obtained. Patients underwent minor (uni- and bisegmentectomies and nonanatomical) and major (three or more segments) resections. Blood tests, including platelet count, International Normalized Ratio (INR), activated partial thromboplastin time (aPTT), and fibrinogen were determined before surgery, immediately after surgery, and daily until values returned to within the normal range. Patients with bleeding diathesis (essential thrombocythemia, idiopathic thrombocytopenic purpura, von Willebrand disease, and hemophilia A and B), neurological dysfunction (multiple sclerosis, subacute myelo-opticoneuropathy or preexisting lower limb neurological deficit), significant respiratory diseases, recent systemic or local infections, history of drug use, or treated with opioids because of chronic pain were excluded from the study. According to institutional policy, we introduced or removed an epidural catheter when INR was 1.5, aPTT was 1.24, and platelet count were more than 100 109/L. We treated our patients according to the recommendations based on a small patient series by Schumann et al. (8), although American Society of Regional Anesthesia Guidelines on Coumadin do not apply to the patients on whom we performed a neuroaxial block because our patients postoperative coagulation derangement may have had multiple etiologies. Patients were randomly allocated to two groups by a computer-generated list: the epidural analgesia group (EP group; n 25) and the intrathecal analgesia group (IN group; n 25). All patients had an IV line connected to a PCA device with IV morphine to improve quality of analgesia. A VAS was used for patient pain assessment. All patients were premedicated with diazepam 0.1 mg/kg orally 1 h before the induction of anesthesia. They were monitored with electrocardiogram, pulse oximetry, capnography, inspired and expired oxygen, anesthetic vapor concentration, systemic (radial) arterial blood pressure, and central venous pressure through the right jugular vein.

In the EP group, an epidural catheter was inserted at the T9-10 or T10-11 level using the loss-of-resistance technique after local anesthesia. Lidocaine 2% 2 mL, plus 2 mL after 5 min, was used to exclude the subarachnoid placement of the catheter. Ropivacaine 0.2% 6 8 mL and morphine 2 mg were injected before the induction of anesthesia. In the IN group, dural puncture was performed at the L3-4 or L4-5 level with a 27-gauge Whitacre spinal needle after local anesthesia. After return of clear, free-flowing cerebrospinal fluid, morphine 0.2 mg in 0.9% saline solution (2.5 mL) was injected. Thiopental sodium 35 mg/kg, fentanyl 23 g/kg IV, droperidol 2.5 mg, and vecuronium 0.1 mg/kg were used for the induction of anesthesia and tracheal intubation. Anesthesia was maintained with desflurane 4%5% in a 50% oxygen-air mixture with minimal flow ventilation, and muscle relaxation was obtained with vecuronium 0.05 mg/kg. Whenever the arterial blood pressure or heart rate increased by more than 30% in the IN group, that depth of anesthesia was judged inadequate and hemodynamic control was ensured with incremental doses of IV fentanyl (50 100 g). Intraoperative analgesia in the EP group was provided with continuous epidural infusion of ropivacaine 0.2% (57 mL/h). A decrease in systolic arterial blood pressure by more than 30% less than the preoperative value was corrected with 500 mL of colloid, with IV ephedrine 5 mg, or with both. Decreases in heart rate to less than 50 bpm were treated with IV atropine 0.25 mg. Total duration of the operation, intraoperative consumption of analgesics, and the time of tracheal extubation were also recorded. The weight of the resected specimen was measured by weighing scales after liver resection, and correlations among weight of resected liver, platelet count, INR, aPTT, and fibrinogen were also investigated. After surgery, patients were admitted to the intensive care unit (ICU) for overnight observation or returned to the ward, according to the preoperative condition, the extent of the liver resection, and the occurrence of intraoperative critical events, such as major bleeding, cardiovascular imbalance, and respiratory dysfunction. In the EP group, epidural analgesia was continued after surgery with an epidural infusion of ropivacaine 0.2% 57 mL/h. In both groups, patients were allowed to receive 1-mg boluses of morphine from an IV PCA pump, with a lockout time of 5 min and a 4-h dose limit of 20 mg. No background infusion of morphine was used. Quality of analgesia was reported by the patient as a pain score on a graded-scale VAS (0 100 mm). Patients with no pain relief or insufficient analgesia (VAS 40) at 30 min after tracheal extubation were excluded from the study, and another analgesic technique was implemented. Patients were taken to the ward or to

ANESTH ANALG 2006;102:115763

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

1159

the ICU when their Aldrete score was 10 and arterial blood gas values were normal (Paco2 50; pH 7.357.45). All patients received oxygen via a facemask with 4 L/min of O2 for at least 4 h after surgery. Noninvasive arterial blood pressure, heart rate, respiratory rate, and pain scores were recorded at 4, 8, 12, 24, 36, and 48 h after surgery using VAS at rest and when coughing. Sedation was assessed on a 5-point scale (I completely awake with the eyes open; II drowsy, closed eyes; III asleep but responding to verbal commands; IV asleep but responding to touch or pain; V not responding). The postoperative clinical monitoring of the patients and the evaluation of VAS were managed by investigators blinded to the analgesic technique used. Four hours after surgery, arterial blood gas evaluation was performed to detect any possible respiratory depression (Paco2 50 mm Hg or respiratory rate 8 breaths/min). Naloxone 0.1 mg IV was administered if the patient was drowsy, respiratory rate was less than 8 breaths/min, or Pco2 50 mm Hg. Patients were excluded from the study if they required sedatives or opioids other then IV PCA morphine. Morphine consumption was recorded in all patients at 4, 8, 12, 24, 36, and 48 h after surgery. Duration of intrathecal or epidural analgesia was defined as the time from patient randomization in the operating room to the time of the first delivery of a morphine bolus from the PCA device. Side effects such as nausea, vomiting, psychomimetic effects, neurological disorders, pruritus, sedation, respiratory depression, and hypotension were evaluated. Patients were evaluated for postdural puncture headache and radicular back pain; muscle weakness and sensory deficit, as early signs of spinal cord compression caused by hematoma, were also evaluated. Where appropriate, results were presented as mean ( sd). Patient characteristics, duration of surgery, intraoperative fentanyl, time to extubation, time to discharge to the ward, and postoperative morphine consumption were analyzed by unpaired Students t-test. The correlation between variables was assessed by linear regression analysis. The Mann-Whitney rank sum test was used to analyze VAS and sedation score. P 0.05 was considered significant. The primary aim of this study was the comparison between intrathecal and epidural analgesia after liver resection, according to VAS. Before starting the study, an estimate for expected standard deviation was entered in the calculation for power analysis and sample size. According to GraphPad StatMate (GraphPad Software, San Diego, CA), a sample size of 25 patients in each group has a 80% power to detect a difference between (VAS) mean values of 2.43 (not less) with a significance level () of 0.05. With NCSS, Number Cruncher Statistical

System, and PASS, Power Analysis and Sample Size Program, (NCSS, Kaysville, UT), a poststudy power analysis was computed, and a noninferiority hypothesis of intrathecal morphine versus epidural naropine was tested using the Mann-Whitney nonparametric adjustment for a uniform distribution. To test noninferiority or equivalence, it is required to define the margin of equivalence (the largest difference that is not of practical significance), and a value of 10 mm was chosen because statistically significant changes of 10 mm in VAS scores are also of clinical significance in a variety of psychophysiological measurements in children (9), adults, and elderly subjects, whereas small differences between VAS scores may be statistically significant but clinically meaningless (10). Therefore, it was accepted that a VAS difference of more than 10 mm between the 2 treatments (mean values) could eliminate the hypothesis of noninferiority of equivalence of intrathecal morphine versus epidural naropine.

Results
Patient age, body weight, surgical indication, and ASA status (I and II) were comparable between the two groups. Types and duration of liver resection were similar in both groups, and no patient was excluded from the study. Tuohy needle insertion for epidural catheter placement required multiple attempts in 4 (16%) patients compared to 2 (8%) patients when dural puncture was performed. No bleeding was observed during needle or catheter placement. Fourteen patients (28%) were admitted to the ICU: six (24%) from the IN group and eight (32%) from the EP group (P not significant). The postoperative mean INR value was 17.4% higher than the mean preoperative value after minor liver resections and 30% higher after major liver resections (P 0.05). The daily mean changes in INR value with respect to the mean preoperative value were significant up to postoperative Day (POD) 5 for both types of liver resection (Fig. 1A). The mean INR never reached 1.3 after minor liver resections. INR was highest on POD1 in all patients (P 0.05) at 1.49 0.26 (close to the accepted threshold value for epidural catheter removal; INR 1.5) after major resections and at 1.02 0.14 after minor resections. The epidural catheter was never removed before POD4 (when the mean INR value was less than 1.3) after major liver resections. The mean platelet count was less than the mean preoperative level up to POD5 (P 0.05) and was smallest on POD3 both after minor and major liver resection (Fig. 1B). Mean fibrinogen levels and mean aPTT values were in the normal range throughout the observation period, and no significant changes were observed.

1160

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

ANESTH ANALG 2006;102:115763

Figure 1. (A) Changes in mean International Normalized Ratio (INR) in major (three or more segments) and minor (uni- and bi-segmentectomies and nonanatomical) resections. Values are expressed as mean sd; *P 0.05 compared with the preoperative value. (B) Changes in platelet count in major and minor sections. Values are expressed as mean sd; no comparison with the preoperative value was significant. PREOP preoperative; POSTOP postoperative; POD postoperative day.

Mean INR value and weight of the resected liver parenchyma were positively correlated on POD1 (r 0.52156) (Fig. 2), but such a correlation was not found with the weight range of the minor liver resections. A significantly larger mean dose of fentanyl was used in the in group compared with the Ep group. Tracheal extubation time and recovery room observation time were significantly shorter in the EP group than in the IN group (Table 1). Moderate to severe pain (VAS 30 mm) was never observed either at rest or when coughing during the first 48 h after surgery in both groups (Fig. 3, A and B). The time to first morphine requirement was significantly longer in EP-group patients. PCA morphine consumption was similar in both groups up to 12 h after surgery; there were significantly more PCA morphine boluses in the IN group (Table 2). PCA morphine was required by 14 of 25 (56%) patients in the EP group and 23 of 25 (92%) patients in the IN group (P 0.05). The mean total amount of morphine used during the first 48 h was larger in the IN group than in the EP group (P 0.01) (Table 2). Three patients in the EP group (12%) required increased fluid support to treat hypotension. Five of 25 patients (20%) in the IN group had a sedation score of III after tracheal extubation. No differences in sedation were observed between the 2 groups beyond 30 min after tracheal extubation. No hypercapnia or hypoxia were observed. Vomiting was 4% in both groups, pruritus was 0% in the EP group and 16% in the IN group (P 0.05), and nausea was 4% in the EP group and 16% in the IN group (P 0.05). No patient had postdural puncture headache, and no spinal hematoma occurred (95% confidence interval, 0%13.71%). According to the poststudy power analysis and the test of noninferiority performed with the NCSS-PASS software package, group sample sizes of 25 and 25 achieved a power 95% at 12, 24, and 48 h, at rest and when coughing, to detect noninferiority of VAS in the IN group versus the EP group, choosing 10 mm to

Figure 2. Correlation between mean International Normalized Ratio (INR) on postoperative Day 1 (POD1) and weight of resected liver (grams).

determine a margin of equivalence (10 mm true difference between the means) and using a one-sided Mann-Whitney test, assuming that the actual distribution is uniform. The significance level () of the test was 0.025. These results can be presented in another way saying that the 95% confidence interval of the treatment effect (the true differences between the means) at rest and when coughing, at 12, 24, and 48 h, lies inside the margin of equivalence.

Discussion
The use of an epidural catheter in patients undergoing liver resection remains controversial. Because of early postoperative changes in the coagulation profile (7), epidural analgesia may not be considered a firstchoice technique for pain relief in liver surgery (11). Liver resection causes coagulation abnormalities caused by decreased levels of clotting factors. Prolongation of prothrombin time is therefore a common finding after major hepatic resection (12). The possibility of a bloody tap from needle or catheter placement or continuing trauma related to an epidural catheter has been described, and it may result in spinal bleeding (13). Schumann et al. (8) showed a transient postoperative coagulopathy after resection in donors for living related liver transplantation, with the highest value of INR recorded on POD1. Matot et al. (13) showed how the extent of liver resection may affect the magnitude and duration of postoperative coagulation disturbances and, therefore, the proper timing of epidural catheter removal. In agreement with Schumann et al. (8) we observed profound and prolonged INR disturbances in patients who had surgery for major liver resection; mean INR was highest on POD1, and a value close to the preoperative value was observed on POD5. We must be aware that the risk of

ANESTH ANALG 2006;102:115763

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

1161

Table 1. Intraoperative and Postoperative Data Presented as Mean (sd) Epidural group (n 25) Intraoperative fentanyl (g kg h ) Extubation time (min) Observation time in recovery room (min)
1 1

Intrathecal group (n 25) 1.95 (0.74) 24.25 (8.6) 40 (5.6)

P-value 0.05 0.05 0.01

0.5 (0.25) 18.13 (7.9) 35 (4.5)

Figure 3. (A) Pain at rest (mean sd). (B) Pain when coughing (mean sd). Both graphs show measurement of pain by a 100-mm visual analog scale (VAS) in the recovery room (RR) and during the first 48 h. No significant differences were observed between groups.

spinal hematoma is increased as long as the coagulation abnormalities persist. PCA with IV morphine and epidural PCA, with epidural local anesthetic and opioids, are two major advances in management of pain after major abdominal surgery (14). Nevertheless, the possibility of using an alternative technique to the epidural route, which guarantees a relatively long period of analgesia while avoiding the placement of an epidural catheter, seems interesting. Intrathecal morphine has become popular in recent years; its analgesic effect has been proven after most types of surgery (15). From 0.1 to 0.2 mg of intrathecal morphine is regarded as the optimal analgesic amount in laparoscopic abdominal and pelvic surgery, whereas it is recommended to avoid a larger dose because side effects increase and quality of analgesia does not (16). However, Devys et al. (17) used an intrathecal dose of 0.3 0.4 mg of morphine to obtain good postoperative analgesia in abdominal surgical procedures not including liver resections. We chose a smaller dose of intrathecal morphine because large reductions of liver parenchyma may decrease drug clearance and disturb drug metabolism (18). Abnormal drug metabolism may lead to excessive sedation, respiratory depression (19,20), and possibly hepatotoxicity1. This study shows that a small single dose of intrathecal morphine, supported by IV PCA morphine, can provide good postoperative analgesia for the first 48 hours after liver resection. Patients treated
1 Goudas LC, Maszczynska-Bonney I, Schumann R, et al. Morphine induces hepatic and renal oxidative stress through a mapkinase signal transduction pathway [abstract]. Liver Transpl 2001; 7:C-88, #350 of Program Abstract for the Joint Meeting of ILTS, ELTA and LICAGE Berlin 2001.

with intrathecal morphine had low VAS scores and rarely required any IV PCA morphine in the first 12 hours after surgery. VAS scores on POD2 were low in both groups, but the mean value of the total IV PCA morphine consumption was significantly higher in the IN group compared to the EP group. The infrequent incidence of nausea and vomiting in our study could be explained by preoperative use of droperidol. Nausea and vomiting are the main side effects of intrathecal morphine, reported by other authors (21) in approximately 50% of patients treated with spinal opioids. However, we observed less vomiting. Nausea and pruritus were more frequent in the IN group patients, who had intrathecal morphine and used more IV PCA morphine. We did not use intrathecal local anesthetic in addition to morphine in order to avoid the wide sympathetic block related to the high subarachnoid anesthesia required for a subcostal incision, and also because the intraoperative analgesic benefit of adding bupivacaine intrathecally was not demonstrated by Motamed et al. (22). The larger intraoperative administration of opioids in the IN group may be explained by the slow onset of the morphine action (17) and by the absence of the preemptive analgesic component provided by the anesthetic used in the EP group. The larger intraoperative amount of opioids may also explain the prolonged time to tracheal extubation and the higher sedation scores observed. The anesthesiologist was not blinded to the type of analgesia administered during surgery, and a bias may have been introduced. Although higher sedation scores were reached in 12% of patients from the IN group 30 minutes after removal of the tracheal tube, the time in the recovery room was not prolonged. Respiratory rate and Paco2 were not different between groups. All patients received oxygen 35% in air for a few hours in the early postoperative period. Respiratory depression was not observed. A few studies demonstrated that the combined use of intrathecal and IV morphine is safe, and respiratory depression rarely occurs (19,20). Cirrhotic patients with compromised liver function or patients who may develop postoperative coagulopathy after a liver resection could benefit from the intrathecal administration of morphine, which avoids the need of an epidural catheter and the risk of its removal. In almost all patients who had surgery for

1162

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

ANESTH ANALG 2006;102:115763

Table 2. Cumulative PCA Use of Morphine 4h EP group IN group 0.4 (1.15) 1.6* (2.3) 8h 0.7 (1.7) 2.5* (3.1) 12 h 1.3 (2.1) 3.4* (3.9) 24 h 2.3 (2.5) 7.2 (3.6) 36 h 2.4 (2.6) 10.2 (5.4) 48 h 3.1 (3.2) 12.1 (5.5)

Values are mean (sd). EP epidural; IN intrathecal. *P 0.05; P 0.01.

major liver resection, an epidural catheter was in situ during the time of coagulation dysfunction. Large clinical trials, which have not been performed (14), are required to guide anesthesiologists in the choice of epidural analgesia or intrathecal morphine. Because complications caused by spinal bleeding are so unusual (1:150,000 and 1:220,000 in epidural anesthesia and intrathecal anesthesia, respectively) (6), it is not an easy task to quantify the risk in patients who need a liver resection. Obviously, the design of this study and the sample size do not allow evaluation of the safety of these techniques. Careful clinical and laboratory monitoring and frequent neurologic testing performed after dural puncture for intrathecal morphine, or after insertion and removal of an epidural catheter, are mandatory in this patient population (23). This study does not try to prove that one treatment is better than another but, rather, aims to show that any difference is of no practical consequence. The test of noninferiority, a subset of the general approach to equivalence statistical analysis, was chosen because to run a standard statistical test and to conclude that the two treatments are equivalent, if the difference is not statistically significant, is an invalid approach and leads to invalid conclusions (24). The question is not whether the two treatments lead to different outcomes (there will always be some difference when two treatments are used) but whether the outcomes differ enough to be clinically or scientifically relevant. Therefore, to test equivalence or noninferiority, we had to define, based on literature and clinical experience, a range of treatment effects (mean VAS) that may be considered scientifically or clinically trivial (a 10-mm margin of equivalence). This decision is not about statistics and must be made on scientific or clinical grounds. In conclusion, intrathecal morphine gives effective control of postoperative pain after liver surgery. It can be used as a valid alternative to continuous epidural analgesia, supported by IV PCA morphine, when the insertion or the removal of an epidural catheter may be considered unsafe because of actual or potential coagulopathy. Further studies with large-scale trials are required to assess the safety of the intrathecal technique compared with continuous epidural analgesia in liver surgery.

References
1. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:74550. 2. Kehlet H, Holte K. Effect of postoperative analgesia on surgical outcome. Br J Anaesth 2001;87:6272. 3. Pelton JJ, Hoffman JP. Comparison of liver function test after hepatic lobectomy and hepatic wedge resection. Ann Surg 1998; 64:40814. 4. Vandermeulen EP, Van Aken H. Anticoagulant and spinalepidural anesthesia. Anesth Analg 1994;79:116577. 5. Horlocher TT, Wedel DJ, Benzon H, et al. Regional anesthesia in the anticoagulated patient: defining the risks (The second ASRA Consensus Conference on Neuroaxial Anesthesia and Anticoagulation). Reg Anesth Pain Med 2003;28:17297. 6. Gunningham JD, Fong Y, Shriver C, et al. One hundred consecutive hepatic resections: blood loss, transfusion, and operative technique. Arch Surg 1994;129:10506. 7. Borromeo CJ, Stix MS, Lally A, et al. Epidural catheter and increased prothrombin time after right lobe hepatectomy for living donor liver transplantation. Anesth Analg 2000;91: 113941. 8. Schumann R, Zabala L, Angelis M, et al. Altered hematologic profiles following donor right hepatectomy and implications for postoperative analgesic management. Liver Transpl 2004;10: 3638. 9. Powell CV, Kelly AM, Williams A. Determining the minimum clinically significant difference in visual analog pain score for children. Ann Emerg Med 2001;37:2831. 10. Bodian CA, Freedman G, Hossain S, et al. The visual analog scale for pain: clinical significance in postoperative patients. Anesthesiology 2001;95:135661. 11. Oguro A, Taniguchi H, Didoh T, et al. Factors relating to coagulation, fibrinolysis and hepatic damage after liver resection. HPB Surg 1993;7:439. 12. Meijer C, Wiezer MJ, Hack CE, et al. Coagulopathy following major liver resection: the effect of rBP121 and the role of decreased synthesis of regulating proteins by the liver. Shock 2001;15:26171. 13. Matot I, Scheinin O, Eid A, Jurim O. Epidural anesthesia and analgesia in liver resection. Anesth Analg 2002;95:117981. 14. Liu SS, Carpenter RL, Neal JM. Epidural anesthesia and analgesia: their role in postoperative outcome. Anesthesiology 1995;82:1474506. 15. Gwirtz KH, Young JV, Byers RS, et al. The safety and efficacy of intrathecal opioid analgesia for acute postoperative pain: seven years experience with 5969 surgical patients at Indiana University Hospital. Anesth Analg 1999;88:599604. 16. Bailey PL, Rhondeau S, Schafer PG, et al. Dose response pharmacology of intrathecal morphine in human volunteers. Anesthesiology 1993;79:4959. 17. Devys JM, Mora A, Jayr C, et al. Intrathecal PCA morphine improves analgesia during the first 24 hr after major abdominal surgery compared to PCA alone. Can J Anaesth 2003;50:35561. 18. Gomez-Lechon MJ, Pensoda X, Jover R, et al. Hepatotoxicity of the opioids morphine, heroin, meperidine and methadone to cultured human hepatocytes. Mol Toxicol 1987 88;4:45363. 19. Cole PJ, Craske DA, Wheatley RG. Efficacy and respiratory effects of low-dose spinal morphine for postoperative analgesia following knee arthroscopy. Br J Anaesth 2000;85:2337.

ANESTH ANALG 2006;102:115763

PAIN MEDICINE DE PIETRI ET AL. TREATMENT OF POSTOPERATIVE PAIN

1163

20. Bailey PL, Lu JK, Pace NL. The effects of intrathecal morphine on the ventilatory response to hypoxia. N Engl J Med 2000;343: 122834. 21. Grattidge P. Nausea and vomiting after major arthroplasty with spinal anesthesia including morphine: a randomized trial of subhypnotic propofol infusion as a prophylaxis. Acta Anaesthesiol Scand 1998;42:1247.

22. Motamed C, Bouaziz H, Franco D, et al. Analgesic effect of low-dose intrathecal morphine and bupivacaine in laparoscopic cholecystectomy. Anaesthesia 2000;55:11824. 23. Horlocker TT, Wedel DJ. Neurologic complications of spinal and epidural anesthesia. Reg Anesth Pain Med 2000;25:8398. 24. Wellek S. Testing statistical hypotheses of equivalence. Chapman & Hall /CRC, 2003.