PAIN AFTER THORACOTOMY is common and often

Transcription

1 A Randomized, Double-Blind Trial Comparing Continuous Thoracic Epidural Bupivacaine With and Without Opioid in Contrast to a Continuous Paravertebral Infusion of Bupivacaine for Post-thoracotomy Pain Jay S. Grider, DO, PhD,* Timothy W. Mullet, MD, Sibu P. Saha, MD, Michael E. Harned, MD,* and Paul A. Sloan, MD* Objective: To compare the results of continuous epidural bupivacaine analgesia with and without hydromorphone to continuous paravertebral analgesia with bupivcaine in patients with post-thoracotomy pain. Design: A prospective, randomized, double-blinded trial. Setting: A teaching hospital. Participants: Patients at a tertiary care teaching hospital undergoing throracotomy for lung cancer. Interventions: Subjects were assigned randomly to receive a continuous thoracic epidural or paravertebral infusion. Patients in the epidural group were randomized to receive either bupivacaine alone or in combination with hydromorphone. Visual analog scores as well as incentive spirometery results were obtained before and after thoracotomy. Methods and Main Results: Seventy-five consecutive patients presenting for thoracotomy were enrolled in this institutional review board approved study. On the morning of surgery, subjects were randomized to either an epidural group receiving bupvicaine with and without hydromorphone or a paravertebral catheter infused bupvicaine. Postoperative visual analog scores and incentive spirometry data were measured in the postanesthesia care unit, the evening of the first operative day, and daily thereafter until postoperative day 4. Analgesia on all postoperative days was superior in the thoracic epidural group receiving bupiv- PAIN AFTER THORACOTOMY is common and often severe. Adequate postoperative analgesia has been shown to be a positive predictive factor in the postoperative recovery of patients after thoracotomy. 1,2 In addition, the severity of postoperative pain may be a factor in the development of sustained chronic pain after thoracotomy. 1-3 The past two decades have yielded significant improvements in post-thoracotomy pain management, with thoracic epidural analgesia emerging as the gold standard treatment of choice. 1,2,4 Epidural analgesia, although clearly effective for managing postoperative pain after thoracotomy, is not without limitations and contraindications. For instance, patients increasingly are presenting for thoracotomy while being maintained on antiplatelet agents, such as clopidogrel. Also, the failure rate for epidural analgesia or catheter malfunction after proper insertion has been reported to be as high as 12%. 5 Intercostal blocks and intrapleural analgesia have been presented as possible alternatives to epidural analgesia for thoracotomy; however, in a recent review, the efficacy of these modalities has been questioned. 1 Paravertebral block, both single injection and continuous infusion, has been reported to be comparable to thoracic epidural with regard to analgesia while avoiding the possibility of hypotension and urinary retention in the postoperative period. 1,6-13 Despite these advantages, it should be noted that percutaneous paravertebral catheter placement carries the same contraindications with regard to anticoagulation as epidural acaine plus hydromorphone. Analgesia was similar in the epidural and continuous paravertebral groups receiving bupivacaine alone. No significant improvement was noted by combining the continuous infusion of bupivacaine via the paravertebral and epidural routes. Incentive spirometry goals were best achieved in the epidural bupivacaine and hydromorphone group and equal in the group receiving bupivacaine alone either via epidural or continuous paravertebral infusion. Conclusions: The current study provided data that fill gaps in the current literature in 3 important areas. First, this study found that thoracic epidural analgesia (TEA) with bupivacaine and a hydrophilic opioid, hydromorphone, may provide enhanced analgesia over TEA or continuous paravertebral infusion (CPI) with bupivacaine alone. Second, in the bupivacaine-alone group, the increased basal rates required to achieve analgesia resulted in hypotension more frequently than in the bupivacaine/hydromorphone combination group, underscoring the benefit of the synergistic activity. Finally, in agreement with previous retrospective studies, the current data suggest that CPI of local anesthetic appears to provide acceptable analgesia for post-thoracotomy pain Elsevier Inc. All rights reserved. KEY WORDS: epidural analgesia, paravertebral analgesia, epidural hydromorphone, thoracotomy analgesia. 14 The use of elastomeric infusion devices with larger volumes of local anesthetic has been demonstrated in a recent retrospective review comparing patients receiving continuous local anesthetic infusion both subcutaneously in the wound and at the intercostal nerves with patients receiving epidural analgesia. 15 The results of this article suggested that the analgesia provided by bupivacaine via continuous infusion at the pericostal sutures and subcutaneously was superior to epidural analgesia in patients undergoing thoracotomy. The retrospective nature and lack of randomization in this study were recognized as significant study limitations; however, the feasibility of the use of elastomeric infusion devices for continuous medication application was established. 15 The aim of the current study was to compare the analgesic efficacy and ability to participate in pulmonary rehabilitation in patients receiving thoracic epidural analgesia (TEA) with local anesthetic (with and without opioid) with continuous extrapleu- From the *Department of Anesthesiology, Division of Pain Medicine, and Division of Cardiothoracic Surgery, University of Kentucky, Lexington, KY. Address reprint requests to Jay S. Grider, DO, PhD, University of Kentucky College of Medicine, 800 Rose Street, Suite N 201, Lexington, KY jsgrid2@ .uky.edu 2012 Elsevier Inc. All rights reserved / $36.00/0 doi: /j.jvca Journal of Cardiothoracic and Vascular Anesthesia, Vol 26, No 1 (February), 2012: pp

2 84 GRIDER ET AL ral paravertebral block with local anesthetic. A secondary aim was to evaluate the efficacy of epidural hydromorphone for thoracotomy analgesia. Additionally, the incidence of ongoing pain after thoracotomy was evaluated as a potential subject for further investigations. METHODS After institutional review board approval and informed consent, 75 patients presenting for thoracotomy were enrolled in this prospective, randomized, double-blinded clinical trial. Inclusion criteria were American Society of Anesthesiologists class I to III subjects younger than 75 undergoing elective anterolateral thoracotomy by one of two cardiothoracic surgeons. Exclusion criteria included preoperative opioid use for more than one month before surgery. All subjects were made aware that if analgesia was not acceptable at any point during treatment they could opt out of participation in the study. Acknowledged postoperative random event factors that had the potential to alter the study dynamics included the following: reintubation and sedation preventing the evaluation of visual analog scores (VAS) or incentive spirometry (IS), catheter malposition or dislodgement, hypotension requiring a reduction of local anesthetic concentration to 0.125% bupivacaine and opioid-induced somnolence or sedation. When these random events occurred, data were collected up to the time of the postoperative factor and included in results using an intent-to-treat analysis. For the purposes of nomenclature in this study, postoperative day (POD) 1 was considered the day of surgery. Figure 1 represents the randomization scheme, the timing of evaluation, and the timing of random event factors for the study. On the day of surgery, patients were assigned randomly to 1 of 3 groups: epidural bupivacaine opioid (EB O, n 25), epidural bupivacaine alone (EB, n 25), and a paravertebral catheter with bupivacaine (paravertebral block [PB], n 25). The bupivacaine concentration used was 0.25% for all subjects receiving local anesthetic whether via the epidural or paravertebral route. In patients developing hypotension or urinary retention, bupivacaine dose was reduced to 0.125% and data noted as a change in dosage. The epidural opioid used was 0.01 mg/ml of preservative-free hydromorphone. Before epidural placement, preoperative IS was performed. All subjects received a thoracic epidural catheter placed at approximately the T6 level. Intravenous midazolam, 2 mg, was administered for sedation during placement. The skin of the thoracic spine was prepared with Chloraprep (CareFusion Corp, Leawood, KS). Skin was anesthetized with 1% lidocaine, and using a paramedian approach contralateral to the operative side, a 17-G Tuohy needle was placed in the epidural space using the loss-of-resistance technique. Catheters were placed 4 cm beyond the depth of the Tuohy needle and secured with Steri-Strips (3M, St Paul, MN), and a clear sterile dressing was placed. The operating room anesthesia staff was instructed not to use the epidural catheters preoperatively and intraoperatively. General anesthesia was induced using propofol, 2 mg/kg, fentanyl, 3 g/kg, and rocuronium. The maintenance of anesthesia consisted of a balanced anesthetic technique of desflurane or isoflurane titrated to clinical efficacy and up to 100 g of intravenous fentanyl as needed intraoperatively. The paravertebral infusion via the elastomeric OnQ system (iflow Corp, Lake Forest, CA) was placed by the surgeon under direct visualization at the conclusion of the procedure in the manner described by Sabanathan et al. 16 Before chest closure, all subjects also received intercostal nerve blocks above and below the surgical incision with single injections of 2 ml of 0.25% bupivacaine. The paravertebral and epidural catheter infusions were begun immediately upon entry to the postanesthesia care unit (PACU). Epidural analgesics (basal 2 ml/h with 1 ml every 10 minutes via patient-controlled analgesia [PCA]) were 0.25% bupivacaine alone or 0.25% bupivacaine with 0.01 mg/ml of hydromorphone, whereas paravertebral infusions were 0.25% bupivacaine at 8 ml/h. The analgesic modality that the patient was not randomized to was in place to ensure blinding of the personnel caring for the patient; however, this modality was not actively infused. All patients could request an increase in the epidural basal rate of 1 ml and, if requested, would be re-evaluated in one hour. This allowed the patients to request titration of their basal rate to satisfactory analgesia. Additionally, if the subject used more than 4 doses via the epidural PCA, then the basal rate was increased by 1 ml and recorded. Subjects also could receive intravenous opioid PCA should the catheter-based system the patient was randomized to fail or was no longer providing adequate analgesia; however, there was no crossover possible without exiting the study. VAS scores were obtained in the PACU before discharge with and without IS, the evening of POD1, and each morning and evening of POD2 through POD4. Likewise, the VAS score was obtained at rest and with IS, with both spirometery results and pain scores recorded each POD morning (6-8 AM) and evening (4-6 PM). According to the center s thoracotomy protocol, every attempt is made to discontinue chest tubes by the end of POD4. If subjects required continuation of their analgesic therapy beyond the morning of POD4, the VAS and incentive spirometery data after POD4 were not included for standardization purposes. The acute pain personnel obtaining the VAS and IS results were blinded as to the route of infusion as was the patient. Subjects were contacted 12 months after thoracotomy to determine the incidence of chronic postsurgical pain. Groups were evaluated for differences in age, sex, American Society of Anesthesiologists classification, and weight using a 3-way analysis of variance. VAS results were analyzed using the Mann-Whitney U nonparametric test to determine differences among groups and are expressed as mean values with standard deviation. Incentive spirometry values as a measure of postoperative pulmonary function were compared using the Student t test and were simply expressed as the ability to achieve a vital capacity of 2 L. Via an a priori power analysis (power 0.80), it was determined that 25 subjects in each group would be sufficient to detect a difference in the VAS score of greater than 0.5. Subjects with incomplete data secondary to a random event (eg, reintubation, hypotension, and catheter dislodgement) not associated with the primary study objective (ie, analgesia) were retained in the study, and when data points could be collected, these points were included for analysis post hoc. For example, patients experiencing hypotension were able to continue to give VAS and IS scores albeit at a lower local anesthetic concentration. In contrast, patients with respiratory distress who were reintubated and sedated on the ventilator were unable to give accurate VAS or IS results. All statistical analysis was performed using Sigma Stat (Aspire Software International, Ashburn, VA). RESULTS Seventy-five patients were recruited into the study. There were no significant differences for sex, age, weight, or American Society of Anesthesiologists classification among groups (Table 1). Data collection on 6 subjects was interrupted midstudy, 2 in the paravertebral group (respiratory distress and reintubation) and 4 in the epidural groups (EB O: 1 respiratory distress/reintubaton, EB: 2 respiratory distress/reintubation and 1 catheter dislodgement) secondary to postoperative factors unrelated to analgesia satisfaction (Fig 1). Three subjects in the EB group experienced hypotension, necessitating a decrease in bupivacaine concentration by 0.125%. The results from these subjects were included using an intention-to-treat analysis. Preoperative VAS scores were not obtained; however, pain scores on discharge from the PACU were comparable across

3 THORACIC EPIDURAL BUPIVACAINE 85 Fig 1. The randomization scheme, the timing of evaluation, and the timing of random event factors. groups (Table 2). On the evening of POD1 and POD3, VAS pain scores were statistically different in the EB O groups versus the EB or PB groups both at rest and with IS (Fig 2). Combining the VAS at rest data from the evening of POD1 to POD4, there was no difference in mean standard error VAS pain scores between the paravertebral (PB, ) and EB ( ) local anesthetic groups; however, the EB O group provided superior analgesia ( , p 0.05) compared with groups receiving local anesthetic alone (Table 3). Five patients in the paravertebral group required supplementation with intravenous PCA, whereas 3 patients in the EB group required intravenous PCA (Table 3). There was no statistically significant difference in the opioid use among patients using intravenous PCA supplementation. Of the 3 subjects in the EB group requiring IV PCA, 1 of these resulted from catheter dislodgement on POD3 (analgesia before catheter loss was adequate). All subjects were able to perform a 2-L incentive spirometry inspiration before surgery (Table 2). Postoperative incentive spirometry of 2 L was best in the EB O group (87%) compared (p 0.05) with PB (78%) or EB (55%) (Table 3). No patients reported urinary retention with either epidural or paravertebral infusions. Six subjects in the EB O group experienced pruritus treated with 12.5 mg of Benadryl (Parkdale Pharmaceuticals, Rochester, MI) every 4 to 6 hours as needed with satisfactory results. The incidences of hypotension and postsurgical respiratory complications are described in Figure 1. Twenty-six patients evenly distributed across all groups required chest tube maintenance longer than 4 postoperative days. However, there were no significant increases in pain reported in the patients who continued chest tube therapy after POD4 (data not shown). All subjects were offered follow-up of ongoing postoperative pain issues in the authors pain center. At the 12-month followup, 11 of the patients had ongoing chronic pain consisting of 6 cases of intercostal neuralgia reported on follow-up by tele-

4 86 GRIDER ET AL Table 1. Patient Characteristics and Demographic Data from Groups at Randomization EB O(n 25) EB (n 25) PB (n 25) Age (y) 52.6 (46-75) 56.9 (51-75) 54.5 (47-75) Sex (M/F) 14/11 18/7 15/10 Weight (kg) 72.1 (55-109) 68.9 (54-110) 71.3 (56-112) ASA class I/II/III 1/15/8 4/16/0 3/17/2 Abbreviations: M, male; F, female; ASA, American Society of Anesthesiologists. NOTE. There were no significant differences with regard to age, sex, or weight. phone contact or in the authors chronic pain clinic and 5 cases of malignancy-related pain. The incidence of chronic pain after surgery could not be linked to a postoperative analgesic technique as cases were evenly distributed among groups (EB 0; 1 intercostal neuralgia report, 3 cases of malignancy related pain; EB; 3 intercostal neuralgia reports, no malignancy related pain; PB 2 intercostal neuralgia reports; and 1 malignancy-related pain report). Seven subjects were lost to follow-up or died of their primary cancer during the ensuing 12 months. Table 2. VAS Score at Rest and With IS, the Number of Subjects Obtaining IS Volume >2 L Data, and Basal Rates of Epidural Group Reflecting Titration of Dose in Milliliters of Solution EB O(n 24) EB (n 21 ) PB (n 23) Preoperative IS 2 L 25/25 25/25 25/25 PACU VAS (rest) * VAS (IS) * IS 24/24 12/21 17/23 Basal (ml) POD1 PM VAS (rest) * VAS (IS) * IS 20/24 11/21 18/23 Basal (ml) POD2 AM VAS (rest) * VAS (IS) * IS 19/24 11/21 17/23 Basal (ml) POD2 PM VAS (rest) * VAS (IS) * IS 20/24 12/21 18/23 Basal (ml) POD3 AM VAS (rest) * VAS (IS) IS 18/24 11/21 18/23 Basal (ml) POD3 PM VAS (rest) * VAS (IS) * IS 22/24 15/21 19/23 Basal (ml) POD4 AM VAS (rest) VAS (IS) IS 23/24 19/21 20/23 Basal (ml) NOTE. PB basal rates were not adjusted as per protocol. VAS scores are reported as mean standard error. PB was continuous paravertebral infusion with bupivacaine via the OnQ Pain Relief System, iflow Corp. *p Three subjects with hypotension receiving 0.125% bupivacaine dose reduction included in the intention-to-treat analysis. EB O incentive spirometry VAS data missing for 1 subject on POD4 (n 23), impacting comparison across groups.

5 THORACIC EPIDURAL BUPIVACAINE 87 Fig 2. A line graph comparison of patient-recorded VAS scores at rest in the PACU on the evening of POD1 and the morning and the evening of POD2 and 3. POD4 represents the morning evaluation. Paravertebral block (PB), epidural bupivacaine opioid hydromorphone (EB O), and epidural bupivacaine alone (EB) are represented; *p < DISCUSSION The current prospective study suggests that paravertebral infusion of bupivacaine alone via the paravertebral route provides comparable analgesia to epidural infusion of bupivacaine alone. This finding agreed with the results of 2 recent metaanalyses. 1,2 In these reviews, both authors concluded that the evidence globally suggests that extrapleural paravertebral analgesia via an indwelling catheter is roughly equivalent to that of thoracic epidural analgesia. 1,2 One of the limitations of the meta-analyses of these studies, however, is the variability of the techniques used, thus leading the authors of both reviews to suggest that further prospective trials are required to adequately address the issue. 1,2 Although the individual studies reviewed by Joshi et al 1 often used different measures and time points, several consistent findings regarding post-thoracotomy analgesia emerged. For instance, in 4 studies using local anesthetic only in the paravertebral and epidural catheters, analgesia was found to be roughly equivalent. 6-9 In contrast, Richardson et al 10 reported superior analgesia with continuous paravertebral infusion (CPI) of 0.5% bupivacaine compared to continuous TEA with 0.25% bupivacaine. Additionally, these authors 6-10 suggested that there are fewer postoperative complications with CPI. One limitation pointed out by the authors in the Richardson study is the fact that more concentrated bupivacaine was used in the CPI compared with the TEA. 10 Although the data appear to suggest that epidural and paravertebral modalities with local anesthetic only provide comparable analgesia, the findings of the Richardson study suggested that further investigation may be warranted and that local anesthetic concentration may play a role. This study examined this issue using a direct comparison of the techniques with equivalent local anesthetic concentrations. In agreement with the balance of the data reported to date, the authors found that CPI and TEA with 0.25% bupivacaine were essentially equivalent with regard to reported VAS scores on all postoperative days as well as functional assessment via incentive spirometry. Although there are many studies comparing TEA and CPI infusion using local anesthetic alone, an important limitation in the current literature lies in the lack of studies comparing CPI with TEA with a combination local anesthetic and opioid technique, 1,2 which is a common practice. 11 Two studies 11,12 provide data comparing CPI with bupivacaine/ropivacaine with TEA with bupivacaine/ropivacaine and a lipophilic opioid such as fentanyl or sufentanil, with one study 11 suggesting that they provide similar analgesia, whereas the second 12 showed improved pain scores in the paravertebral group. With regard to post-thoracotomy epidural analgesia, Joshi et al 1 suggested that the combination of lipophilic opioid with local anesthetic was superior to opioid alone. It recently has been suggested that analgesia with lipophilic opioids administered via the epidural route may be primarily from a systemic rather than neuraxial effect Conversely, hydrophilic opioids, such as morphine and hydromorphone, commonly are used in continuous epidural infusions and may provide more reliable neuraxial analgesia than the more lipophilic opioids such as fentanyl and sufentanil. 24 The present study findings suggested that epidural analgesia with bupivacaine and hydromorphone combined is superior to that of TEA or CPI with bupivacaine alone. This observation is important because, to the authors knowledge, this direct comparison has not been reported previously in thoracotomy patients. Moreover, there are few randomized trials evaluating administration of epidural hydromorphone for most surgical procedures despite the common usage of this opioid for systemic and epidural analgesia clinically. 25 A recent large-scale randomized trial demonstrated that epidural hydromorphone in combination with dilute bupivacaine, 0.06% provided excellent analgesia for postoperative pain following orthopedic surgery. 25 Before this report, only 2 other randomized, controlled trials evaluated epidural hydromorphone for postoperative analgesia; none involving thoracotomy patients. 26,27 Further, lipophilic agents, such as fentanyl and sufentanil, are the opioids that have been studied in conjunction with TEA for post-thoracotomy pain. 1,2 Table 3. VAS Score, Incentive Spirometry Data, and Intravenous PCA Incidence From All Collection Periods EB O(n 24) EB (n 22 ) PB (n 23) Mean standard error VAS pain score * Basal rate 3.24 ml Basal rate 5.1 ml Intravenous PCA 1/24 3/18 5/23* Spirometry ( 2 L) 21/24* 14/18 18/23 *p Three subjects with hypotension receiving 0.125% bupivacaine dose reduction included in the intention-to-treat analysis. EB O incentive spirometry VAS data missing for 1 subject on POD4 (n 23) impacting comparison across groups.

6 88 GRIDER ET AL Although these opioids do provide analgesia as alluded to earlier, it has been shown that this pain relief is via systemic uptake and redistribution to the brain and not via a spinallymediated action of more hydrophilic opioids. 24 The results from the current study suggested that epidural hydromorphone is effective for post-thoracotomy analgesia with predictable and manageable side effects or minor adverse events common to epidural opioid administration; most commonly, pruritus and possible urinary retention (although this was not assessed in the present study). It has been suggested that CPI results in fewer postoperative adverse events than TEA because epidural-related hypotension is among the most commonly reported adverse events with TEA. 1 Mukergee et al 28 recently reported that systolic and mean arterial blood pressures were lower in the immediate postoperative period in TEA subjects compared with CPI. In keeping with these findings, the present results suggested that TEA with bupivacaine alone was more likely to result in hypotension than CPI. In this study, unresolved hypotension was a criterion for study alteration with dose of local anesthetic concentration reduction because of the common practice of fluid restriction concomitant with lung resection. Interestingly, similar degrees of hypotension were not seen in the TEA bupivacaine plus opioid group. Post hoc analysis revealed that basal infusion rates to achieve acceptable analgesia in the TEA bupivacaine without opioid group (EB) was higher than the basal infusion rates of the TEA group with opioid group (EB O), possibly resulting in a greater degree of sympathectomy and, ultimately, hypotension (Table 2). These findings suggested that the addition of opioid may play an important role in lessening the amount of basal infusion required to achieve analgesia. These findings are in keeping with other studies that suggested a synergistic relationship between opioid and local anesthetics when administered epidurally. 1,2 All other adverse events observed during the current trial were considered of a minor nature routinely encountered during epidural analgesia. The rates of urinary retention were not obtained because all TEA subjects were maintained with a Foley catheter during infusion. However, urinary retention is an important limitation of epidural analgesia not seen with CPI. Additionally, there were no cases of pruritus in the CPI or TEA subjects receiving bupivacaine alone. In contrast, approximately 35% of subjects in the TEA bupivacaine and hydromorphone group reported pruritus that required treatment. Although not a primary outcome of the current study, these rates of pruritus would be similar to those seen with morphine in combination with local anesthetic. 1,2 The use of the elastomeric infusion device (OnQ Pain Relief System) has not been established for use in paravertebral analgesia. Although also not the primary focus of the study, it should be pointed out that this system has been used for the continuous infusion of local anesthetic, resulting in acceptable analgesia in other surgical settings. 29,30 A 2005 retrospective study reported that for those undergoing thoracotomy this infusion system would provide acceptable analgesia when one catheter was placed at the pericostal sutures near the intercostals nerve and a second placed in the subcutaneous fascia. 15 The current study used this elastomeric infusion system to provide constant local anesthetic flow using the catheter placement technique previously described by Sabanathan. 16 As in the study by Wheatly et al, 15 the present results suggested reliable infusion and, ultimately, analgesia via CPI of bupivacaine. One growing area of interest among pain practitioners is the rate of conversion in patients from acute thoracotomy pain to chronic pain of a nonmalignant origin. It has been reported that up to 50% of patients undergoing a thoracotomy may experience ongoing pain after surgery. 3 In this report, most patients described their ongoing pain as mild. 3 Furthermore, a correlation was made between the severity of postoperative pain experienced and the incidence of the development of chronic pain. 3 This correlation suggested that adequate analgesia is essential in the immediate postoperative period to prevent the conversion from acute to chronic pain. 3 Although not a primary focus of the present study, approximately one-fifth of subjects were experiencing ongoing chronic pain after thoracotomy, with many experiencing pain consistent with intercostal neuralgia/chest wall pain. Limitations of the current study included the relatively small numbers of subjects in each group and the number of data points absent because of the inability to obtain VAS and IS on patients requiring ventilator support. Additionally, the increased basal rate required in the bupivacaine-alone group may have contributed to the incidence of hypotension in that group. The hypotension rate and concomitant higher basal rate required in this group to provide analgesia suggested the synergistic benefit of the addition of an opioid to the epidural solution for thoracotomy patients. Another issue of concern during the institutional review board process centered on the placement of an analgesic device that would not be activated and the ethics concerning this experimental protocol. Because there certainly is morbidity and possibly mortality associated with inadequate analgesia after thoracotomy, it was believed by the institutional review that the gold-standard therapy (ie, a thoracic epidural) should be in place and ready for activation should the patient fail to receive adequate analgesia from the study intervention or in combination with IV PCA. It should be noted that no subject required withdrawal and conversion to TEA in the CPI. The current study provided data that fill gaps in the current literature in 3 important areas. First, this study found that TEA with bupivacaine and a hydrophilic opioid hydromorphone may provide enhanced analgesia over TEA or CPI with bupivacaine alone. Second, the increased basal rates in the bupivacainealone group required to achieve analgesia resulted in hypotension more frequently than in the bupivacaine/hydromorphone combination group, underscoring the benefit of the synergistic activity. Finally, in agreement with previous retrospective studies, the current data suggested that CPI of local anesthetic appear to provide acceptable analgesia for post-thoracotomy pain.

7 THORACIC EPIDURAL BUPIVACAINE Joshi GP, Bonnet F, Shah R, et al: A systematic review of randomized trials evaluating regional techniques for post-thoracotomy analgesia. Anesth Analg 107: , Davis RG, Myles PS, Graham JM: A comparison of the analgesic efficacy and side effects of paravertebral vs. epidural blockade for thoracotomy A systematic review and meta-analysis of randomized trial. Br J Anaesth 96: , Plumis WA, Steegers AH, Verhagen AFTM, et al: Chronic postthoracotomy pain: A retrospective study. Acta Anaesthesiol Scand 50: , Soto RG, Fu ES: Acute pain management for patients undergoing thoracotomy. Ann Thorc Surg 75: , Hansdottir V, Philip J, Olsen MF, et al: Thoracic epidural versus intravenous patient-controlled analgesia after cardiac surgery. Anesthesiology 104: , Dhole S, Mehata Y, Saxena H, et al: Comparison of continuous thoracic epidural and paravertebral blocks for post-operative analgesia after minimally invasive direct coronary artery bypass surgery. J Cardiothorac Vasc Anesth 15: , Matthews PJ, Govenden V: Comparison of continuous paravertebral and extradural infusions of bupivacaine for pain relief after thoracotomy. Br J Anaesth 62: , Perttunen K, Nilsson E, Heinonen J, et al: Extradural, paravertebral and intercostals nerve block for post-thoracotomy pain. Br J Anaesth 75: , Wedad M, Zaki MK, Haleem M: The effect of additional wound infiltration with local anaesthetics to interpleural block on post-thoracotomy pain, pulmonary function and stress response in comparison to thoracic epidural and paravertebral block. Egypt J Anaesth 20:67-72, Richardson J, Sabanathan S, Jones J, et al: A prospective, randomized comparison of preoperative and continuous balanced epidural or paravertebral bupivacaine on post-thoracotomy pain, pulmonary function and stress response. Br J Anaesth 83: , DeCosmo G, Aceto P, Campanale E, et al: Comparison between epidural and paravertebral intercostals nerve block with ropivacaine after thoracotomy: Effects on pain relief, pulmonary function, and patient satisfaction. Acta Med Romana 40: , Kaiser AM, Zollinger A, De Lorenzi D, et al: Prospective, randomized comparison of extrapleural versus epidural analgesia for post-thoracotomy pain. Ann Thorac Surg 66: , Brimston DN, McGee JP, Liptay MJ, et al: Continuous paravertebral extrapleural infusion for post-thoracotomy pain management. Surgery 126: , Horlocker TT, Wedel DJ, Rowlingson JC, et al: Regional anesthesia in the patient receiving antithrombotic or thrombolytic therapy: American Society of Regional Anesthesia and Pain Medicine evidence-based guidelines (Third Edition). Reg Anesth Pain Med 35:64-101, 2010 REFERENCES 15. Wheatly GH, Rosenbaum DH, Paul MC, et al: Improved pain management outcomes with continuous infusion of a local anesthetic after thoracotomy. J Thorac Cardiovasc Surg 130: , Sabanathan S, Bickford Smith PJ, et al: Continuous intercostals nerve block for pain relief after thoracotomy. Ann Thorac Surg 59: , Guinard JP, Mavrocordatos P, Chiolero R, et al: A randomized comparison of intravenous versus lumbar and thoracic epidural fentanyl for analgesia after thoracotomy. Anesthesiology 77: , Loper KA, Ready LB, Sowney M, et al: Epidural and intravenous fentanyl infusion are clinically equivalent after knee surgery. Anesth Analg 70:72-75, Ellis DJ, Millar WL, Reisner LS: A randomized double-blind comparison of epidural versus intravenous fentanyl infusion for analgesia after cesarean section. Anesthesiology 72: , Ginosar Y, Riley ET, Angst MS: The site of action of epidural fentanyl in humans: The difference between infusion and bolus administration. Anesth Analg 97: , Berti M, Casati A, Fanelli G, et al: 0.2% ropivacaine with or without fentanyl for patient controlled epidural analgesia after major abdominal surgery: A double blind study. J Clin Anesth 12: , Miguel R, Barlow I, Morrell M, et al: A prospective, randomized, double-blinded comparison of epidural and intravenous sufentanil infusions. Anesthesiology 81: , Bernards CM, Hill HF: Morphine and alfentanil permeability through the spinal dura, arachnoid, and pia mater of dogs and monkeys. Anesthesiology 73: , Umenhoffer WC, Arends RH, Shen DD, et al: Comparative spinal distribution and clearance kinetics of intrathecally administered morphine, fenatanyl, alfentanil and sufentanil. Anesthesiology 92: , Liu SS, Bieltz M, Wukovits B, et al: Prospective survey of patientcontrolled epidural analgesia with bupivacaine and hydromorphone in 3,736 postoperative orthopedic patients. Reg Anesth Pain Med 35: , Liu S, Carpenter RL, Mulroy MF, et al: Intravenous versus epidural administration of hydromorphone. Effects on analgesia and recovery after radical prostatectomy. Anesthesiology 82: , Parker RK, White PF: Epidural patient-controlled analgesia: An alternative to intravenous patient-controlled analgesia for pain relief after cesarean section. Anesth Analg 75: , Mukerjee M, Goswami A, Gupta SD, et al: Analgesia in postthoracotomy patients: Comparison between thoracic epidural and thoracic paravertebral blocks. Anesth Essays Res 4:75-80, Sanchez B, Waxman K, Tatevossian R, et al: Local anesthetic infusion pumps improve postoperative pain after inguinal hernia repair: A randomized trial. Am Surg 70: , Ashcraft EE, Baillie GM, Shafizadeh SF, et al: Further improvements in laparoscopic donor nephrectomy: Decreased pain and accelerated recovery. Clin Transplant 15:59-61, 2001 (suppl 6)