| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Departments of *Anesthesiology and Intensive Care, Orthopedics; and Institute of Medical Biometry, Charité Universitätsmedizin Berlin, 10117 Berlin, Germany.
Address correspondence and reprint requests to Thomas Volk, MD, Department of Anesthesiology and Intensive Care, Campus Mitte, Charité Universitätsmedizin Berlin, Schumannstrasse 20/21, 10117 Berlin, Germany. Address e-mail to thomas.volk{at}charite.de.
Abstract
BACKGROUND: Spinal fusion surgery causes severe postoperative pain, hampering reconvalescense. We investigated the efficacy of patient-controlled epidural analgesia (PCEA) in a prospective, double-blind, randomized, controlled comparison with patient-controlled IV analgesia (PCIA).
METHODS: After lumbar anterior-posterior fusion receiving an epidural catheter intraoperatively, 72 patients were given either PCEA (ropivacaine 0.125% and sufentanil 1.0 µg/mL at 14 mL/h; bolus: 5 mL; lockout time: 15 min) and IV placebo or PCIA (morphine 2.0 mg/mL; bolus: 3 mg; lockout time: 15 min) and epidural placebo. Pain levels (visual analog scale 0-10), functional capabilities (turning in bed, standing, and walking), analgesic consumption, and side effects were evaluated until 72 h after surgery.
RESULTS: Fourteen patients were excluded by predetermined criteria, leaving 58 patients for data analysis. Pain levels at rest and during mobilization were significantly lower in the PCEA when compared with that in the PCIA group throughout the study period (P < 0.0001 in all cases). Time until able to turn in bed was achieved earlier in the PCEA group (P < 0.05). Patients in the PCEA group were significantly more satisfied with pain therapy (P < 0.01).
CONCLUSION: We conclude that PCEA with ropivacaine and sufentanil, using intraoperatively placed epidural catheters, provides superior analgesia and higher patient satisfaction when compared with PCIA after spinal fusion surgery.
Major spinal fusion surgery causes severe postoperative pain, which persists for at least 3 days (1–6). Bianconi et al. (1) reported mean maximal pain scores at rest which were 73 ± 9 (visual analog scale, VAS score 0-100) 4 h after posterior fusion surgery and which declined to approximately 35 on the third postoperative day. Efficient and safe methods for postoperative analgesia after spinal fusion surgery are, therefore, mandatory. Intrathecal analgesia has been used after "lumbar spine procedures," (7) spinal fusion (8), laminectomy, discectomy, hemilaminotomy, and foraminotomy (9). Epidural analgesia was used after lumbar laminectomy (10), posterior, anterior or combined fusion (11), posterior fusion (2,4,12,13) with or without decompression, (14) and after posterior or anterior-posterior lumbar fusion (3).
For continuous postoperative analgesia, epidural administration is an established and safe method used routinely for other operative procedures. Epidural opioids (10), ropivacaine (3,4), or a combination of opioids with bupivacaine (2,11,13,14), have been reported to be useful after spine surgery. Patients’ postoperative functional rehabilitation is hampered by intense postoperative pain. Although fairly high doses of IV opioids are typically necessary, epidural analgesia may result in fewer side effects.
The efficacy of patient-controlled epidural ropivacaine with sufentanil (PCEA) when compared with conventional IV patient-controlled morphine treatment (PCIA) has not been evaluated in this patient group. Therefore, our primary outcome measure was postoperative pain. We hypothesized that pain levels, both at rest and during mobilization maneuvers would be lower when epidural analgesia is used compared to PCIA. Furthermore, we chose to determine the impact of these analgesic regimens on patient satisfaction and on defined postoperative mobilization maneuvers.
METHODS
The study was approved by our local ethics committee (Berlin, Germany). Seventy-two patients, ASA class I-III aged between 18 and 75 yr, were included after obtaining written informed consent. All patients had lumbar spondylolisthesis and were scheduled for combined anterior-posterior spine fusion surgery. Exclusion criteria were mental illness, inability to use a patient-controlled pump, drug addiction, renal or hepatic insufficiency, and treatment with opioids or CYP1A2-inhibitors such as fluvoxamine or enoxacin. Patients were excluded from the study if the dura was damaged intraoperatively, and if the epidural catheter became dislocated. Before surgery, patients were thoroughly examined with particular emphasis on their neurological status. Patients were familiarized with the use of a patient-controlled analgesic system (PCA), the quantification of pain with a VAS, and potential adverse drug events related to epidural analgesia, such as sensory or motor block or clinical signs of toxicity. Patients were premedicated (1.0 mg flunitrazepame at night; 0.1 mg/kg midazolam on the day of surgery). Anesthesia was induced with propofol 2 mg/kg, fentanyl 2.0 µg/kg, and cisatracurium 0.1-0.15 mg/kg and maintained with propofol 6-8 mg · kg–1 · hr–1, cisatracurium 1.0-2.0 mg · kg–1 · min–1 and fentanyl. Patients’ lungs were ventilated with an oxygen-air mix (Fio2, 0.5).
The anterior spinal fusion was performed first with the patient in the supine position. The intervertebral disk material was removed and the anterior longitudinal ligament was cut for segmental release. Titanium interbody cages were placed in the disk spaces and filled with lyophilized, allogenic spongiosa. For posterior fusion, pedicle screws were inserted as anchor points and connected with a fixateur interne for transpedicular stabilization and distraction.
After posterior instrumentation, the surgeon inserted the epidural catheter (Portex® DuraFlexTM epidural catheters 18 G; NH) in the midline in both groups, one segment above the fusion and 3-6 cm into the epidural space. Intervertebral puncture level and catheter depth in the epidural space were documented. Patients were evaluated neurologically after tracheal extubation to detect new deficits. The epidural catheters were tested by aspiration and the administration of a test dose (3 mL bupivacaine 0.5% with epinephrine 1:200.000).
Before the operation, patients were randomized by a computerized list to the PCEA or PCIA group. The study was double-blind. Postoperative analgesic therapy was initiated with a bolus. Patients in the PCEA group received epidural ropivacaine 0.2% (14 mL) with sufentanil 10 µg (2 mL) and IV placebo NaCl 0.9% (5 mL). Patients in the PCIA group received epidural placebo NaCl 0.9% (16 mL) and IV morphine 5 mg (5 mL).
Thereafter a PCA pump (CADD-PCA® model 5800, Smiths Medical MD, St. Paul, MN) was connected to patients (reservoir: either ropivacaine 0.125% with sufentanil 1.0 µg/mL or morphine 2.0 mg/mL). To proceed with a double-blind design, the pump was placed in a nontransparent sealed bag. Both epidural catheter and central venous line ended in the bag, but only the appropriate line was connected to the reservoir.
Patients in the PCEA group received a continuous epidural infusion of 14 mL/h with additional patient-controlled bolus doses of 5 mL (lockout time 15 min). Patients in the PCIA group titrated their analgesia with IV bolus doses of 3 mg (lockout time 15 min).
Independent of the examiners, acute pain physicians not involved in the study had the choice of adjusting the infusion rate according to clinical demands. The rate was increased with pain VAS >3 at rest or VAS >6 with movement. The rate was decreased when patients had intolerable relevant motor block (Bromage score >0) or sensory disturbances (numbness), or hypotension (systolic blood pressure <90 mm Hg). Epidural catheters were removed on the fourth postoperative day. IV rescue analgesia was metamizole (1 g per 6 h). All clinical assessments were performed at regular intervals on the day of surgery: 1, 2, and 4 h after the initial bolus; twice daily at 9:00 a.m. and 6:00 a.m. until day 3 after surgery.
Pain was assessed using the VAS ranging from "0" (no pain) to "10" (worst imaginable pain). Pain was evaluated at rest, while coughing, and during mobilization. Maneuvers of particular clinical importance for postoperative mobilization (alone and with help) were chosen: Turning in bed, standing in front of the bed and walking, and using the toilet without help. The time needed until the patient could first successfully perform these maneuvers was documented.
For assessment of patients’ satisfaction with postoperative pain management a verbal rating score was used (4 = very satisfied, 3 = satisfied, 2 = neutral, 1 = dissatisfied, and 0 = very dissatisfied).
Motor block was quantified with the Bromage scale (0 = free movement of legs and feet, 1 = just able to flex the knees with free movement of the feet, 2 = unable to flex the knees but with free movement of the feet, and 3 = unable to move legs or feet). Patients were asked about sensory deficits.
Verbal rating scores were used for sedation (0 = awake, 1 = sleepy, 2 = sleeping, rousable, 3 = not rousable), nausea and vomiting (0 = no nausea or vomiting, 1 = nausea, 2 = vomiting, 3 = nausea and vomiting) and the incidence of pruritus (0 = no pruritus, 1 = low-grade, 2 = severe, and 3 = unbearable).
In an observational study (5), we evaluated the efficacy of epidural pain therapy after major spinal surgery by bolus administration. In this pilot study, the mean pain score (VAS ranging from 0 to 10) was 4.8 (sd 2.4) at rest after surgery and before initiation of postoperative epidural analgesia. The relevant value assumed for better pain therapy was two. A type I 
error of 5% and a type II ß error of 90% indicated that a sample size of 31 patients per group would be necessary to show a difference. To compensate for potential exclusions or withdrawals, we decided to include five more patients per group.
Data were expressed as medians (ranges) if not declared otherwise. Demographic data, predetermined functional targets (functional results), and the incidence of adverse events at determined time points were compared with the Mann-Whitney U-test. Significance was assessed at the P < 0.05 level. A nonparametric analysis of variance including multiple comparison adaptation (Brunner analysis) was performed to compare group differences during the whole course of treatment with respect to pain and satisfaction.
RESULTS
Of the 72 patients entering the study, 14 were excluded, with no significant difference between groups (P = 0.55). Reasons for the exclusion of six patients in the PCIA group were: intraoperative leakage of cerebrospinal fluid (n = 3) and withdrawal of consent (n = 3). Reasons for the exclusion of eight patients in the PCEA group were intraoperative leakage of cerebrospinal fluid (n = 2), inability to place the epidural catheter (n = 1), blood aspiration at testing (n = 1), catheter occlusion (n = 2), and withdrawal of consent (n = 2). Demographic and perioperative patient data included in the study did not differ between groups (Table 1). Epidural catheter insertion levels in the PCEA/PCIA group, respectively, were at L2/L3 for 1/2, at L3/L4 for 11/11, and at L4/L5 for 16/17 patients. The median intraepidural insertion depth of the catheters was 5 cm (2-6 cm) in the PCEA group and 4 cm (2-6 cm) in the PCIA group. None of the patients had neurological deficits caused by surgery at the neurological examination before the administration of the initial bolus. All patients received their initial bolus treatment after testing.
|
Consumption of epidural ropivacaine and IV morphine is summarized in Table 2. Patients in both groups did not use the maximal possible PCA daily dose of either morphine or epidural ropivacaine/sufentanil.
|
Metamizole rescue analgesia did not differ between groups (P = 0.227). The median (min-max) consumption per day in the PCEA and PCIA groups was 0.6 g (0-4.0 g) vs 0.4 g (0-4.0 g), respectively.
Mean (±sd) pain scores at rest for the PCEA group versus the PCIA group at 1, 2, and 4 h after surgery were 0.1 (±0.2) vs 1.7 (±1.9), 0.2 (±0.8) vs 1.6 (±1.8), and 0.3 (±0.8) vs 1.1 (±1.7). On the first, second, and third postoperative day, the pain scores were 0.7 (±1.6) vs 1.9 (±2.0), 0.4 (±1.1) vs 1.9 (±2.1), and 0.4 (±1.3) vs 1.4 (±1.7), respectively.
Early mean (±sd) pain reduction 1 h after the initial bolus in the PCEA group was –4.5 (±3.4) and in the PCIA group –2.7 (±3.0).
Patients in the PCEA group had significantly less pain when compared with patients in the PCIA group during all predetermined maneuvers: at rest, while coughing, while turning in bed, while standing, and while walking (P always <0.0001 for the whole study period). Only patients in the PCEA group who had experienced no sensory or motor deficits (n = 13) caused by epidural analgesia during the study showed similar significant differences: pain at rest (P < 0.005), while coughing (P < 0.005), while turning in bed (P < 0.05), while standing (P < 0.005), and while walking (P < 0.05). Within 20 min, pain levels at rest in PCEA patients decreased 
3 in 96% (27/28) whereas only 63% (19/30) in the PCIA group had pain levels 3 (P < 0.001). One hour after the initial bolus, 90% of patients in the PCEA group and 47% of patients in the PCIA group were completely pain-free (P < 0.0001). Median pain levels at rest and while coughing are given in Figures 1A and B. On the first postoperative day, pain levels 3 for turning around in bed were achieved in 68% of patients (19/28) in the PCEA group and 27% (8/30) in the PCIA group (P < 0.001). The percentage of patients with no pain at all, both while standing and while walking on Day 2, were 39% (11/28) in the PCEA group and 10% (3/30) in the PCIA group. Pain levels were lower in the PCEA group both while standing (P < 0.01) and walking (P < 0.05) (Figs. 2 A–C).
|
|
Time to turn in bed without assistance was decreased in the PCEA group (Table 3).
|
Although there were no sensomotoric deficits caused by the operation, 46% (13/28) of patients in the PCEA group after surgery had transient sensory deficits attributable to the local anesthetic. The corresponding numbers on the first, second, and third postoperative day were 39% (11/28), 32% (9/28), and 32% (9/28), respectively.
Transient motor block (Bromage score >0) in the PCEA group was present in five of 28 patients (18%) on the day of surgery. Two of these patients had a complete motor block after the initial epidural bolus. Two, one, and no patients had motor block on the first, second, and third postoperative day, respectively. The number of patients with Bromage scores of 0/1/2/3 on the day of surgery, first, and second postoperative day were 23/2/1/2, 26/1/1/0, and 27/1/0/0, respectively. All motor blocks and sensory deficits were reversible and attributable to the local anesthetic.
The median sedation score 1 h after the initial bolus was 1 (0-2) in both groups. The median sedation score decreased to 0 (0-2) thereafter in both groups.
The incidence of nausea or vomiting within the observation period was insignificantly less frequent in the PCEA group (n = 32) when compared with that in the PCIA group (n = 53; P = 0.356). The median pruritus score was 0 with no significant difference between groups (P = 0.45).
Patients in the PCEA group were significantly more satisfied with the postoperative analgesic treatment compared to patients in the PCIA group (P < 0.008) at all observation times and over the whole study period (Fig. 3).
|
DISCUSSION
We compared the efficacy of PCEA sufentanil with PCIA morphine. Throughout the observation period, patients in the PCEA group had significantly less pain at rest and during mobilization maneuvers, which were important for rehabilitation and they were more satisfied with their pain therapy when compared with patients in the PCIA group.
Despite excellent pain control, patients in the PCEA group did not gain functional capabilities at a significantly earlier stage when compared with those in the PCIA group, except for turning in bed without assistance. There were no substantial reasons such as postural hypotension, dizziness or a notably frequent incidence of motor block, or surgical reasons to explain this. When the study was initiated, epidural analgesia after major spine fusion surgery with its potential for pain control and early mobilization was a relatively new procedure. Retrospectively, we speculate that patients may not have been asked empathically to perform the mobilization maneuvers, which is a limitation in our study.
Although preoperative placement of epidural catheters may interfere with spine surgery, intraoperative placement has been introduced as an effective measure for treating postoperative pain (5). Serious complications related to this technique are rare. Cassady et al. (13) reported one case of wound infection, which they did not attribute to the intraoperatively placed epidural catheter. Shaw et al. (11) reported one case of respiratory depression, which occurred in a 16-year-old boy, who "was given the epidural infusion within several minutes" and became apneic. He had to be tracheally intubated for a short time, but made an uneventful recovery thereafter. Joshi et al. (10) described one case of somnolence in 10 patients. Variable technical problems were described by Gottschalk et al. (3), Fisher et al. (14), and Shaw et al. (11). Common side effects associated with epidural administration of local anesthetics or opioids, such as nausea and vomiting or pruritus were not evaluated systematically by all authors. Pruritus was described with an incidence between 7% and 43% (2,4,10–13), nausea and vomiting with an incidence between 14% and 86% (2–4,10–13).
For safety reasons, we excluded patients in whom the dura was damaged, because unpredictable amounts of the local anesthetic might have reached the intrathecal space.
Cohen et al. (2) compared fixed continuous epidural analgesia (bupivacaine 0.0625%) via intraoperatively placed epidural catheters two or three levels above the cephalad operative level with PCIA (morphine 1 mg/mL) after major spine surgery and found no differences in postoperative pain levels. Mean VAS pain scores in the epidural group on postoperative Days 1, 2, and 3 were 4.5 (±0.5 se), 3.5 (±0.5 se), and 3.8 (±0.6 se). The authors concluded that the dilute local anesthetic may not have reached the operative field in sufficient amounts.
In a case series by Lowry et al. (6) after anterior spinal fusions, 10 patients received epidural fentanyl and hydromorphone together with ropivacaine 0.125% leading to a mean pain score (VAS 0-10) of 2.3 (sd 0.9) on the first postoperative day. Gottschalk et al. (3) used epidural ropivacaine without opioid in 13 patients, after lumbar spinal fusion surgery, at a concentration of 1.0 mg/mL (cumulative systemic piritramide of approximately 100 mg after 72 h was necessary).
The excellent pain control in the present study was probably due to the higher concentration of ropivacaine, the higher infusion rate (median patient-controlled rate was 15.8 mL/h) and the use of an epidural opioid (15). Blumenthal et al. (4) used two epidural catheters after scoliosis correction surgery with an even higher concentration of ropivacaine (3 mg/mL) in 15 patients and compared it with IV morphine in an unblinded design. Mean epidural infusion rates were not specified (4-10 mL/h in each catheter), but mean pain levels (VAS 0-100) on postoperative days 1, 2, and 3 were extremely low with 8, 8, and 10, respectively. Fixed co-analgesic treatment (rofecoxib:25 or 50 mg; acetaminophen: 4 g/day) makes the results difficult to compare with our regime. These non-opioids are known for their potential to reduce postoperative opioid consumption (16,17).
Higher concentrations of local anesthetic may lead to motor block, which is counterproductive for early mobilization strategies and may confound the diagnosis of neurodeficits due to a surgical etiology. In our study, motor block occurred after the initial bolus of 14 mL of 0.125% ropivacaine in five of 28 patients in the PCEA group. Motor block was not reported after 10 mL of ropivacaine 0.1% by Gottschalk et al. (3). It is interesting to note that in the study by Blumenthal et al. (4), the higher concentration of ropivacaine (0.3%) caused a transient initial Bromage score >1 in only four of 15 patients.
To our knowledge, this is the first study comparing postoperative physical capabilities after major spine surgery based on the type of analgesia. The optimal concentration and dose of local anesthetic for epidural pain control after major spine surgery remains undefined. Even though the analgesic regime produced optimal pain control, it is unclear how these results affect functional variables relevant to postoperative mobilization. We chose mobilization maneuvers (turning in bed, standing, walking, and walking to the toilet) that are relevant to postoperative rehabilitation and patient autonomy in daily routines. Patients in the PCEA group were more satisfied with their pain therapy and experienced significantly less pain during mobilization maneuvers. Although we did not report a difference in the time to achieve functional capabilities, it is possible that with the improved pain control, a more rigorous integration of physiotherapeutic aims may enable earlier mobilization.
Footnotes
Accepted for publication July 25 2006.
This work should be attributed to the Department of Anesthesiology and Intensive Care, Campus Mitte, Charité Universitätsmedizin Berlin, Schumannstrasse 20/21, 10117 Berlin, Germany.
Supported by Charité Universitätsmedizin Berlin, Germany.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  O. Ahlers, I. Nachtigall, J. Lenze, A. Goldmann, E. Schulte, C. Hohne, G. Fritz, and D. Keh Intraoperative thoracic epidural anaesthesia attenuates stress-induced immunosuppression in patients undergoing major abdominal surgery Br. J. Anaesth., December 1, 2008; 101(6): 781 - 787. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
