The hypothesis is that epidural pethidine is an effective form of pain relief following lumbar spinal surgery, resulting in significantly lower usage of concomitantly administered (intravenous) patient-controlled analgesia (PCA) pethidine.

Rationale Lumbar laminectomy is commonly performed to alleviate local or nerve root pain in the lower back and legs. Post-operative pain relief may consist of oral or parenteral medication, with an increasing body of research focussing on the use of epidural analgesics and anaesthetics (see references). It is the current practice of surgeons at this hospital to use an epidural infusion of pethidine in some patients.

Epidural medication has sometimes been administered at the conclusion of an operation as a single dose, in the form of a spray1 2 3, paste 4, or been sponged onto the exposed epidural space.5 Agents used in this manner have included morphine, tramadol, buprenorphine and methylprednisolone. In one such study Bourke6 found that a single dose of epidural morphine 3 mg reduced pain scores and analgesic requirements for 24 hours when compared to the same dose given intramuscularly, suggesting that direct epidural administration is superior to parenteral administration.

As an alternative, analgesics have been infused post-operatively via and epidural catheter placed at the end of surgery under direct vision. Cohen7 compared an epidural infusion of morphine and bupivacaine to PCA morphine in 54 patients and found no difference in outcome between groups. Other studies which have been uncontrolled or unblinded have demonstrated the efficacy of epidural narcotic administration (morphine, fentanyl and hydromorphone) following lumbar laminectomy surgery.8 9 10 11 12 Thus there is some evidence supporting the use of this form of analgesia, despite a report of an incidence of technical failure in surgically-positioned epidural catheters. 13

To date, no published study has investigated the use of epidural pethidine following lumbar laminectomy. Pethidine has some theoretical advantage over other agents used.14 As an opioid with intermediate lipid solubility, it is less likely to leave the epidural space than fentanyl. The low lipid solubility of morphine increases its likelihood of causing central respiratory depression as it crosses the dura to the site of action at a slower rate than pethidine.14 Pethidine also has an intrinsic local anaesthetic activity.

An early study showed that patients receiving a single dose of epidural pethidine at the end of major surgery had a longer period before further analgesics were required than patients receiving equipotent doses of morphine or pethidine. Blake15 compared patient-controlled epidural pethidine (PCEA) with intravenous patient-controlled analgesia (PCA) pethidine following abdominal aortic surgery, where all patients also received epidural bupivicaine. The PCEA group had lower pain scores, despite having lower plasma pethidine concentrations, confirming the central action of epidural pethidine. In a similar study, Chen16 compared epidural and PCA pethidine in 37 patients following gastrectomy. Pain and satisfaction scores were similar, but the epidural group had lower plasma concentrations of pethidine and norpethidine.

The current proposed study will compare the efficacy of a continuous infusion of epidural pethidine through a surgically-placed catheter to PCA pethidine in patients who have undergone a lumbar laminectomy. Both strategies are considered usual practice for this institution. PCA pethidine has been used for postoperative analgesia following a variety of surgeries. Compared to PCA morphine and fentanyl it has been shown to associated with similar patient satisfaction17. Pethidine also caused less vomiting and pruritis than other agents.18

One potential consideration with the use of pethidine is its metabolite norpethidine. Norpethidine is an excitatory neurotoxin that has been shown to cause seizures in animals. It is renally excreted, and hence may accumulate in patients with renal failure. There are case reports of patients experiencing seizures during the use of PCA pethidine.19 20 21 In each of these cases the patients had received high doses, long-term therapy, or both. The average reported dose administered in the 24 hours prior to the seizure was 1900 mg. By contrast, the mean 24-hour dose in the Woodhouse study18 was 410 mg, and none of the 135 patients in those studies18 17 exhibited any excitatory side effects.

Hypothesis The hypothesis is that epidural pethidine is an effective form of pain relief following lumbar spinal surgery, resulting in significantly lower usage of concomitantly administered (intravenous) PCA pethidine.

Methods A double-blind randomised controlled trial comparing a continuous epidural pethidine infusion to a placebo normal saline epidural infusion following lumbar laminectomy surgery. Both groups will receive intravenous PCA pethidine analgesia as well as routine analgesic adjuvants (paracetamol, diazepam). The follow-up period will be 48 hours.

Inclusion criteria:

• Adults undergoing lumbar spinal surgery

Exclusion criteria:

• Lack of informed patient consent
• Acute or chronic renal failure
• Known allergy or intolerance to pethidine or tramadol
• Chronic respiratory insufficiency
• Epidural contraindicated (coagulopathy, systemic infection)

All consenting patients will be randomised to one of two groups, the epidural pethidine group (group P) or the placebo group (group N). All patients will receive fentanyl 2 - 3 mcg/kg intra-operatively as part of a balanced anaesthesia. Otherwise the anaesthesia will not be controlled. At the conclusion of surgery, all patients will have an epidural catheter placed under direct vision by the consultant surgeon. In the recovery room, all patients will have a patient-controlled analgesia (PCA) machine attached to the side arm of the intravenous infusion line. All participants will receive written as well as standard verbal instruction on the use of the PCA. The 30 ml PCA syringe will contain pethidine 10 mg/ml saline. Patients will be given standard instructions on the use of the PCA machine. The PCA machine will be set to deliver a one millilitre bolus with a five minute lockout, and a maximum dose of 20 ml (200 mg) over four hours. In addition, an infusion will be connected to the epidural catheter, containing either pethidine 250 mg in 250 ml saline (group P) or saline 250 ml (group N). The infusion will be commenced at 10 -20 ml/h according to the preference of the attending anaesthetist. This will generally depend on the age and weight of the patient.

Patients will in the high-dependency area of the neurosurgical ward and will be reviewed frequently in the usual manner of patients receiving parenteral narcotics. Patients who develop pain will be encouraged to use the PCA. Any patient expressing dissatisfaction in their pain relief will be offered an injection of intramuscular morphine (50 mcg/kg) as a rescue medication up to every 2 hours. At the same time, the epidural infusion will be increased by 10 ml/h to a maximum of 40 ml/h. All patients will be visited by blinded study personnel at 1, 4, 24 and 48 hours for collection of data, and also twice daily by the hospital acute pain service. At the end of 48 hours, the epidural catheter will be removed and patients will be treated according to usual hospital pain management practices.

In the event of a severe adverse reaction, or uncontrollable pain, the patient will be excluded from further participation in the study, and will be unblinded. Pain management will be in accordance with the wishes of the acute pain team and treating unit. Data from these patients will be analysed according to "intention to treat" criteria.

Primary end-point:

• Cumulative 24-hour pethidine consumption

Patient data:

• Age
• Gender
• Body mass index
• Current medications
• Preoperative opioid analgesic use (none, low, high - high dose being > 300 mg codeine or dextropropoxyphene daily or oral morphine or parenteral opioid use)
• Compensable status (HNC, TAC, WCV or PMI/SUR) 22

Surgical data:

• Anatomical extent of surgery (number of spinal levels)
• Highest anatomical surgical level
• Experience level of primary surgeon (trainee or consultant)
• Anatomical level of epidural tip (identified from postoperative X-Ray when taken as routine care only)
• Dural tear at the time of surgery (yes/no)
• Spinal instrumentation (yes/no)

Other end-points (all at 1, 4, 24 and 48 hours unless stated):

• Cumulative pethidine dose
• Cumulative morphine dose
• VAS scores for pain at rest and during movement
• Sedation score (1 - 4) 16
• VAS scores for nausea, pruritis
• Other adverse events: agitation, tremor, hallucinations, seizure
• Patient satisfaction scale for pain control during study (48 hours) (very dissatisfied, dissatisfied, neutral, satisfied, very satisfied) 17
• Plasma pethidine and norpethidine levels (24 hours)
• Physiotherapy assessment of ability to deep breathe & cough (unable, poor, adequate, good)
• Length of inpatient stay

Statistical considerations The study will be powered to detect a difference in cumulative 24-hour pethidine consumption. Based on a baseline 24-hour consumption of 410 mg 18 with a standard deviation of 100 16 , a power of 0.8 and a significance level of 0.05, the sample size required to detect a 20% difference in pethidine consumption is 24 subjects per group. This difference would be considered clinically significant. A total of 60 patients will be enrolled to allow for loss of data due to participant drop-out.

The analysis will be on an intention-to-treat basis. The unpaired Student's t-test and the Mann-Whitney U-test will be used to analyse univariate parametric and non-parametric data respectively. Serial measurements will be analysed by using repeated measures analysis of variance; any significant overall differences between groups will be investigated with t-tests at individual time intervals. The Chi-squared test will be used to compare categorical data. A p value < 0.05 is considered significant. The number and reason for participants not completing the study will be reported.

Adverse event reporting Information concerning adverse events is to be collected as end-points for the study. This includes pain scores, sedation scores, respiratory depression, nausea, pruritis, agitation, tremor, hallucinations, and seizures.

All other adverse events, whether resulting from the use of study medication or procedural errors, or apparently independent of the study, will be recorded.

A serious adverse event is defined as one which causes patient death, is life threatening, prolongs hospital stay or results in a persistent or significant disability or incapacity (TGA). An independent physician will be appointed to review all serious adverse events. This review will occur within 24 hours of the adverse event occurring. The monitoring physician may suspend the study if warranted on the basis of the occurrence of adverse events. All serious adverse events will be reported to the Drug Trial Subcommittee of the Human Research Ethics Committee as soon as possible.

Inclusion Criteria:

• Adults undergoing lumbar spinal surgery

Exclusion Criteria:

• Lack of informed patient consent
• Acute or chronic renal failure
• Known allergy or intolerance to pethidine or tramadol
• Chronic respiratory insufficiency
• Epidural contraindicated (coagulopathy, systemic infection)