Comparative study of epidural application of morphine versus gelfoam soaked in morphine for lumbar laminectomy.

BACKGROUND: Epidural application of morphine has been used for postoperative analgesia following spine surgery but short duration of action of single application limits its widespread use.
MATERIALS AND METHODS: One hundred and fifty patients undergoing lumbar laminectomy were randomly allocated to two groups of 75 patients each. Anesthetic technique was standardized in both the groups. In Group I, at the completion of laminectomy, a 5 × 1-cm strip of gelfoam soaked in 5 mg morphine (1 mg/ml) was contoured to be placed in the epidural space whereas, in group II, gelfoam soaked in saline was placed in the epidural space and 5 mg morphine (1mg/ml) was instilled over the intact epidural space. Analgesic consumption for 48 hours, time-of first analgesic request, time of ambulation, time of discharge from post anesthesia care unit (PACU) and hospital and adverse effects were recorded. The data was analyzed using appropriate statistical tests.
RESULTS: Mean analgesic consumption in 48 hours was significantly less in group I (8.47 ± 3.674 mg) as compared to group II (24.80 ± 6.009 mg). Supplemental analgesia was requested at 30.03 ± 6.796 hours in Group I, vs 10.25 ± 2.243 in group II (P < 0.001). Group I patients were discharged earlier from PACU as compared to group II (P < 0.001) though time of discharge from hospital was similar in both the groups. There were no major adverse effects except pruritis, which was observed in 30.6% patients in group I and 37.3% in group II (statistically insignificant (P > 0.01)).
CONCLUSION: Epidural application of morphine soaked in gelfoam is an effective method for prolonging the postoperative analgesia after spine surgery.

RCT Entities:   Population Interventions Outcomes

Introduction

Parenteral opioids have been the mainstay of treatment for postoperative pain after lumbar laminectomy.[1] The biggest drawback of parenteral opioids is that these drugs are usually given with relatively large time lapses, so that there are wide fluctuations in clinical effect. Ideal postoperative analgesia should provide continuous pain relief, in an alert patient who can be mobilized early.[2]

Alternatives to standard pain management include, epidural administration of narcotics as a single dose at the time of the surgery,[3] or via an epidural catheter postoperatively.[4] Epidural catheters are difficult to manage and maintain after spine surgery. In addition, there is always a concern of infection, restricting its widespread application.[5] Narcotics administered directly over the duramater (dura) before closure have been used safely in a number of studies,[6] but has limited application due to short duration of action (3-24 hours).[7]

To prolong the effect of epidural morphine, studies have been conducted by applying gelfoam soaked in opiods.[89] However, direct epidural instillation of morphine has not been compared with morphine soaked gelfoam. Thus conclusions cannot be drawn whether soaking morphine in gelfoam prolongs the analgesic effect of morphine or not. In this context, the present study was designed to evaluate the analgesic efficacy of gel foam soaked in preservative free morphine placed in epidural space and to compare it with direct application of morphine in patients undergoing laminectomy surgery under general anesthesia.

Materials and Methods

The study was conducted in a prospective, randomized, double-blinded manner on patients suffering from lumbar canal stenosis posted for lumbar laminectomy in a tertiary care hospital. Ethical clearance was obtained from the institutional research and ethical committee. A total of 150 patients of either sex between the age group of 25-50 years operated between July 2011 to December 2012 belonging to American Society of Anesthesiologists (ASA) grade I or II were included.

Patients undergoing cervical or thoracic level laminectomies, spine-fixation surgery, those with herniated sequestered disc, previous history of spine surgery, patients having significant medical diseases, neuromuscular disease or psychological disease, history of substance abuse or inability to comprehend visual analog scale (VAS) were excluded from the study. Patients, who had accidental dural tear while undergoing laminectomy, were excluded from the study.

All patients were pre-medicated with tab 0.05 mg/kg lorazepam and tab ranitidine 150 mg orally, the night before surgery. In the pre-operative holding area, inj midazolam 2 mg, inj metoclopramide 10 mg and inj glycopyrolate 0.2 mg were administered intravenously (IV) half-an-hour prior to induction of anesthesia. After shifting to operation theatre, standard monitoring (i.e., electrocardiogram with two derivations, pulse oximetry (SpO2), non-invasive arterial pressure measurement, End tidal CO2(EtCO2), body temperature) were employed and baseline parameters recorded. This monitoring was continued till reversal and extubation.

Anesthesia was induced with 2 μg/kg fentanyl, 2 mg/kg propofol, and 0.5 mg/kg atracurium IV followed by endotracheal intubation. Anesthesia was maintained with 60% N2O in O2, infusion propofol, and fentanyl 0.5-1 μg/kg per hour, as needed, to maintain mean arterial pressure between 65 and 85 mmHg. Neuromuscular relaxation was maintained with intermittent atracurium bolus (0.15 mg/kg every 15-20 minutes). Ventilation was adjusted to maintain EtCO2 between 30 and 35 mm Hg.

Patients were divided into two groups (group I and group II) of 75 each by computer-generated randomization. Preservative free morphine was diluted by the scrub nurse preparing a concentration of 1 mg/ml. At the end of the operation, before final closure, in group I, a 5 × 1-cm piece of absorbable gelatin sponge soaked in 5 mg of diluted morphine was placed in the epidural space, whereas in group II a saline-soaked gelfoam was placed in the epidural space and 5 ml of 1 mg/ml morphine was instilled over the intact epidural space by the Neurosurgeon.

All patients were given inj diclofenac 75 mg infusion over 30 min and inj ondansetron 100 mcg/kg IV half-an-hour before reversal. After surgery was over, patients were turned supine. Neuromuscular blockade was reversed with inj Neostigmine 0.05 mg/kg and inj glycopyrrolate 0.02 mg/kg IV, after return of spontaneous respiration, trachea extubated and patients shifted to post anesthesia care unit (PACU) where continuous monitoring was carried out.

In PACU, all patients were administered supplemental oxygen for first 2 hours. After this time, oxygen was administered, if patient had saturation below 92%. Pulse rate, mean arterial pressure (MAP), respiratory rate, oxygen saturation, VAS, and sedation score were recorded hourly for the first 6 hours, after that at 2 hourly intervals till 12 hours and at 4 hourly intervals till 48 hours by the staff nurse / resident anesthesiologist, who was unaware of the group allocation.

Subjective pain was measured by VAS on a scale of 0 to 10 with 0 representing no pain and 10 worst imaginable pain.[9] All patients were acquainted with VAS pre-operatively. Patients were administered supplemental analgesic with inj morphine 5 mg IV if they complained of pain or VAS ≥ 3, which could be repeated after 20 minutes if patient still complained of pain. Opioids other than inj morphine and all other analgesic or anti-inflammatory agents were prohibited for the first 48 hours after study dose. After 48 hours, alternate opioid therapies were permitted at the investigator's discretion. Naloxone was permitted for treatment of opioid-related adverse events.

The duration of pain relief was defined as the time from the end of the operation until the patient requested supplementation of analgesic. Time of first demand of analgesic and total analgesic consumption in the initial 24 and 48 hours were recorded.

Level of sedation was assessed by 5-point ordinal scale,[10] where,0 means alert patient, 1 stands for patient who is drowsy, 2 is for frequently drowsy and disoriented, 3 is very drowsy and disoriented and 4 is Stuporose and difficult to arouse.

Patients were encouraged to ambulate 6 hours after surgery if they felt comfortable. Time of ambulation was counted from the time at which patients were shifted to the PACU until the time patients could be ambulated and it was recorded in all cases. Patients were discharged from PACU to ward at any time after 8 hours if modified Aldrete's score was ≥9,[11] patient had voided and started accepting orally. At our institute, patients are usually kept in PACU for 4 hours. However, for the purpose of study, the time of discharge from PACU was kept 2 hours later than time of ambulation so that most patients could be ambulated in the PACU itself before shifting to ward to rule out pressure symptoms and respiratory depression. Ambulation, though was not mandatory before discharge from PACU. Time of discharge was noted and pulse oximetry monitoring continued in ward for 48 hours.

Patients were also observed for postoperative complications or adverse effects such as respiratory depression, nausea, vomiting, bradycardia, pressure symptoms and need for supplemental oxygen.

For the purpose of study, bradycardia was defined as heart rate less than 50 beats per minute. Respiratory depression was defined as respiratory rate less than 10 per minute, need of oxygen supplementation was recorded if patients had oxygen saturation of less than 93%, urinary retention was labeled as absence of spontaneous voiding more than 7 hours after removal of bladder catheter.[12]

Severity of pruritis was graded as:[13]

0. Absent.

1. Mild — restricted to one area not troubling the patient and reported only on prompting the patient.

2. Moderate — affecting a large area, not disturbing the patient and not requiring treatment.

3. Severe — generalized, often disturbing the patient and requiring treatment.

Pressure symptoms were recorded by observing for aggravation of preoperative symptoms and neurologic signs such as pain, including radicular pain, paresthesia, hyperesthesia, bowel and bladder dysfunction (after removal of the urinary catheter) and muscular weakness. These were observed for 48 hours postoperatively, at the time of discharge and again when the patients reported for follow-up after 2 weeks. At hospital discharge, all patients were told to report whenever they noticed any deterioration in their neurologic symptoms.

All patients were also observed for nausea and vomiting. In the nausea category, patients who had only nausea were included and those who had vomiting with or without nausea were included in the vomiting category. Nausea or vomiting was treated with inj ondandetron 100 μg/kg.

Statistical methods

The previous multicenter study found that total 48-hour postoperative rescue supplemental opioid consumption (in IV morphine milligram-equivalents) was 25 ± 21 mg in the extended release epidural morphine (EREM) 10 mg group compared to 47 ± 34 mg in the conventional epidural 5 mg morphine group.[12] Based on these findings, a hypothesis was generated that for a dose of 5 mg in both groups of gelfoam soaked in morphine and direct epidural instillation, a sample size of 70 patients was needed to detect 10% difference with 90% power and α of 0.05.

Descriptive statistics were used to summarize demographic, outcome, and adverse events data. Outcome measures of interest between the two groups were compared using Student's t-test for normally distributed variables and longitudinal data analyses were performed using repeated measures ANOVA taking time as a repeated measure. Analyses were performed using Statistical Package for the Social Sciences software (SPSS, version 12.0, SPSS Inc, Chicago, IL). The level of significance was fixed at a minimum of 5%.

Results

Both the groups were demographically similar with respect to age, weight, height and sex distribution [Table 1].

Table 1

Demographic data

Mean VAS score was zero till 3 hours in both the groups, after which mean VAS score started to increase in group II. Highest mean VAS recorded in group II was 3.61 ± 1.541 at 12 hours. In group I, mean VAS score was 0 till 4 hours after which it steadily increased and reached a peak of 2.93 ± 1.781 at 40 hours. Thus highest VAS value was greater in group II and the peak in VAS also occurred earlier. Mean VAS score at most time intervals was greater in group II as compared to group I except at 40 hours when mean VAS score was higher in group I. [Figure 1] This difference in mean VAS was statistically significant.

Figure 1

Mean VAS and sedation score vs time

Patients in group II received rescue analgesia much earlier than group I and this difference was statistically significant [Table 2].

Table 2

Time of first supplemental analgesia, ambulation, discharge from PACU and hospital

Mean consumption of morphine in first 24 hours, between 25-48 hours and total morphine consumption in 48 hours was significantly greater in group II as compared to group I (P value < 0.05) [Figure 2].

Figure 2

Morphine consumption in 0-24 hours and 25-48 hours

Mean sedation scores at a maximum time intervals of 24 hours, 36 hours and 40 hours were statistically significantly higher in group II as compared to group I. It was only after 24 hours that sedation scores were similar in both the groups [Figure 1].

Mean time to ambulation was shorter in group I as compared to group II though this difference was statistically not significant [Table 2].

Mean time to discharge from PACU in group I was 10.23 ± 3.547 hours as compared to 13.35 ± 4.660 hours in group II. This difference was statistically significant [Table 2].

Mean time to discharge from hospital was similar in both the groups with no statistically significant difference [Table 2].

Mean heart rate and MAP was lower in group 2 at most time intervals (P < 0.05) [Figure 3].

Figure 3

HR and MAP vs time

Respiratory rates were similar in both groups with no significant change from baseline. The incidence of adverse effects is given in Table 3. There was no difference in the incidence of nausea, vomiting, bradycardia, urinary retention, respiratory depression, need for naloxone and supplemental oxygen. Both the groups had similar incidence of pruritis. Total of 29.3% in group I reported mild to moderate pruritis not requiring any treatment and 1.3% of patients had severe pruritis-requiring treatment. In group II, 37.3% reported mild to moderate pruritis not needing treatment whereas 1.3% reported severe pruritis-needing treatment. None of the patients reported any pressure symptoms like radicular pain, muscle weakness or bladder and bowel involvement.

Table 3

Incidence of adverse events

Discussion

Epidural morphine sulphate has proven analgesic efficacy and superiority over systemically administered morphine for the treatment of postoperative pain.[4141516]

The easy access to epidural space during laminectomy surgery can be utilized for applying morphine directly over the exposed dura mater. However, pain relief after a single epidural injection of morphine may last less than 24 hours.[7] To overcome this, various investigators have attempted to prolong the duration of analgesic action by morphine-adon L compound,[17] Oxiplex/SP gel-morphine,[18] Vaseline-Sterile-Oil-Morphine.[19] Many of these techniques lack wider applicability. Liposomal based EREM has been recently introduced and successfully used extending the duration of analgesic action of morphine.[2021] However, this preparation is costly and not universally available.

The routine use of surgical gelfoam in the epidural space at the completion of surgery prompted the use of gelfoam as an extended release drug delivery system in this study. Earlier studies have tried application of gelfoam soaked in buprenorphine[9] and morphine[8] in epidural space, but since direct epidural application of morphine was not compared with gelfoam soaked in morphine, definitive conclusions cannot be drawn whether or not the duration of action of opioid is prolonged due to its application via a gelfoam.

The dose of epidural morphine used in our study corresponds to the doses used in earlier studies. Gibson Kevin used single dose of 4 mg in lumbar spine surgery.[8] Tzu zeng chen used 3 mg epidural morphine in lumbar laminectomy surgery.[22] Carvalho et al., used 5 mg epidural morphine in obstetric patients.[23] Gambling used 5 mg morphine in epidural space for lower abdominal surgeries.[8]

We observed that the mean VAS score in the group I remained less as compared to group II at most time intervals studied. This implies that patients in group I had less pain as compared to group II. In group II, total analgesic consumption was almost greater than 3.5 times and supplemental analgesia was required almost 20 hours earlier as compared to group I. The mean duration of action of a single dose epidural instillation of morphine was 10.25 hours in group II and it prolonged to 30.25 hours in group I. The duration of action of single dose epidural morphine in our study corresponds to that of Yadanapudi et al., who also reported that the effect of single dose of epidural morphine lasted 10.1 hours.[24] However, Carvalho reported single dose of epidural morphine to provide analgesia for 3 hours.[23] This difference could be due to the fact that the study was performed in post-cesarean patients, and pregnant state is known to alter pain perception.

Our results are similar to those of Mishra et al., who reported decreased pain scores and reduced analgesic consumption when buprenorphine-soaked gelfoam was applied.[9] Gibson et al., reported that 60% of patients did not require supplemental analgesia on the day of surgery when gelfoam soaked in morphine was placed in the epidural space.[8] However, in that study, dose of morphine injected in the gelfoam was not standardized and methylprednisolone was also injected in the gelfoam. Epidural steroids reduce inflammation at nerve roots thereby decreasing the pain.[25]

Patients in the group I had lower sedation scores conceivably because patients in group II had higher consumption of IV morphine. Similarly patients in group I could be ambulated earlier and discharged from the PACU. The explanation for earlier discharge from PACU was that most patients in group I were pain-free and alert. This is important because earlier ambulation lifts the morale of the patient, relieves the surgeon's anxiety and also reduces the chances of postoperative complications. There was no significant difference in time to hospital discharge between the two groups, hence the benefit provided by gelfoam-soaked morphine does not translate into earlier hospital discharge rate.

MAP and heart rate were lower in the group I compared to the group II pointing towards a better pain control in group I. One patient in each group had heart rate below 50/min, however, in both the cases the baseline heart rate of patient was low and this bradycardia did not cause any symptoms or required any treatment.

The incidence of other adverse effects in our study was small with 29% patients reporting pruritis, however, this pruritis was mild to moderate and rarely disturbing to the patient or required any treatment. Nausea and vomiting were more in group II as compared to group I though the difference was statistically not significant. This could be due to greater consumption of supplemental analgesia in group II. It is of significance that none of the patients in either of groups had any pressure symptom. This is extremely important because compression of nerve roots can cause motor or sensory deficits in the area of distribution. At our institution, it is a routine practice to place gelfoam at the end of spine surgery so the absence of any neurological deficits was reassuring.

Epidural morphine acts on receptors in the dorsal horn instead of spinal nerve roots thus causing pain relief without sympathetic or motor blockade. This is of relevance after spine surgery, in which motor blockade as seen after epidural local anesthetic can interfere with the postoperative neurologic assessment and delay mobilization.[26] Early mobilization, minimizes complications due to pain and immobility.

Hydrophilic drugs like morphine administered via the epidural route can cause delayed respiratory depression.[27] However, we did not observe any case of respiratory depression. The possible explanation could be that dose of morphine used by us was small. Thus cerebro spinal fluid (CSF) concentration of morphine remains less and thus there was little cranial ascent.

The rational for choosing gelfoam as an extended delivery device was that gelfoam has the capability to absorb fluid several times its weight. Free opiates placed at the time of the surgery are diluted by blood and tissue fluids, are lost into the disc space, and are absorbed systemically, thus limiting its duration of action.[19] Morphine soaked in gelfoam thus remains in epidural space for greater time. It is possible that morphine is slowly released from gelfoam so its dilution by blood and tissue fluids is reduced accounting for its longer duration of action.[9]

Some authors have tried using even smaller doses of epidural morphine in an attempt to reduce the incidence and severity of adverse effects. Wu et al., used 1 mg doses of morphine using a microfibrillary paste, however, using such smaller doses also reduces duration of analgesic effect.[28] Future studies are needed to arrive at the most effective dose of morphine for its application in gelfoam. One limitation of our study was that we did not have long term follow-up of patients to evaluate for chronic pain and the for any pressure symptoms due to epidural gelfoam.

Conclusion

To conclude, epidural administration of morphine using gelfoam as a vehicle for sustained release of morphine is a safe and effective method of postoperative analgesia and results in a prolonged analgesic duration devoid of any side effects.