Postoperative analgesia by epidural methylprednisolone after posterolateral thoracotomy

Y. Blanloeil*, P. Bizouarn, Y. Le Teurnier, C. Le Roux, J.C. Rigal, E. Sellier and B. Nougarède

Service d’Anesthésie et de Réanimation Chirurgicale, Hôpital G et R Laënnec, Boulevard J Monod, CHU Nantes, F-44093 Nantes Cedex 01, France*Corresponding author

Accepted for publication: April 10, 2001


    Abstract
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
The aim of this study was to evaluate the potential analgesic effect of epidural methylprednisolone (MP) after posterolateral thoracotomy (PLT). Adult male patients undergoing PLT for lung surgery were included in a prospective, randomized, double blind study. Peroperative analgesia (bupivacaine plus sufentanil) was given by a thoracic epidural catheter associated with general anaesthesia. After surgery, patients received either MP 1 mg kg–1 followed by a continuous epidural infusion of MP 1.5 mg kg–1 during 48 h (MP group) or 0.9% saline as a bolus injection and continuous epidural infusion (P group). Additional morphine analgesia was administered by i.v. patient-controlled analgesia. Pain was assessed at rest and with mobilization every 4 h after operation during 48 h with a visual analogue scale (VAS). The primary end-point was the total morphine requirements during the 48 first postoperative hour. Twenty-four patients were allocated to MP (n=12) and P (n=12) groups. Characteristics of the two groups were similar. There were no differences between groups for morphine requirements (median and interquartile range) during the 48 h: 59 mg (40–78) in MP group vs 65 mg (59–93) in P group. There were no differences between groups for morphine requirements every 4 h during the 48 h and VAS for pain at rest and evoked pain. No side effects were reported. It was concluded in this small study that these results did not support the use of epidural steroids for postoperative analgesia after PLT.

Br J Anaesth 2001; 87: 635–8

Keywords: analgesia, postoperative; analgesic techniques, epidural; analgesics anti-inflammatory, steroid; surgery, thoracic


    Introduction
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
Posterolateral thoracotomy (PLT), is one of the most painful surgical procedures.1 Various techniques have been developed and evaluated to treat it, particularly epidural analgesia.1 Epidurally administered glucocorticoids have been used in rheumatology since the 1960s for treatment of the chronic radicular pain of sciatica as a result of herniated nucleus pulposus.2 Their more recent use in the treatment of acute pain after laminoarthrectomy has suggested a satisfactory analgesic effect during the postoperative period.3 4 In this situation, glucocorticoids act probably at the surgical site by a local anti-inflammatory effect. But glucocorticoids may have a central effect.5 We conducted a prospective randomized, double blind, placebo-controlled clinical study to determine whether methylprednisolone (MP) improved analgesia when administered epidurally in this after PLT.


    Methods and results
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
After approval by the Ethics Committee and written informed consent, 27 adult male patients undergoing lobectomy or pneumonectomy for cancer, were included. The same team surgeon operated all patients by PLT including preservation of the serratus anterior muscle.

The primary end-point was the total i.v. morphine requirements by patient-controlled analgesia (PCA) pump during the first postoperative 48 h. Secondary end-points were the 4-hourly interval morphine requirements during 48 h, visual analogue scale (VAS) pain assessment at rest and mobilization, quality of pain relief appreciated by the patient himself and the incidence of adverse events.

During the preoperative visit, the patient was instructed in the use of the VAS and PCA use. After premedication with hydroxyzine 1.5 mg kg–1, an epidural catheter was inserted at T4–T5 or T5–T6. A mixture of 100 mg bupivacaine with 0.1 mg epinephrine, 20 µg sufentanil and saline solution (total volume of 30 ml), was injected at a rate of 5 ml every 5 min until a block was obtained extending from T1 to T10. General anaesthesia was induced with thiopental 6 mg kg–1 and pancuronium bromide 0.1 mg kg–1, orotracheal intubation was performed after local anaesthesia of the glottis, using a double lumen 39–41 tube. Anaesthesia was maintained by inhalation of nitrous oxide and isoflurane. Patients were assigned randomly to two groups the day before surgery. At the end of the operation, patients in the MP group received epidural administration of 1 mg kg–1 of MP diluted in 10 ml of saline solution followed by a continuous epidural infusion of 1.5 mg kg–1 of MP diluted in 24 ml of saline solution. Patients in the placebo (P) group received identical volumes of saline solution epidurally, followed by a continuous epidural injection at the same rate. The various solutions were administered in double blind manner and maintained for 48 h postoperatively.

An initial morphine-loading dose was given when the patient complained of pain (VAS >3 cm). Titration of 1 mg boluses followed until pain was relieved (VAS <3). Subsequently, the PCA pump was set on PCA mode alone (10-min lockout, 1 mg bolus and no maximal dose).

Haemodynamic data, ventilatory frequency, SpO2 and level consciousness were evaluated every hour for 48 h. Consciousness was scored on a scale of 1–5 (1= asleep and not wakeable; 2=asleep but wakeable with altered vigilance; 3=asleep but wakeable and vigilant; 4=spontaneously awake; and 5=restless). The patient evaluated his pain (at rest and on coughing) every 4 h using a VAS (0=absence of pain; 10=unbearable pain). The patient evaluated the quality of relief of pain using a questionnaire insufficient, good, excellent at 72 h postoperatively. Adverse effects were recorded during the entire period.

Data are presented as mean (SD) or median with interquartile range. Statistical analysis was performed using the Mann–Whitney rank sum test for comparison of quantitative variables. Multivariate analysis of variance (MANOVA) for repeated measurements was used to compare VAS pain scores and morphine use over time between the groups or within each group throughout the study (Systat IM 7.0 software). The significance threshold for the tests was set at P<0.05.

Three patients were excluded because of misunderstandings with respect to the use PCA or technical failure. Therefore, 24 patients were included in the statistical analysis, 12 in each group. There were no significant differences between the two groups with respect to patient, surgical and anaesthetic characteristics.

Total morphine requirements during the study were not significantly different between the two groups (median 59 (40–78) for MP group and median 65 (59–93) for P group) (Fig. 1).



View larger version (25K):
[in this window]
[in a new window]
 
Fig 1 Box-plots of 4 hourly morphine requirements during the first postoperative 48 h. No significant differences between groups (P=0.197).

 
VAS pain scores at rest and for evoked pain did not differ between groups (P=0.59 and 0.14 for pain at rest or evoked pain, respectively). VAS pain scores at rest were different across time in each group (P=0.029) but no interaction between groups and time was observed. VAS pain scores after evoked pain did not differ significantly within each group throughout the study (P=0.25). Patient satisfaction was not different between groups. There was one case of pruritus and sedation in the MP group, none in P group and two and three cases of nausea and vomiting, respectively, in MP and P groups. No patient experienced bradypnea or required naloxone.


    Comment
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
This prospective, randomized, double blind study showed no significant difference in morphine requirement and pain intensity after epidural administration of MP for the treatment of postoperative PLT pain compared with placebo.

The conclusions of studies concerning the analgesic efficacy of glucocorticoids are controversial. Some data support a central effect of glucocorticoids.5 Other studies have shown a lack of efficacy.6 Epidural glucocorticoids have been tested for postoperative pain after spinal surgery. Dexamethasone reduced postoperative lumbar pain after laminoarthrectomy.3 Epidural administration of methylprednisolone alone or in association with morphine decreased postoperative pain significantly after surgery on spinal stenosis, but not after disk surgery.4 However, it should be noted that methylprednisolone in spinal surgery acts at the surgical site essentially by a local anti-inflammatory effect. In postoperative pain as a result of PLT, this effect is probably less important.

It is concluded that our data do not support the use of epidural steroids for postoperative analgesia after PLT.


    References
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
1 Kavanagh BP, Katz J, Sandler AN. Pain control after thoracic surgery. Anesthesiology 1994; 81: 737–59[ISI][Medline]

2 Carette S, Marcoux S. Epidural corticosteroid injection for sciatica due to herniated nucleus pulposus. N Engl J Med 1997; 336: 1634–40[Abstract/Free Full Text]

3 Ang ET, Goldfarb G, Kobn S, et al. Analgésie postopératoire: injection péridurale de phosphate sodique de dexaméthasone. Ann Fr Anesth Réanim 1988; 7: 289–93[Medline]

4 Mc Neil TW, Anderson GB, Schell B, Sinkora G, Nelson J, Lavender SA. Epidural administration of methylprednisolone and morphine for pain after a spinal operation. A randomized, prospective comparative study. J Bone Joint Surg (Am) 1995; 77: 1814–8[Abstract]

5 Hall ED. Glucocorticoids effects on central nervous excitation and synaptic transmission. Int Rev Neurobiol 1982; 23: 165–95[ISI][Medline]

6 Abram SE, Marsala M, Yaksh TL. Analgesic and neurotoxic effects of intrathecal corticosteroids in rats. Anesthesiology 1994; 81: 1198–1205[ISI][Medline]