Secondary Logo

Journal Logo

Original Articles – Obstetrics and Regionals

The effective duration of analgesia after intrathecal morphine in patients without additional opioid analgesia: a randomized double-blind multicentre study on orthopaedic patients

Gehling, MHGa; Luesebrink, Ta; Kulka, PJb; Tryba, Ma

Author Information
European Journal of Anaesthesiology (EJA): August 2009 - Volume 26 - Issue 8 - p 683-688
doi: 10.1097/EJA.0b013e328329b045
  • Free



Intrathecal morphine added to a spinal anaesthesia reduces acute pain after orthopaedic surgery [1]. Effective analgesia can be obtained with doses ranging between 0.01 and 0.5 mg [1–5]. The addition of intrathecal morphine resulted in a significant reduction in morphine requirement via patient-controlled analgesia (PCA) and an increase in the time until first opioid request [5]. Some patients did not demand any systemic opioid during the first 24 h after intrathecal morphine [6–9]. In particular, the long duration of analgesia made intrathecal morphine a valuable tool in postoperative pain management after surgery in patients with spinal anaesthesia. The application of PCA devices may be unnecessary in patients who do not need systemic opioids due to a long lasting analgesia after intrathecal morphine.

However, adding 0.5 mg morphine to spinal anaesthesia reduced respiratory frequency, indicating an increased risk of respiratory depression [10]. With lower doses, this complication decreases [1,6,8,11]. Several dose-ranging studies [5,10,12,13] looked into the dose of intrathecal morphine that provides sufficient analgesia without an increased risk of respiratory depression.

However, data on the duration of analgesia beyond 24 h are lacking so far. For the purpose of this study, the effective time of analgesia was defined as the duration of the postoperative period without pain that needed opioids.

To optimize the results of our study, the administration of metamizol for nonopioid analgesia was part of the postoperative pain management. A Kaplan–Meier analysis of the time until first opioid request provided information on the time course of the ratio of patients without opioids.


A randomized controlled multicentre study was conducted at six teaching hospitals in Germany in 2002 and 2003. The ethics committee of each participating centre approved the study protocol. Written, informed consent was obtained from all included patients the day before surgery.


Patients older than17 years of age, ASA physical status I–III and who were scheduled for elective spinal anaesthesia were eligible if they had no clotting abnormality, infection at the injection site, a known sleep apnoea, pregnancy or addiction. A premedication with up to 0.2 mg kg−1 bodyweight dipotassium chlorazepate, a benzodiazepine, was allowed in study patients. This dose was chosen because it had a good anxiolytic and weak sedative effect.


Spinal anaesthesia was induced at spinal levels L3/4 or L4/5 with 15 mg bupivacaine and the study medication, which was morphine sulphate (Mundipharma GmbH, Limburg Germany) in the morphine groups.

Intraoperatively, heart rate, blood pressure and oxygen saturation were monitored. On request, sedation was performed with 0.5–3.0 mg kg−1 h−1 propofol. After surgery, the patients were monitored for 1–3 h in a postanaesthesia care unit (PACU) until muscular force in the lower extremities returned to almost normal strength. Patients were returned to the ward when they had normal values of cardiorespiratory function and no relevant sedation. In the ward, heart rate, blood pressure, respiratory frequency and sedation score were monitored every hour until the next morning. If respiratory frequency was of 10 min−1 or less, patients were referred to the ICU for monitoring and therapy.

After surgery, all patients with pain received a 30 min infusion of 1000 mg metamizol that could be repeated every 4 h with a maximum daily dose of 6000 mg.

If no sufficient analgesia could be achieved in the PACU, the patients received 5 mg morphine intravenously, repeated every 10 min if necessary.

In the ward, the patients received a nurse-controlled analgesic regimen with the opioid used for postoperative analgesia in each hospital. If sufficient analgesia could not be obtained, the study was stopped and an individual analgesic regimen was commenced. PCA, epidural analgesia or peripheral regional blocks were then applied at the discretion of each study centre.

Respiratory depression is the most important complication of intrathecal morphine. The risk of respiratory depression was monitored by measuring the respiratory frequency. Patients with fewer than 10 breaths per minute were referred to an ICU. They received naloxone according to a protocol.


The objective of this study was the effective time of analgesia as a function of intrathecal morphine dose. For the purpose of this study, we defined effective analgesia as the period of time without an additional opioid analgesia request, because this is the length of time that additional PCA would have been unnecessary.


We hypothesized that 0.1 and 0.2 mg intrathecal morphine are effective for postoperative pain control after surgery performed in spinal anaesthesia. Effective pain control was defined as an increase in the number of patients not requiring additional opioid analgesia until the first morning after surgery.

Outcome parameter

The primary outcome parameter was the number of patients without additional opioid request at any time during the 72 h observation. Secondary parameters were pain intensity on a visual analogue scale (VAS) or numerical rating scale (NRS), side effects and complications. The primary analysis compared the results of each morphine group with the control group. In a secondary step, we also analysed differences between the two morphine groups.

The morning after surgery, adverse events were documented with a standardized questionnaire. Side effects were rated by the patients on a NRS from 0 indicating not at all to 10 indicating worst possible side effect. The need for urinary catheterization was defined as urinary retention and an NRS of 10 (patients who could not pass urine at all).

The frequency of respiratory depression was derived from a monitoring protocol that listed the breathing rate every hour.

Sample size

Based on a pilot study, we expected a rate of opioid requests of 20% in patients with 0.1 or 0.2 mg intrathecal morphine and 80% in the placebo group [14]. For a type I error of α of 0.10 and a type II error of β of 0.05, a significance level of P value less than 0.05 and a drop out rate of 10%, we calculated a sample size of 64 patients in each group.


After inclusion in the study, a consecutive case number for each patient was documented. On the basis of a computer-generated randomization list, we allocated patients randomly to receive placebo, morphine 0.1 mg or morphine 0.2 mg in addition to 15 mg bupivacaine intrathecally. Each participating clinic received a randomization list of its own.

An anaesthetist otherwise not involved in the study prepared the study medication outside the operating room. The anaesthetist inducing anaesthesia, the patients and study doctors all were blinded to the study group. The randomization was finally disclosed with the statistical analysis. For safety reasons, a complete randomization list was available to the consultant anaesthetist in charge at any time of the study.


Dichotomous parameters were analysed with the χ2 test or Fisher's exact test if appropriate. Continuous numerical data were analysed with an analysis of variance (ANOVA). A survival analysis of the time until first opioid request was performed using a Kaplan–Meier calculation. Statistical significance was defined as P value less than 0.05.


Demographic data

In this multicentre study, 188 patients were included and randomized. We obtained datasets of all included patients (Fig. 1). However, in some patients, the documented data were incomplete.

Fig. 1
Fig. 1

As we did not stratify the groups according to a planned type of surgery, there are differences in the type of surgery between the groups (Table 1).

Table 1
Table 1:
Patients' characteristics


The primary end point of our study was the frequency of additional opioid requests. In the placebo group, many patients required opioid analgesia immediately after surgery (Fig. 2). Eight hours after surgery, 50% had received additional opioids. The median time until opioid request was 7 h. In the placebo group, 20% did not ask for any opioid analgesia in addition to the applied scheme with metamizol.

Fig. 2
Fig. 2

Patients with 0.1 mg intrathecal morphine showed a significant reduction in opioid requests. Only 25% in the 0.1 mg morphine group needed additional systemic opioids in the first 8 h after surgery. The median time until opioid request increased to 24 h with 0.1 mg intrathecal morphine (Table 2).

Table 2
Table 2:
Proportion of patients with opioid request

In the group with 0.2 mg intrathecal morphine, only 10% required systemic opioids 8 h after surgery. The median time until first opioid request was 45 h in this group.

The Kaplan–Meier curve shows the time course of patients censored for opioid request over a period of 72 h after surgery. Both active treatments were associated with a significant reduction in opioid required by patients compared with placebo (P = 0.0001). Analysing significance levels after 12, 24, 36, 48 and 72 h provides more detailed information on differences between the two active groups (Fig. 2). During the first 12 h after surgery, 0.1 and 0.2 mg intrathecal morphine resulted in similar reductions in the number of patients with additional opioid for postoperative analgesia (P = 0.0574). However, 24 and 36 h after surgery, 0.2 mg of intrathecal morphine reduced the number of patients requiring opioids compared with 0.1 mg morphine. The difference was statistically significant only at 24 h (P < 0.05).

The reduction in opioids in the active treatment groups was not associated with increased levels of pain intensity (Fig. 3). Pain during movement was significantly lower in patients receiving intrathecal morphine between 1 and 8 h after surgery (P < 0.01). Both active treatment groups did not differ significantly (P > 0.05).

Fig. 3
Fig. 3

Metamizol was a standard nonopioid analgesic in this study. During the first 24 h after surgery, the placebo patients received (mean ± SD) 2352 ± 1033, patients in the 0.1 mg group 1977 ± 1111, and in 0.2 mg group 1878 ± 928 mg.


After surgery, placebo patients showed a trend towards increased respiratory frequency (Fig. 4). We found no differences in oxygen saturation between the study groups during 72 h after application of intrathecal morphine.

Fig. 4
Fig. 4

A respiratory frequency equal to or less than 10 min−1 was observed in one patient in the placebo group and in one patient in the morphine 0.1 group, and in two patients in the 0.2 mg morphine group. All episodes of decreased respiratory frequency were observed within 8 h after induction of anaesthesia. The patients were then monitored in an ICU. They received supplemental oxygen, but no one needed naloxone or ventilation.

Side effects

Patients receiving intrathecal morphine had no significant increase in the intensity of sedation or nausea. However, intrathecal morphine does dependently increased the intensity of pruritus compared with placebo (Fig. 5, P < 0.05). Between 1 and 3 h after surgery, the intensity of pruritus in patients with 0.2 mg morphine was significantly increased compared with patients in the placebo and patients with 0.1 mg morphine groups (Fig. 5).

Fig. 5
Fig. 5


In this randomized double-blind placebo-controlled multicentre study, intrathecal morphine was associated with better postoperative analgesia in patients receiving a nonopioid analgesic on a regular basis.

Intrathecal morphine decreased the number of patients requiring opioids after surgery during the study period. The present study showed a significant reduction in opioid requests in patients receiving 0.2 mg compared with 0.1 mg intrathecal morphine at 24 h after surgery. We conclude that 0.2 mg morphine increased the duration of effective analgesia compared with 0.1 mg significantly in our patients. Thus, intrathecal morphine dose-dependently increased the ratio of patients without additional opioids for 72 h after surgery.

Intrathecal morphine resulted in more intensive pruritus, but no differences in sedation and nausea scores. In general, no decreased respiratory function was observed. However, one patient with placebo or 0.1 mg intrathecal morphine and two patients with 0.2 mg intrathecal morphine needed continuous monitoring because of an episode of bradypnoea with a respiratory frequency of 10 min−1 or less. The presented data do not have the power to describe risks and side effects precisely.

Although we investigated a heterogeneous sample of different operations, our results represent primarily the situation for orthopaedic patients.

In the first dose–response study [10], 1 and 2.5 mg of intrathecal morphine were associated with profound respiratory depression. Unfortunately, analgesia after 0.3 mg was unsatisfactory in 30% of orthopaedic patients [10]. Therefore, the authors did not recommend any ideal dose of intrathecal morphine. In another study, 0.025–0.2 mg intrathecal morphine was compared in patients undergoing total hip replacement [5]. In this investigation, 0.1 and 0.2 mg intrathecal morphine decreased total morphine consumption over 24 h after surgery from 24 mg to 10 mg per day [5]. Stratifying investigation groups for hip or knee replacement confirmed the efficacy of low-dose intrathecal morphine in hip but not in knee surgery [12].

In the former studies [5,10,12], nonopioid analgesics were not administered. As several studies showed better analgesia with intrathecal morphine in patients with additional nonopioids, we investigated patients with metamizol on a regular basis. Metamizol is a potent nonopioid analgesic with an efficacy similar to diclofenac or ibuprofen in acute pain. Metamizol can be given intravenously. Some doctors refuse to use the substance because of the risk of agranulocytosis that was discussed in the context of metamizol applications.

Our results show a dose-dependent analgesia in a heterogeneous group of patients, including knee replacement.

In another study [13], 0.2 and 0.5 mg intrathecal morphine were compared in patients after knee replacement. In this investigation, all patients received paracetamol and diclofenac on a regular basis. The dose of 0.5 mg intrathecal morphine increased the time to first opioid request significantly compared with 0.2 mg morphine from 10 to 24 h without differences in the incidence of side effects [13]. As the median tramadol consumption in both groups was rather low, the question is raised of whether or not the difference of 0.5 mg intrathecal morphine (0 mg) compared with 0.2 mg (100 mg) was clinically important. Unfortunately, no data beyond 24 h are available.

Another clinically important factor may be the start of mobilization, because mobilization results in more intensive pain. However, in our study, patients with intrathecal morphine did not document increased pain intensity during movement.

Our study showed a significant prolongation of effective analgesia with 0.2 mg intrathecal morphine compared with 0.1 mg morphine or placebo over a period of 72 h.

In elderly patients (>65 years), 0.1 and 0.2 mg resulted in equal levels of analgesia [8]. No significant differences in postoperative nausea, sedation and respiratory depression were observed in this study [8]. Intrathecal morphine was associated with increased pruritus [8]. These results are comparable to our observations of side effects.

Several studies [5,10,12,13] demonstrated a reduced morphine consumption after intrathecal morphine in orthopaedic patients during the first 24 h after operation. Our investigation documents the duration of effective analgesia defined as time without opioid request. Both active groups had a significantly lower number of patients requiring opioids compared with placebo patients. Compared with 0.1 mg morphine, the dose of 0.2 mg intrathecal morphine resulted in a significant reduction in the number of patients asking for opioids over 24 h, but not thereafter. The advantage of 0.2 mg over 0.1 mg intrathecal morphine seems to be an increased time of effective pain control without additional opioids in patients receiving metamizol regularly. The type of surgery, comedication with nonopioid analgesics and an early start of functional rehabilitation, that is, movement, can influence the analgesic efficacy of intrathecal morphine. Therefore, the dosing of intrathecal morphine should not only be based on small statistical differences found in this study but also on clinical factors.

There are two conclusions that can be drawn from dose-ranging studies of intrathecal morphine.

  1. In the majority of studies, a dose of 0.1–0.2 mg morphine provided clinically important analgesia. The lowest effective dose of intrathecal morphine should be applied to avoid side effects and risks.
  2. In patients receiving analgesia with a combination of intrathecal morphine and systemic nonopioid analgesia, the use of a PCA device could be avoided.


The following persons and institutions participated in the study:

Dr B. Bang-Vojdanovski, Orthopaedische Klinik Kassel; Professor Dr H. Harke, Klinikum Krefeld; Dr Reinhard Sittl, Institut für Anästhesiologie, Universitätskliniken Erlangen/Nürnberg; Professor Dr I Linde, Annastift Hannover; Th.L., Dr M.G., Professor Dr Tryba, Klinikum Kassel; Professor Dr Zenz, Universitätsklinik Bergmannsheil Bochum.

The study was funded by an unrestricted grant from Mundipharma GmbH. No author has a conflict of interest related to the study.


1 Kalso E. Effects of intrathecal morphine, injected with bupivacaine, on pain after orthopaedic surgery. Br J Anaesth 1983; 55:415–422.
2 Drakeford MK, Pettine KA, Brookshire L, Ebert F. Spinal narcotics for postoperative analgesia in total joint arthroplasty. A prospective study. J Bone Joint Surg Am 1991; 73:424–428.
3 Fogarty DJ, Milligan KR. Postoperative analgesia following total hip replacement: a comparison of intrathecal morphine and diamorphine. J R Soc Med 1995; 88:70–72.
4 Niemi L, Pitkanen MT, Tuominen M, Rosenberg PH. Comparison of intrathecal fentanyl infusion with intrathecal morphine infusion or bolus for postoperative pain relief after hip arthroplasty. Anesth Analg 1993; 77:126–130.
5 Slappendel R, Weber EW, Dirksen R, et al. Optimization of the dose of intrathecal morphine in total hip surgery: a dose-finding study. Anesth Analg 1999; 88:822–826.
6 Grace D, Bunting H, Milligan KR, Fee JP. Postoperative analgesia after co-administration of clonidine and morphine by the intrathecal route in patients undergoing hip replacement. Anesth Analg 1995; 80:86–91.
7 Johnson A, Bengtsson M, Soderlind K, Lofstrom JB. Influence of intrathecal morphine and naloxone intervention on postoperative ventilatory regulation in elderly patients. Acta Anaesthesiol Scand 1992; 36:436–444.
8 Murphy PM, Stack D, Kinirons B, Laffey JG. Optimizing the dose of intrathecal morphine in older patients undergoing hip arthroplasty. Anesth Analg 2003; 97:1709–1715.
9 Schaer H, Baasch K, Prochacka K. Intrathecal morphine for postoperative pain. Anaesthesist 1992; 41:689–693.
10 Jacobson L, Chabal C, Brody MC. A dose–response study of intrathecal morphine: efficacy, duration, optimal dose, and side effects. Anesth Analg 1988; 67:1082–1088.
11 Tan PH, Chia YY, Lo Y, et al. Intrathecal bupivacaine with morphine or neostigmine for postoperative analgesia after total knee replacement surgery. Can J Anaesth 2001; 48:551–556.
12 Rathmell JP, Pino CA, Taylor R, et al. Intrathecal morphine for postoperative analgesia: a randomized, controlled, dose-ranging study after hip and knee arthroplasty. Anesth Analg 2003; 97:1452–1457.
13 Bowrey S, Hamer J, Bowler I, et al. A comparison of 0.2 and 0.5 mg intrathecal morphine for postoperative analgesia after total knee replacement. Anaesthesia 2005; 60:449–452.
14 Gehling M, Tryba M. Intrathecal morphine in orthopaedic surgery patients. Optimised dose in patients receiving dipyrone. Anaesthesist 2008; 57:347–354.

morphine; postoperative pain; spinal anaesthesia

© 2009 European Society of Anaesthesiology