Abstract

Objective

To compare the analgesic efficacy and side effects of magnesium and fentanyl as an additive to intrathecal bupivacaine.

Methods

Ninety adult patients scheduled for femur surgery under spinal anesthesia were randomly allocated to one of the following three groups to receive intrathecally: bupivacaine 15 mg combined with 0.5 mL magnesium 10%; bupivacaine 15 mg combined with 0.5 mL fentanyl; or bupivacaine 15 mg combined with 0.5 mL distilled water (control). The time to first analgesic request, sensory and motor blockade onset time, duration of sensory and motor blockade, analgesic requirement in the first 12 hours after surgery, and the incidences of hypotension, bradycardia, hypoxemia and ephedrine were recorded.

Results

Magnesium caused a significant delay in the onset of both sensory and motor blockade compared with the fentanyl (95% CI 3 to 4; p < 0.001) and control (95% CI 3.5–5; p < 0.001) groups. The duration of spinal analgesia in group F (fentanyl) was significantly greater than in group C (control) (95% CI 365–513; p < 0.001) and group M (magnesium) (95% CI 385–523; p < 0.001). The total amount of methadone consumption over 12 hours was significantly lower in the magnesium and fentanyl groups than in the control group (5 mg vs. 5.666 ± 1.728 mg; p = 0.04).

Conclusion

Addition of intrathecal magnesium sulfate to spinal anesthesia induced by bupivacaine significantly prolonged the onset of both sensory and motor blockade compared with fentanyl. Although magnesium failed to prolong the time to first analgesic requirement as seen with fentanyl, it reduced the total consumption of opioids in the first 12 hours postoperatively compared with the control group.

Keywords

magnesium; fentanyl; injections, spinal; pain;

---

1. Introduction

Co-administration of small doses of opioids and bupivacaine for spinal anesthesia reduces the postoperative analgesic requirements. However, the simultaneous use of these drugs is limited because of significant adverse effects such as pruritus, hemodynamic instability, respiratory depression, urinary retention, nausea and vomiting.1, 2 Adequate pain management to facilitate rehabilitation and to accelerate functional recovery after femur surgery is essential to enable patients to resume normal activity as soon as possible. Postoperative pain is associated with neuroendocrine responses, catecholamine release, and increased morbidity and the central sensitization is believed to be among the mechanisms implicated in the persistence of postoperative pain.3, 4 Central sensitization has been shown to depend on the activation of dorsal horn N-methyl-D aspartate (NMDA) receptors by excitatory amino acid transmitters such as aspartate and glutamate.4, 5 It has been reported that the μ-opioid receptor activation leads to a sustained increase in glutamate synaptic effectiveness at the NMDA receptor level and is associated with central hypersensitivity to pain.⁶ Single injection of the μ-opioid receptor agonist heroin is described to induce a reduction in nociceptive threshold (hyperalgesia) for several days in rats.⁷ This phenomenon is inhibited by the NMDA-receptor antagonist.⁷ Magnesium blocks the NMDA channels in a voltage-dependent fashion.⁸ It has been suggested that the NMDA receptor antagonists induce preemptive analgesia when administrated before tissue injury occurs, because these can prevent the induction of central sensitization from peripheral nociceptive stimulation.9, 10 Also, magnesium is called nature's physiological calcium channel blocker which is reported to have antinociceptive effects in rats and in morphine-potentiating chronic pain patients.11, 12, 13 

The safety of intrathecal magnesium (Mg) has been extensively evaluated in animals and humans.14, 15, 16, 17, 18, 19, 20, 21, 22 In previous studies, it was demonstrated that the intrathecally (IT) administered Mg⁺² could prolong the spinal opioid analgesia both in rats and humans.19, 20, 21, 22 Also it has been shown that an intrathecal injection of a large dose of magnesium sulfate (MgSO₄) (1260 mg) causes a complete sensory and motor block (e.g. spinal anesthesia) lasting for 60 minutes.²³ In contrast, in some studies, the administration of magnesium was found to have no direct benefit in pain or analgesic reduction.24, 25

Our hypothesis was that the addition of low-dose MgSO₄ without opioid supplements to spinal bupivacaine may potentiate the analgesic effect of opioids to be used postoperatively, without prolonging the duration of sensory and motor block. Therefore, it would cause a similar reduction of analgesic, equivalent postoperative pain relief, and lower the ambulation time and side effects compared with intrathecal fentanyl. To test our hypothesis, we designed this randomized double-blind, placebo-controlled study to compare the analgesic efficacy and side effects of magnesium with fentanyl as an additive to IT bupivacaine in patients undergoing lower limb orthopedic surgery.

2. Methods

This trial was registered with the United States National Institutes of Health at www.clinicaltrials.gov, under the number NCT01205997. Following the approval by the institutional research ethics board and obtaining informed patient consent, ninety patients 20–60 years old with American Society of Anesthesiologists (ASA) physical status I or II, scheduled for femur surgery under spinal anesthesia, were studied in a prospective, double-blinded, randomized way. The Consolidated Standards of Reporting Trials (CONSORT) recommendations for reporting the randomized, controlled clinical trials²⁶ were followed (Fig. 1).

Exclusion criteria were significant co-existing conditions such as hepatorenal and cardiovascular diseases, any contraindication to regional anesthesia like local infection or bleeding disorders, allergy to opioids, long-term opioid use, and a history of chronic pain. Patients were randomly allocated to one of three groups of 30 subjects each. Blinding was achieved through the use of equal amounts of drugs (3.5 mL) while the syringes used were labeled as A, B or C, according to their content. Identical coded syringes, prepared by people not involved in the study, were randomly handed to the anesthetists, who were unaware of the identities of the drugs. The magnesium group (group M) received intrathecally 15 mg bupivacaine (0.5% hyperbaric solution) combined with 0.5 mL of 10% MgSO₄; the fentanyl group (group F) received 15 mg bupivacaine (0.5% hyperbaric solution) combined with 0.5 mL (25 μg) fentanyl; and the control group (group C) received 15 mg bupivacaine (0.5% hyperbaric solution) combined with 0.5 mL distilled water. Time to the first requirement of analgesic supplement, sensory block onset time, onset of motor block, maximal level of sensory block, duration of blockade, the incidence of hypotension, ephedrine requirement, bradycardia, hypoxemia (SpO₂ < 90%), postoperative analgesic requirements, and adverse events such as sedation, dizziness, pruritus, nausea, and vomiting were recorded. All patients received an intravenous preload of 5–7 mL/kg lactated Ringer's solution before the subarachnoid block. With aseptic technique, a 25-gauge Quincke needle was inserted intrathecally via the L4-5 interspaces by midline approach with the patient in the sitting position. After a successful dural puncture, the anesthetic solution was injected. In this study, the postoperative analgesia in terms of definition was time to first requirement of analgesic supplement from the time of injection. No additional analgesic was administered unless requested by the patient. Sensory block was assessed by a pin-prick test. The onset of sensory block was defined as the time between the end of injection of the intrathecal anesthetic and the absence of pain at the T10 dermatome. The duration of sensory block was defined as the time for regression of two segments from the maximum block height evaluated by pin prick. The maximal level of sensory block was evaluated by pin prick 20 minutes after the completion of injection. Motor block was assessed by the modified Bromage score (0, no motor loss; 1, inability to flex the hip; 2, inability to flex the knee; and 3, inability to flex the ankle). The onset of motor block was defined as the time from intrathecal injection to Bromage block 1, whereas the duration of motor block was assumed when the modified Bromage score was 0.

The duration of spinal anesthesia was defined as the period from spinal injection to the first time when the patient complained of pain in the postoperative period. Patients were instructed preoperatively to use the verbal rating scale (VRS) from 0 to 10 (0, no pain; 10, maximum imaginable pain) for pain assessment. If the VRS exceeded 4 and the patient requested a supplement analgesic, methadone 5 mg was given intravenously, for postoperative pain relief. The quality of anesthesia during surgery was judged by an anesthetist at the end of surgery as excellent (no pain or sensation), good (mild pain or discomfort), fair (mild discomfort and in need of analgesia), and poor (patient in moderate or severe pain and in need of general anesthesia). Systolic and diastolic blood pressures (SBP, DBP), heart rate (HR), and peripheral oxygen saturation (SpO₂) were recorded by an anesthetist blinded to the patient group 5 minutes before the intrathecal injection and 2, 4, 6, 8, 10, 15 and 20 minutes after the injection. If the SBP was 20% below the baseline (measured in the ward) or less than 90 mmHg, ephedrine 5 mg was administered intravenously. Also, if the HR was less than 50 beats/minute, a dose of 0.5 mg of atropine sulfate was administered intravenously. A follow-up telephone call was made 24 hours after the surgery and again 1 month later, during which the patients were asked about the side effects and dysesthesia of the lower limbs or buttocks. 

A sample size of 20 patients per group was needed to detect a difference of at least 20% in fentanyl consumption (α = 0.01, two-sided, power = 90%) with two-sample t test.27, 28 We included 30 patients in each group to allow for drop-outs and protocol violations. Data were analyzed using SPSS (SPSS 15.0, SPSS Inc, Chicago, IL, USA). Continuous variables were tested for normal distribution by the Kolmogorov–Smirnov test. Parametric data were expressed as mean and standard deviation (SD) and analyzed using the independent t test. Non-parametric data were expressed as median and interquartile range (IQR) and analyzed using the Mann Whitney U test. The effect of time on hemodynamic parameters was analyzed using repeated measurement analysis of variance. The differences between the highest mean arterial pressure, heart rate, and the baseline pressure in each patient were compared between the study groups using the one-factor ANOVA test. Within each group, comparison with baseline was made using the paired Student t test with Tukey correction. The Chi-square test was used to analyze the incidence of adverse events. A p value < 0.05 was considered statistically significant.

3. Results

Among 105 patients initially enrolled in this study, 15 patients had to be excluded because of logistical reasons or violations of the study protocol. Ninety patients were included and randomly assigned to the treatment groups (Fig. 1).

There were no significant differences in sex, age, height and weight among the three groups. The duration of surgery was also similar (Table 1).

As shown in Table 2, the difference in mean onset time of sensory block between groups M and F (95% CI 3–4; p < 0.001) and C (95% CI 3.5–5; p < 0.001) was significant.

Table 2 also shows a significant difference in duration of sensory block between groups F and M (95% CI 148–162.5; p < 0.001) and C (95% CI 147.5–162.5; p < 0.001). The difference in duration of sensory block time between groups M and C was shown to be insignificant (p = 0.714). In addition, as demonstrated in Table 2, there was a significant difference in onset time of motor block between groups M and F (95% CI 4.5–6; p < 0.001) and C (95% CI 5–6.5; p < 0.001). Also, Table 2 is indicative of a significant difference in mean duration of motor blockade time between groups F and M (95% CI 140–155; p < 0.001) and C (C95% CI 137.5–150; p < 0.001) groups. Moreover, no significant difference in duration of motor block between groups M and C was found (p = 0.697). Patients who were given fentanyl demonstrated a significantly prolonged duration of anesthesia compared with the control (C95% CI 152.5–164; p < 0.001) and magnesium groups (C95%CI 152.5–165; p < 0.001) (Table 2). With the duration of anesthesia, there was a significant difference in mean time to first analgesic request between groups F and M (95% CI 385–523; p < 0.001) and C (C95%CI 365–513; p < 0.001) (Table 2). No significant difference in mean time to first analgesic request between groups M and C was observed (p = 0.333) (Fig. 2).

The quality of spinal analgesia was excellent in all group F patients, compared with 56.7% in group C and 56.7% in group M (p < 0.001). However, none of the patients in these groups was found to have a fair or poor quality of analgesia. The total analgesic consumption during the first 12 hours after surgery was devoid of any significant difference between groups F and M (p = 1.000). Also, the total amount of methadone consumption over 12 hours was significantly lower in the magnesium and fentanyl groups than in the control group (5 mg vs. 5.666 ± 1.728 mg; p = 0.04).

Transient hypotension occurred at various times in groups F and C despite a pre-block volume loading. These patients were treated with 5-mg boluses of ephedrine IV to maintain their SBP within 20% of baseline values or 90 mmHg. There was a significant difference in occurrence of hypotension episodes between groups F and M (p = 0.001) and C (p = 0.001), yet no significant difference in occurrence of hypotension was established between groups M and C (p = 0.52). Although none of the patients in group M showed a hypotensive episode requiring treatment (Table 3), the overall difference in ephedrine requirement between the three groups was insignificant, statistically (p = 0.632). In addition, no statistically significant difference in SpO₂ values between the three groups was demonstrated. As shown in Table 3, the two groups were significantly different in terms of intraoperative and postoperative side effects (p < 0.001).

There was no report of postdural puncture headache after surgery and also no patient in either group with any sensory or motor complications was identified in the first month following surgery.

4. Discussion

We have demonstrated that the addition of MgSO₄ (50 mg) to 15 mg of spinal bupivacaine (0.5%) hyperbaric prolonged the onset time of sensory and motor blockade without prolonging the duration of spinal anesthesia. Although magnesium failed to prolong the time to first analgesic requirement, as seen with fentanyl, it reduced the total opioids consumption in the first 12 hours postoperatively as compared with the control group.

In the current study, the onset of both sensory and motor blocks in the fentanyl group, as compared with groups C and M, were shorter while the duration of sensory and motor blocks were longer. This finding is in agreement with the previous report by Unlugence et al.²⁹ In the present study the onset time of sensory and motor blockade in the magnesium group were longer without prolonging the duration of spinal analgesia. These findings are consistent with a recent report by Khalili et al.³⁰ Our results are also supported by Hung et al who reported that the addition of MgSO₄ to either 2% lidocaine, 0.25% bupivacaine or 0.5% ropivacaine shortened the duration of sciatic-nerve block in rats.³¹ In our study, the onset of both sensory and motor blocks were longer in the magnesium group compared with the fentanyl and control groups, whereas in the study by Unlugence et al²⁹ although the onset of both sensory and motor blocks were longer in the magnesium group compared with the fentanyl group, there was no significant difference in the onset of sensory or motor blockade between groups M and C. The difference between the results of the two studies may be because of different doses of bupivacaine and the use of distilled water in the control group (used in our study) with different pH and baricity of solution compared with the study by Unlugence et al2⁹ in which normal saline was used in the control group. In a recent study by Khalili et al,³⁰ it was declared that in patients undergoing lower extremity orthopedic surgery under spinal anesthesia, the addition of 100 mg IT MgSO₄ to 15 mg of isobaric 0.5% bupivacaine caused a significant change in the duration of sensory block, while in our study, the difference in duration of sensory block time between groups M and C was shown to be insignificant. The difference between the results of the two studies may be caused by different doses of magnesium sulfate or the baricity of bupivacaine. 

In our study, contrary to the previous study,²⁹ none of the patients in the magnesium group demonstrated a hypotensive episode requiring treatment. The different findings of the two studies may be attributable to gradual onset of sympathetic blockade in the magnesium group in our study. Our results regarding fentanyl are consistent with previous studies.20, 22, 29, 32, 33 Fentanyl binds to opioid receptors32, 33 and also exerts a supraspinal action by intrathecal cephalic spread.³³ The selected dose of intrathecal magnesium in the current study was based on previous studies.22, 29, 34 Saeki et al³⁴ claimed that a dose of 1 mg/kg represents the maximum tolerable dose in rabbits from a neurotoxic standpoint and hence the selection of a dose of 50 mg intrathecal MgSO₄ in our study was based on the fact that several previous studies have shown that the use of such a dose could prolong the duration of intrathecal opioid analgesia without additional side effects.18, 20, 35, 36 Furthermore Khalili et al³⁰ also demonstrated that the application of a larger dose (100 mg) could not produce any further desirable effects compared to 50 mg MgSO₄ except prolonging the duration of sensory block with no effect on duration of motor block or duration of spinal analgesia. The authors of the present study speculate that there are a few possible reasons why magnesium failed to prolong the time to the first analgesic requirement. First of all, it has been claimed that the effects of MgSO₄ on the NMDA receptor complex are weaker than those of some other NMDA receptor antagonists.34, 37 The second possible cause is that MgSO₄ which most likely vasodilates the tissues around the injection site, will eventually accelerate the systemic uptake of local anesthetic, thereby prolonging the onset time of block.31, 38 Thirdly, according to the results of Okutomi's experimental study,3⁹ MgSO₄ might activate bupivacaine hydroxylation by the cytochrome P450 (CYP).³⁹

Therefore, the addition of intrathecal MgSO₄ to spinal bupivacaine may alter bupivacaine pharmacokinetics and cause a more rapid elimination of bupivacaine. Furthermore, MgSO₄ is reported to have been successfully used to attenuate the bupivacaine-induced toxicity in the central nervous system and heart.39, 40 In the present study although magnesium failed to prolong the time to first analgesic requirement as seen with fentanyl, it reduced the total opioid consumption in the first 12 hours postoperatively compared with the control group. This finding is in line with the recent report by Khalili et al.³⁰ The possible explanation to this finding could be augmentation by intrathecal MgSO₄ of the opioid effects, used as postoperative pain relief, and thereby preventing the subsequent NMDA activation. The NMDA receptors antagonist potentiates the opioid antinociception by blocking the spinally-mediated facilitators component evoked by repetitive C-fibre stimulation.¹⁹ Analgesic consumption has been shown to be correlated with primary hyperalgesia caused by the increased responsiveness of primary afferent nociceptors rather than by central sensitization.⁴¹

5. Conclusion

Based on the data found in our study, it could be concluded that addition of MgSO₄ (50 mg) intrathecal to 15 mg of spinal bupivacaine (0.5% hyperbaric solution) prolonged the onset time of sensory and motor blockade without prolonging the duration of spinal anesthesia. Although MgSO₄ failed to prolong the time to first analgesic requirement, as seen with fentanyl, it reduced the total opioid consumption in the first 12 hours postoperatively compared with the control group. Furthermore, MgSO₄ administration could provide a more stable hemodynamic profile, lower the ambulation time and cause less side effects compared with the fentanyl group. Further studies are needed to evaluate the analgesic efficacy of magnesium with other neuraxial drug combinations such as clonidine, neostigmine, and epinephrine to provide better analgesia and reduce the incidence and severity of side effects.