Intrathecal neostigmine has been used as an adjunct to intrathecal local anesthetic or opioid to prolong regional analgesia and improve hemodynamic stability, with variable results. Escalating doses of intrathecal neostigmine (10 to 100 µg) followed by 2% epidural lidocaine resulted in improved analgesia in a dose-independent manner after cesarean delivery.28 The reduction in morphine requirements lasted up to 10 hours without adverse fetal effects, but the incidence of nausea varied from 50% to 100% in patients. In another study, intrathecal neostigmine (10 µg) alone was ineffective for labor pain relief, but when combined with intrathecal sufentanil, reduced the ED50 of sufentanil by approximately 25%.29
Epidural administration of neostigmine (100 to 200 µg) appears to avoid these clinically troublesome adverse effects while still improving local anesthetic-induced analgesia.30,31 Combinations of epidural neostigmine with local anesthetics, opioids, or clonidine for labor analgesia displayed analgesic effectiveness, potentiating the analgesic effect of opioids and clonidine.32–34 Epidural neostigmine does not affect motor blockade. Higher doses of intrathecal neostigmine can cause mild sedation.27,35
A meta-analysis evaluated the effectiveness and side effects of intrathecal neostigmine in the perioperative and peripartum settings. The authors concluded that adding intrathecal neostigmine to other spinal medications improves perioperative and peripartum analgesia marginally when compared with placebo. It is associated with significant side effects and the disadvantages outweigh the minor improvement in analgesia achieved.36 Nausea and vomiting were seen less frequently in epidural neostigmine studies. This is thought to be due to the lower amount of neostigmine that reaches the CSF and the absence of cephalic distribution.37 Neostigmine stimulates muscarinic receptors in the bronchial smooth muscles and leads to bronchospasm. In intrathecal neostigmine studies, except at very high doses (e.g., 750 µg), no change has been detected in oxyhemoglobin saturation and in end-tidal carbon dioxide levels.38
Intrathecal neostigmine at a dose of 1 µg/kg has been used in pediatric lower abdominal and urologic surgeries where it was found to increase analgesia.39,40 Adverse gastrointestinal effects have made neostigmine an unpopular choice for neuraxial adjuvant therapy. Unlike intrathecal neostigmine, epidural neostigmine is not associated with an increased risk of nausea and vomiting; however, doses greater than 100 µg have been associated with sedation. It does not cause respiratory depression or pruritus either alone or in combination with neuraxial opioids.
Ketamine
Anesthetic and subanesthetic doses of ketamine have analgesic properties as a result of noncompetitive antagonism of N-methyl-D-aspartate (NMDA) receptors. With prolonged, repetitive nociceptive stimulation, NMDA receptors are activated, releasing excitatory neurotransmitters glutamate, aspartate, and neurokinin.41 Its primary analgesic effect is mediated by antagonizing NMDA receptors located on secondary afferent neurons in the dorsal horn of the spinal cord thus preventing enhancement of excitatory neurotransmission. These neurotransmitters are associated with many activities including central sensitization, wind-up, and the plasticity of various systems such as memory, vision, motor function, and spinal sensory transmission.
Neuraxial ketamine must be administered in a preservative-free solution to avoid neurotoxic effects.42–45 Naguib et al.46 studied epidural doses of 10 mg and 30 mg of ketamine and found that a 30-mg dose produced excellent postoperative pain relief. A low dose of ketamine at 4, 6, and 8 mg epidurally was found to be ineffective for postoperative analgesia.47,48 Caudally administered ketamine 0.5 mg/kg along with 0.175% levobupivacaine 1 mL/kg has been used successfully without adverse effects in children for lower abdominal and urologic surgeries.49 Epidural infusion of 0.25 mg/kg per hour of S(+)-ketamine during thoracic surgery provides better postoperative analgesia than epidural 0.25% ropivacaine (Fig. 8-2).50 Both epidural infusions were started before skin incision and were run at 6 mL per hour for the duration of surgical procedure in the previously cited studies.
Combination of epidural ketamine with local anesthetic and/or opioid infusions results in improved analgesia without significantly increasing adverse effects.1,51,52 A bupivacaine-ketamine mixture provided better analgesia than bupivacaine alone (Fig. 8-3).51 Side effects such as motor weakness or urinary retention were not observed in the ketamine group.51 Adding low-dose ketamine to a multimodal epidural analgesia regimen provides better postoperative analgesia and reduces morphine consumption in thoracic, upper abdominal surgery, and lower abdominal surgeries.50 Ketamine acts synergistically with opioid, dopaminergic, serotoninergic, and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor agonists to produce dissociation between the thalamocortical and limbic systems. AMPA is a non-NMDA-type ionotropic glutamate receptor that mediates fast synaptic transmission in the CNS. Ketamine has also shown efficacy in the management of neuropathic pain, and it is believed to work through one or more of these mechanisms. In high doses, ketamine may have additional minor analgesic effects by modulating descending inhibitory pathways through inhibition of reuptake of neurotransmitters.
Reported side effects of epidural ketamine include sedation, headache, and transient burning back pain during injection with doses greater than 0.5 mg/kg. There has been no reported respiratory depression, hallucinations, cardiovascular instability, bladder dysfunction, or neurologic deficit with epidural doses up to 1 mg/kg. The incidence of nausea, vomiting, and pruritus when combined with neuraxial opioids is similar to that reported with neuraxial opioid alone.
There are limited human studies on the use of intrathecal ketamine due to the potential risk of neurotoxicity from its preservative benzalkonium chloride. Intrathecal ketamine has been shown to decrease morphine requirements in patients with terminal cancer and is useful in opioid-tolerant patients. The intrathecal administration of ketamine, however, did not prolong or improve the quality of anesthesia from bupivacaine, but increased adverse effects.41 Ketamine has been administered intrathecally to 16 patients with war injuries of the lower limbs in varying doses from 5 to 50 mg in a volume of 3 mL of 5% dextrose.53 In these doses, intrathecal ketamine resulted in a distinct sensory level in all patients and satisfactory surgical analgesia. Central effects (drowsiness, dizziness, and nystagmus) occurred in nine patients, but they remained conscious throughout; one patient experienced no central effects, and one patient developed dissociative anesthesia. Ketamine alone did not produce motor block, but addition of epinephrine resulted in complete motor block and may have intensified sensory blockade.53
The advantage of intrathecal ketamine is the lack of cardiovascular effects and respiratory depression. The main drawbacks of intrathecal ketamine are the potential for psychomimetic reactions, inadequate motor blockade, and short duration of action. Clinical manifestations of myelopathy suggestive of spinal cord injury were observed in a terminally ill cancer patient after continuous infusion intrathecal preservative-free ketamine at a rate of 5 mg per day for a duration of 3 weeks.54
Midazolam
Intrathecal midazolam produces analgesia by acting on γ-aminobutyric acid (GABA)-A receptors and reducing spinal cord excitability. GABA-A receptors are ligand-gated receptors located throughout the CNS and GABA is a major inhibitory neurotransmitter of the CNS, although glycine is prominent in the spinal cord. GABA binding results in a change of receptor configuration, causing an ion channel to open which allows chloride ions to flow down their electrochemical gradient into the cell. This results in hyperpolarization of the neuron and reduced action potential propagation.
A high density of benzodiazepine receptors (GABA-A) have been found in lamina II of the dorsal horn of the spinal cord, suggesting a possible role in pain modulation. Benzodiazepines have also been shown to act at opioid receptors.55 The δ-selective opioid antagonist, naltrindole, suppresses the antinociceptive effect of intrathecal midazolam, suggesting that intrathecal midazolam is involved in the release of endogenous opioid acting at spinal δ receptors.56 The substantia gelatinosa of the dorsal horn of the spinal cord contains a high density of GABA-A receptors. Benzodiazepines are likely to mediate their analgesic effect by increasing inhibition of nociceptive neurons in this area.
Midazolam administered epidurally57–61 or intrathecally55,62–67 has been shown to have an analgesic effect (Fig. 8-4). Adding midazolam (10 to 20 mg for 12 hours) to continuous epidural infusion of bupivacaine (100 mg) for postoperative pain provided better analgesia than bupivacaine alone without sedative effects.26,68,69
Midazolam added to fentanyl-ropivacaine epidural analgesia was associated with a significant reduction in the incidence of postoperative nausea and vomiting compared with fentanyl-ropivacaine alone, and a significant decrease in the amount of patient-controlled epidural analgesia (PCEA) administered without a significant increase in adverse events in these patients who underwent subtotal gastrectomy and postcesarean delivery. The exact mechanism by which midazolam exerts its antiemetic action is not fully understood. Postulated mechanisms include glycine mimetic inhibitory effects, enhancement of the inhibitory effects of GABA, inhibition of dopamine release, and augmentation of adenosine-mediated inhibition of dopamine in the chemoreceptor trigger zone.68–70 No sedation is observed at doses of 1 and 2 mg of epidural midazolam; however, sedation has been reported at higher doses.
Animal study has shown synergism between intrathecal midazolam and morphine.71 Subsequent clinical studies have evaluated intrathecal and epidural midazolam for treatment of postoperative and chronic pain.72 Intrathecal midazolam has also been investigated in combination with an opioid for labor analgesia.64 These studies support the analgesic efficacy of intrathecal midazolam at doses <2 mg and in concentrations <1 mg/mL. Intrathecal midazolam is more effective for treatment of somatic pain than visceral pain.72 The addition of intrathecal midazolam also decreases postoperative analgesic requirements. The incidence of postoperative nausea and vomiting is much less with intrathecal midazolam than that seen with intrathecal fentanyl.73 Midazolam can be successfully combined with other drugs such as opioids and clonidine for additive effects66 and has been used as a continuous infusion (12 mg per day) in patients with refractory pain.62
A serious risk of intrathecal drug administration is neurotoxicity and such neurotoxic effects have been demonstrated in animal studies. Most animal studies examining intrathecal administration of midazolam have demonstrated no neurotoxic effects, although two of the earliest studies reported signs of neurotoxicity. Current evidence suggests that the addition of midazolam in doses of 1 to 2 mg intrathecally is beneficial in the treatment of perioperative and chronic pain. Current evidence supports the use of midazolam in doses not exceeding 1 to 2 mg at concentrations not exceeding 1 mg/mL. Considerable experience in humans with the use of perioperative midazolam suggests no evident deleterious neurologic effects under these conditions.63,74–76 The story of the experimental work on intrathecal midazolam in animals and humans is a cautionary tale in drug development. Investigators proceeded with clinical trials in humans at a time when the only available animal data suggested that intrathecal administration of midazolam might well be neurotoxic. Only after the human trials were performed did additional animal data emerge to support the lack of neurotoxicity,74 raising significant ethical concerns about the progression of investigational work from animals to humans.
Tramadol
Tramadol is an analgesic combining mainly µ-opioid and monoaminergic activity through the inhibition of the neuronal uptake of serotonin and norepinephrine.77 Animal studies have confirmed the analgesic effect of intrathecally administered tramadol. However, there is limited available data in humans. Epidural administration of tramadol has been the subject of some study78–81 and did not demonstrate effective postoperative analgesia with epidural administration.82,83
The effect of intrathecal administration of tramadol to patients showed contradicting results.84–86 Tramadol 1 to 2 mg/kg has also been administered caudally in children for postoperative analgesia.87
Droperidol
Epidural droperidol is effective for reducing pruritus and postoperative nausea and vomiting.88 Long-term administration of intrathecal droperidol proved to be an excellent antiemetic in patients with nonmalignant pain.89 It has been suggested that droperidol exerts direct actions on the brainstem chemoreceptor trigger zone. Although no side effects were observed, it is important to recognize the lack of laboratory data documenting the safety of neuraxial droperidol (including the potential for neurotoxicity).90
Adenosine
In the spinal cord, adenosine receptors are located in the superficial layers of the dorsal horn. Adenosine shows antinociceptive activity at adenosine A1 receptors located in laminae I and II of the dorsal horn of the spinal cord.91 Another proposed mechanism is enhancement of spinal norepinephrine release.92
Initial studies confirmed relative safety of intrathecal administration of adenosine in human volunteers with no reported clinical toxicity.93,94 Intrathecal adenosine does not inhibit acute pain95 but is effective in treating allodynia and hyperalgesia. Experimental hyperalgesia and allodynia is reduced by intrathecal adenosine in a non–dose-dependent fashion96; however, in clinical settings, it did not change the anesthetic requirement or postoperative analgesia.97 Similarly, in combination with an opioid, intrathecal adenosine did not prolong analgesia during labor.98
Adenosine appears be effective in the treatment of neuropathic pain. Intrathecal adenosine is not associated with hypotension, motor blockade, or sedation. Following many clinical trials involving animal subjects, intrathecal adenosine 500 to 2,000 µg in human volunteers was shown to decrease allodynia in phase I clinical trials. The only side effect observed was transient lumbar pain after a dose of 2,000 µg.99–101 The role of neuraxial adenosine in the armamentarium for treatment of acute or chronic pain awaits further delineation.
Conopeptides
Ziconotide
Ziconotide is a synthetic 25-amino acid, polybasic peptide with three disulfide bridges and is a derivative of an omega conotoxin found in the venom of the marine snail conus magnus. Ziconotide acts as a selective antagonist of neuronal (N-type) voltage-sensitive calcium channels within presynaptic neuronal terminals in the dorsal horn, thereby inhibiting nerve transmission. Ziconotide directly inhibits norepinephrine release and functions as a sympatholytic, resulting in decrease in mean and diastolic pressure, most profoundly when administered intravenously and normally negligible when dosed intrathecally.
Highly polar and water soluble, ziconotide is hypobaric at clinically useful concentrations and has a relatively large molecular weight. This agent has narrow therapeutic window, with neuropsychiatric side effects appearing in nearly all patients at higher doses or when dose escalation is too rapid. Initial infusion rates should be limited to 0.1 µg per hour with stepwise increase of this rate over time; CNS adverse effects are to be expected.102–104 Ziconotide is the only FDA-approved, nonopioid approved for intrathecal administration for the treatment of neuropathic pain.
Ziconotide produces marked spinal antinociception in animal models of acute and persistent pain105 and additional reports described its intrathecal administration to relieve severe neuropathic pain.106–109 Following extensive demonstration of safety in animal models, clinical trials in humans suffering with poorly controlled pain associated with advanced illness demonstrated that side effects occurred in the majority of patients (92.9%).110 Significant adverse events reported in the ziconotide group were dizziness, confusion, ataxia, abnormal gait, and memory impairment. Suicidality was increased with ziconotide as compared to placebo. Most of the side effects are self-limiting with cessation of therapy. The marked expense and nearly universal appearance of side effects in those receiving ziconotide have limited the use of this agent in clinical practice. A small number of patients with chronic pain gain significant, ongoing pain reduction with few or tolerable side effects when receiving intrathecal ziconotide infusions via chronic, implanted intrathecal drug delivery systems.
Other Investigational Conopeptides
Xen 2174
Xen 2174 is a conopeptide derived from a marine snail. The drug is found to inhibit norepinephrine transport and activate noradrenergic inhibitory pathways causing antihyperalgesic, antiallodynic, and antinociceptive effects.111
CGX-1160
CGX-1160 is a conopeptide that produces analgesia by activation of neurotensin receptor type 1 (NTR1). The drug has been found to be safe in a small number of patients with neuropathic pain related to spinal cord injury.112
The therapeutic index of these newer conopeptides may well be superior to that of ziconotide, but much additional investigational work is needed before they can reach clinical use. The promise of a novel nonopioid analgesic with significant efficacy in the treatment of neuropathic pain remains elusive but is the most alluring promise of this novel class of analgesics.
Octreotide