To understand the pharmacologic treatment of neuropathic low back pain, one must appreciate the pathophysiology of low back pain. Diagnosing the source of the structural spinal pathology alone is not enough. While it is important to understand the degenerative cascade of the lumbar spine, one must also incorporate the concept of chronic pain as a concurrent disease. Both peripheral and central mechanisms play a role in the pathogenesis of chronic neuropathic low back pain. The clinician can approach effective pharmacotherapy via an understanding of peripheral concepts such as neurogenic spread of chronic inflammatory pain, peripheral hyperalgesia and allodynia, highly activated sodium channels and ectopic neural triggering. Ultimately the central effects of these peripheral processes on the development and maintenance of wind-up pain plays a critical role in pharmacotherapeutic interventional strategies. The general rule in pharmacological pain treatment is that of rational polypharmacology, rather than trying to address the entire problem with one class of medication. The work of crafting a mosaic of treatment can be accomplished, when the clinician understands the various contributors to the loss of homeostasis in the bodyÕs own attempt to modulate pain.
Traditionally neuropathic low back pain has been correlated with radiculopathy only. The problem with this limited view is that it ignores aggravation of neural structures other then the nerve root via peripheral and central mechanisms. Controversy and confusion have long reigned regarding the upheaval caused by the degenerative cascade of lumbar motion segments. While numerous studies point out an extremely high volume of non-painful degenerative discs, the corollary that all lumbar motion segment degeneration is normal is incorrect. [Kawakami, M Anatomy, Biochemistry and Physiology of Low Back Pain, In Spine Care, Edited by White & Schofferman, St Louis; Mosby, 1995 p84-103.]
A motion segment of the low back is made up of two posterior Zygapophyseal (Facet) joints and an intervertebral disc, with their supportive muscles and connective tissue. When any of these tissues are injured an inflammatory response may ensue, causing increased pain. [Selby, D The Structural Degenerative Cascade: Lumbar, In White and Schofferman, editors, Spine Care, St. Louis; Mosby, 1995, p8-15.] (Figure 1)
Figure 1 (©Moskowitz, 2002)
The myth that the lumbar discs are not innervated with nociceptors (pain transmitting nerves embedded in peripheral tissues) has long been put to rest. In fact the outer third of the disc has three separate nociceptive nerve groups all feeding afferent branches to the nerve root.[Bogduk, N, Tynan, W, Wilson, AS: The nerve supply to the human vertebral disc, J Anat, 132:39, 1981] When a disc degenerates, a crack in the inner wall (annulus) may allow nuclear material to herniate outside of the disc margins causing compression of neural structures or to spread out within the disc causing inflammatory activation of nociceptors.[Crock, HV, Internal disc disruption: a challenge to disc prolapse fifty years on, Spine 11: 650, 1986.] Additionally herniated discs may compromise the neuroforamen with resultant nerve root compression.[Selby, D, The Structural Degenerative Cascade: Lumbar, p. 15, in White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby.] (Figure 2) The nerves can also become inflamed and swell in the intervertebral foramina.[Kawakami, M, Kenici, C, Weinstein, J: Anatomy, Biochemistry and Physiology of Low Back Pain, pp 92-94, in White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby.]
Internal Disc Disruption occurs when there is a tear in the inner wall of the annulus and inflammatory nuclear material seeps into the damaged disc wall. (Figure 2) Not only is the outer third of the disc wall richly innervated by nociceptive nerve endings from the sinovertebral nerves, the gray rami communicantes and the lumbar ventral rami, but chronic inflammation of these nerve endings and damage to the annulus results in peripheral sprouting with extension of these nerves into more central aspects of the disc wall. (Bogduk, et al, Discography, p. 221-224. In White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby. ]
Figure 2 The process of disc herniation can lead to compression of nerve roots causing neuropathic back and leg pain, when nuclear material breeches the intervertebral disc margins and compresses exiting spinal nerve roots. If the disc margins are not breeched and nuclear material spreads out within the intervertebral disc, loss of mechanical advantage of the disc and chronic inflammation of nociceptors can cause back and leg pain. (©Moskowitz, 2002)
The spine can also degenerate without any internal disc disruption. Pain occurs in this situation due to several different etiologies. Vertebral endplates are actually part of the vertebral bodies, not the intervertebral discs. They serve to diffuse oxygen and nutrients into the avascular vertebral discs, keeping them healthy and functioning. During injuries to the low back the endplates may either tear away from the intervertebral discs or actually fracture. [Bogduk, et al, Discography, p. 221-224. In White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby.] The separation of the vertebral disc from its nutrient supply may go on to rapid degeneration. (Figure 4) If the course is that of degeneration, cracks begin to develop in the disc wall, which loses its elasticity. The disc begins to dry out in a process called desiccation and the disc loses height. Zygapophyseal joints, which take on 20% of the weight bearing of the lumbar spine, begin to become overloaded, as synovitis gives way to destruction of the articular surface and subluxation of the joint with laxity of the joint within its capsule, a process known as spondylosis. Further instability of the entire motion segment occurs,
As the intervertebral discs lose height, the Zygapophyseal joints begin to spur, with calcium deposits in an attempt to stabilize, but this results in bone spurs growing into an already compromised neuroforamen. At this stage there is nerve root compression, which can be intermittent, painful, and position dependent. [Selby, D The Structural Degenerative Cascade: Lumbar, In White and Schofferman, editors, Spine Care, St. Louis; Mosby, 1995, p8-15.] (Figure 3)
Figure 3 (©Moskowitz, 2002) Disc degenerates and loses height compressing the motion segment composed of disc and superior and inferior facet joints. Neuroforamen becomes compromised, compressing nerve root. Compressed facets degenerate and hypertrophy, ultimately forming osteophytes (bone spurs) that may compress medial branch or nerve root. Internal Disc Disruption causes inflammatory irritation of annulus nociceptors.
Ultimately the neuroforamen becomes compromised enough to prevent egress of the nerve root and combined with increasing instability of the motion segment, pain and weakness ensue with a neuropathic pattern to the pain causing both low back and referred pain. The ongoing anatomical compromise of the exiting nerve root can cause permanent damage to the nerve and through local inflammatory chemicals, may cause further scar entrapment of the nerve root. (Figure 4) Nerves may even begin to be tethered in the central spinal canal if inflammation extends into the canal.
Figure 4 Degeneration occurs with loss of integrity of vertebral endplates with subsequent interruption of osmotically diffused nutrients to the intervertebral disc. Degeneration can result in compression of the nerve root in the neural foramen, producing a pattern of neuropathic low back and leg pain. (©Moskowitz, 2002)
Within the central spinal canal painful conditions can occur from nerve inflammation or compression.[Kawakami, M, Kenici, C, Weinstein, J: Anatomy, Biochemistry and Physiology of Low Back Pain, pp 92-94, in White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby.] A clumping of nerves in the central canal by scar tissue (arachnoiditis) can also cause nerve compression and pain.[Burton, CV: Lumbosacral arachnoiditis, Spine 3: 24, 1978.]
Injury to non-neural tissue in the low back results in activation of nociceptors. This starts as a normal process invoking synergistic mechanisms between the peripheral tissue injury, the spinal cord and the brain to regulate acute pain and bring it to rapid resolution. Chronic inflammation via unresolved peripheral chemical cascades and ongoing mechanical irritation of these inflamed areas sets off a series of events involving recruitment of nerves outside of the zone of injury and sensitization of neurons in the Dorsal Root Ganglia and Dorsal Horn of the Spinal cord.[Weinstein, JN: Anatomy and neurophysiologic mechanism of spinal pain, In Frymayer, JW, et al., editors: The Adult Spine: Principles and Practice, New York, 1991, Raven Press, p593.] Chronically inflamed nociceptors themselves release Calcitonin Gene Related Peptide (CGRP), Substance-P and ATP in the peripheral nervous system and recruit normally non-pain transmitting large diameter highly myelinated A-§ mechanoreptors and touch receptors to rapidly transmit pain to the spinal cord. (Figure 5) Over time this process spreads outside of the zone of inflammation, giving a look of non-anatomical pain. At this point the line between chronic inflammatory and neuropathic pain is blurred, with increased pain to stimuli that would normally produce less pain (hyperalgesia) and pain production from non-painful stimuli (allodynia) predominating. [Neuman S, Doubell TP, Woolf CJ, Inflammatory pain hypersensitivity mediated by phenotypic switch in myelinated primary sensory neurons, Nature 384; 360-364, 1996.]
Figure 5 When inflammation occurs blood vessels, fibroblasts, macrophages and mast cells release chemicals including Cholecystokinen, corticotropin releasing factor, leukotriens, prostaglandins, and arachidonic acid. When the inflammation is chronic non-painful A-§ mechanoreceptors ( ) are activated and begin to transmit pain through release of CGRP, Substance-P and ATP. When A-§ receptors are activated allodynia occurs. The continuation of this pain can lead to central nervous system wind-up pain. (©Moskowitz, 2002)
Central activation of sympathetic efferents may then feedback upon the peripheral injury causing spontaneous, unprovoked pain increases.[Roberts, WJ, A hypothesis on the physiological basis for causalgia, Pain, 1986; 24: 297-311.] This can worsen with peripheral activation by peripheral sympathomimetic inflammatory products and supersensitization of adrenergic receptors.[Wasner, G, Drummond, P, Birklein, F, Baron, R: The role of the sympathetic nervous system in autonomic disturbances and Òsympathetically maintained painÓ in CRPS, In Harden, et al, editors, Complex Regional Pain Syndrome, Seattle, 2001, IASP Press, p107-109.]
Injury to the nerve root causes ectopic firing of injured tissue via increased sodium channel activity. This ectopy requires no particular stimulus and fires at random from short bursts of activity (paroxysmal allodynic neuropathic pain) to sustained volleys of activity (constant allodynic neuropathic pain). Painful stimuli are exaggerated (hyperalgesia). In turn changes occur at the Dorsal Root Ganglia (DRG) that cause in-growth of normally non-pain transmitting A-§ nerve terminals into the superficial pain transmitting areas of the Dorsal Horn. (Figure 6) At the same time the DRG neurons produce mRNA delivered messages that cause atrophy in these same areas of normal C-fiber nociceptors This is known as nerve sprouting and causes non-painful sensory input to be interpreted and passed to the brain as a pain producing signal (Allodynia). [Black, J et al, Sodium channels as therapeutic targets in neuropathic pain, In Hansson, et al, Neuropathic Pain: Pathophysiology and Treatment, Seattle, 2001, IASP press, p22-29.]
Figure 6 Damage to nerve causes sprouting of A-§ Fibers into superficial lamina, while C-fiber terminals atrophy. (©Moskowitz, 2002)
Another problem with damage to nerves can occur if there is a loss of sensory input into the dorsal horn of the spinal cord, a process known as deafferentation. This often results in pain signals transmitted from the spinal cord to make up for the lack of sensory input into the Dorsal Horn. In a sense the spinal cord second order neurons make up their own response to the lack of normal background information.[Fields H.L. and Hill R.G: The near and far horizon, In Neuropathic Pain: Pathophysiology and Treatment, Seattle, 2001, IASP press, p 258-261.] This can result in severe neuropathic pain, sometimes delayed for years after the injury, which arises suddenly without clear provocative incident.
Any of these peripheral pain processes may precipitate the centerpiece of the development of the disease of chronic low back pain, activation of normally dormant N-Methyl D-Aspartate (NMDA) receptors by Glutamate.[Woolf, CJ, Salter, MW: Neuronal plasticity: increasing the gain in pain, Science 200; 288: 1765-1788.] These receptors are located throughout the Spinal Cord and Brain. Their usual function is that of destroying dying neurons and their terminals in order to replace them with newly developed neurons.[Olney JW and Ishimaur MJ; Exitotoxic cell death, In Koliatsos and Ratan, Cell Death and Diseases of the Nervous System, Humana Press, New Jersey, 1999, 199-202.] Normally the NMDA receptor is blocked by Magnesium, preventing activation by Glutamate. Due to this block synaptic calcium cannot move through closed channels in the NMDA receptor to the interior of the post synaptic cell. [Coderre C: Excitatory amino acid antagonists: potential analgesics for persistent pain, In Sawynok and Cowan, editors, Novel Aspects of Pain Management, Wiley Liss, New York, 157-178) (Figure 7)
Figure 7 First order neuron terminal synapsing with second order neuron in the Dorsal Horn of the spinal cord. Moderate amounts of Glutamate, an excitatory amino acid and the most common neurotransmitter in the nervous system, are released into the synapse. Glutamate can activate non-pain transmitting AMPA receptors, but not NMDA receptors blocked by Magnesium in the normal physiologic state. Substance-P remains in the presynaptic terminal. Excessive Ca is prevented from entering the post-synaptic cell. (©Moskowitz, 2002)
Constant activation of non-pain producing AMPA postsynaptic receptors by Glutamate, paired with Substance-P attached to Neurokinin-1 (NK-1) receptors, results in release of the Magnesium block in NMDA receptors. This allows the most ubiquitous and excitatory neurotransmitter Glutamate to attach to sites in the NMDA receptor and results in subsequent influx of Calcium ions into the postsynaptic cell. (Figure 8)
Figure 8 The pairing of Glutamate and substance-P activating postsynaptic AMPA and NK-1 receptors causes the Magnesium block to be removed from NMDA receptors and allows Glutamate to bind to sites in the NMDA receptor. This sensitizes the receptor, opening Calcium channels and resulting in the influx of Calcium ions into the postsynaptic cell. (©Moskowitz, 2002)
Calcium activates internal processes causing synthesis, membrane placement and activation of NMDA receptors and influx of more Calcium. Calcium also activates the free radical, Nitric Oxide, which diffuses across the synapse to presynaptic cells. More substance P is released setting up a vicious cycle. The result is that pain winds up and amplifies on its way to the brain. When the pain signal arrives, the brain interprets it as greater than the peripheral stimulus (Wind-up).[Woolf, CJ, Salter, MW: Neuronal plasticity: increasing the gain in pain, Science 200; 288: 1765-1788.] NMDA receptor activation also is a major contributor to opioid tolerance.[Mao, J, Price, DD, Mayer, DJ, Mechanisms of hyperalgesia and morphine tolerance: a current view of their possible interactions, Pain 1995a, 62: 353-364.] (Figure 9)
Figure 9 Protein Kinase-C (PKC) is increased in its inactive form by opioid binding to Mu-Opioid receptors. Glutamate activated NMDA receptors allow calcium into the cell and this activates the increased PKC. Activated PKC attaches to cell membranes and increases the sensitivity of NMDA receptors to allow in more calcium, which in turn activates mRNA to create more NMDA receptors. This allows more calcium in the cell, which then works with Nitric Oxide Synthetase on L-Argenine to increase postsynaptic nitric oxide. It diffuses across the synapse to the presynaptic cell and releases more substance-P, defeating the action of presynaptic opioids on Mu receptors. Substance-P attaches to postsynaptic sites defeating the effect of postsynaptic opioids on Mu receptors, resulting in opioid tolerance and wind-up pain. (©Moskowitz, 2002)
A physiologic source of control over wind-up is the GABAergic system. GABA is the main inhibitory neurotransmitter in the Central Nervous System. While it is ubiquitous in its distribution, there is a particularly high concentration of GABA in the most superficial lamina 1, 2 and 3 of the Dorsal Horn of the spinal cord. This also happens to be an area of high concentration of Glutamatergic first order neurons.[Todd AJ and Maxwell DJ; GABA in the mammalian spinal cord, In Martin and Olsen, GABA in the Nervous System, Lippincott, Williams and Wilkins, 2000 p439-458.] In normal physiologic functioning Dorsal Horn GABA serves as a brake upon the excitatory activity of Glutamate. A balance is struck between these two prominent neurotransmitters. In this balanced state, GABA receptors on Glutamatergic afferent terminals prevent excessive release of Glutamate and Substance P.[Roberts, E., ÒAdventures with GABA,Ó In Martin and Oslen, GABA in the Nervous System, Lippincott, Williams and Wilkins, 2000, p1-24.] (Figure 10)
Figure 10 The action of GABA on Glutamatergic afferent terminals prevents the release of Substance-P and maintains homeostasis between excitatory and inhibitory CNS activity. (Moskowitz, 2002)
In chronic pain states this balance is lost and the excitatory aspect of the nervous system overwhelms GABAÕs inhibitory influences.[Yaksh TL, Behavioral and anatomic correlates of the tactile-evoked allodynia produced by spinal glycine inhibition: effects of modulatory receptor systems and excitatory amino acid antagonists, Pain, 1989;37, p111-123.] (Figure 11) Excessive excitatory neurotransmitter activity in the Central Nervous System is a problem in chronic pain conditions and in chronic affective disorders.[Blackburn-Munro G; Chronic pain, chronic stress and depression: coincidence or consequence, Journal of Neuroendocrinology 2001 Dec;13(12):1009-23] Both are mediated through the action of excitatory amino acids on activated NMDA receptor sites. [Mathew, Sanjay J., et al ÒGlutamate-Hypothalamic-Pituitary Adrenal Axis Interactions: Implications for Mood and Anxiety Disorders,Ó CNS Spectrums, 6(7): 555-564, 2001.
Figure 11 Without adequate GABAergic restraint upon first order Glutamatergic neurons; excessive release of Glutamate and Substance-P activates second order AMPA, NK-1 and NMDA receptors (not shown). Activation of GABA receptors results in decreased release of Glutamate and Substance-P (©Moskowitz, 2002)
Combined Peripheral and Central Pain Mechanisms
Complex Regional Pain Syndrome (CRPS) 1 and 2 represent yet another aspect of the overall picture of neuropathic low back pain. Although we tend to think of this syndrome as occurring more frequently with peripheral limb and nerve injuries, it occurs with significant frequency secondary to damage to nerve roots and following injury or surgery on the low back. The pathological changes that lead to the development of this condition remain unclear, but most current theories indicate that the precipitation of this condition is multifactorial in nature and includes central process and peripheral processes.[Kramis RC, Gillette RG Roberts WJ; Neurophysiology of chronic idiopathic back pain, p11-113, White and Schofferman, editors, Spine Care, St. Louis, 1995, Mosby. Although it is generally accepted that both CRPS 1 and 2 are sympathetically maintained pain conditions, the mechanism for this is not yet understood and there are many who feel that the sympathetic system involvement represents a reaction rather than an etiology. Other theories advanced in these conditions include recruitment of uninjured afferents in injured and anatomically contingent nerves, exaggerated peripheral inflammatory responses, central sensory disturbances, antidromal neurogenic inflammation, psychophysiologic activation, and genetic predisposition[Janig W; CRPS1 and CRPS2 a strategic view, in Hardin RN, Baron R, Janig W, editors, Complex Regional Pain Syndrome, p 3-15, IASP Press, Seattle, 2001].
The decision process regarding which medications to use for treating neuropathic low back pain must take into account several factors. Random Controlled Trials (RCTs) of medication for this type of treatment are lacking. One can infer from RCTs done on other types of neuropathy the type of medications that might be helpful, but the great majority of studies have been done on Post Herpetic Neuralgia and on Diabetic Neuropathy, neither of which plays any significant role in low back pain. One well designed longitudinal study on opioids in treating chronic low back pain [Schofferman, J, Opioid analgesic therapy for severe intractable low back pain, Clinical Journal of Pain, June; 15(2): p136-140.] has been published, but most other studies involve short term treatment of patients with chronic low back pain.[Schnitzer TJ, Gray WL, Paster RZ, et al. Efficacy of tramadol in treatment of chronic low back pain, Journal of Rheumatology (Canada), Mar 2000, 27(3) p772-8] and [Simpson RK, Edmondson EA, Constant CF, et al. Transdermal fentanyl as treatment for chronic low back pain, Journal of Pain Symptom Management (United States), Oct 1997, 14(4) p218-24] While experience does not equal evidence, much effective pharmacological treatment for neuropathic low back pain fills the void left by the lack of evidence based studies. Additionally physicians can use known analgesics and apply these in the treatment of neuropathic low back pain, despite the dearth of specific RCTs regarding their effectiveness. Finally we can make pharmacologic choices based upon known action of existing pharmacologic agents released to treat other conditions, combined with our knowledge of the basic science of the disease of chronic pain. This last choice accounts for most of the current innovative pharmacologic treatment for chronic low back pain. Furthermore, designing medication treatment strategies to diminish neuropathic low back pain must take into account the structural disease, alteration of normal neuroanatomy and neurophysiology, and likely sites of medication action. Joining together a mosaic of medications and routes of administration is far more likely to be effective than the use of any one medication, medication class, or delivery system. Much of that which follows will be based upon my own clinical experience in over 15 years of pharmacological treatment of patients with chronic low back pain.
Starting with structural disease the first determination should be whether the neuropathic pain is most likely caused by intermittent entrapment with resultant nerve root inflammation and swelling leading to periodic pain symptomatology or by permanent damage to the nerve. If clinical correlation points to the former, a combination of topical lidocaine patch, opioid and antiinflammatory may be the best choice. If the structural lesion appears to be caused by permanent damage to a nerve, the topical lidocaine patch, tricyclic antidepressants and anti-epilepsy drugs (AEDs) may prove to be most effective. Inflammation of the surrounding tissue of a disc or facet joint injury usually responds well to antiinflammatories, while inflammatory injury to the internal disc wall tends to have a poor response to these medications due compromised blood supply. In this situation opioids tend to be a better choice, because their main location of action is in the Central Nervous System. If fractures have occurred or if one or more segments of the lumbar spine are in the process of fusing from surgical intervention, antiinflammatories should not be used due to their role in retardation of bone fusion. Arachnoiditis is most likely to respond to opioids and AEDs, as is a pattern of constant referred pain to the lower extremities from intervertebral neuroforaminal entrapment from scar tissue, large osteophytes or severe stenosis.
Turning to the disease process of chronic pain, the processes that lead to allodynia and hyperalgesia present interesting targets for medication intervention. These include activation of normally non-nociceptive A-§ Fibers by chronically stimulated C-fibers, recruitment of nociceptors and non-nociceptors beyond the area of injury or inflammation, activation of central and peripheral sympathetic processes, deafferentation, and NMDA receptor activation with wind-up and loss of brain spinal cord synergy. Chronic C-fiber stimulation by inflammatory cascade chemicals may improve with antiinflammatory treatment. When allodynia involves pain to touch use of topical agents and compounded transdermal agents can be quite helpful, as can AEDs and tricyclic antidepressants. Deeper allodynia may also respond to these agents. Sympathetically maintained pain also responds to AEDs, topical and transdermal approaches. Deafferentation pain is notoriously difficult to treat, but appears to respond best to anti epilepsy drugs and opioids. Activation of NMDA receptors, as the central disease process in chronic pain presents with many interesting approaches and challenges. The obvious target of action is that of blockade of the receptor, but issues of efficacy and side-effects have made this treatment a tricky one. New approaches are being developed. Most currently available agents have weak to moderate efficacy, with one agent, Ketamine, having potent efficacy, but significant side effect potential. Spinal cord calcium channel blockade is yet another strategy, as is increased activation of the GABAergic system. Current Calcium Channel blockers may be helpful, but more potent N-type Calcium Channel blockers are in development. [Hunter JC, Voltage-gated ion channel modulators, in Sawynok J and Cowan A, editors, Novel Aspects of Pain Management, Wiley Liss, New York, 1999, p329-333.] Since GABA based inhibition of Glutamate is overwhelmed in chronic NMDA receptor activation, targeting the GABAergic system is another approach to controlling wind-up pain. Several of the AEDs have GABAergic action, including Tiagabine, Topirimate and possibly Gabapentin. Baclofen is a non-AED with GABAergic spinal cord activity.
Sites of medication action, hinted upon in the above several paragraphs, provide yet another strategy for dealing with neuropathic low back pain. Antiinflammatories and opioids are most effective at peripheral injury sites, where antiinflammatories interfere with the inflammatory cascade[Gaslinger, G and Yaksh, T, Spinal actions of cox inhibitors, p 773, In Devor et al, editors, Proceedings of the 9th World Congress on Pain, Seattle, 2000, IASP press.] and opioids address diminution of release of Substance-P.[Ossiprov, M, et al, Recent advances in the pharmacology of opioids, In Sawynok and Cowens, editors, Novel Aspects of Pain Management: Opioids and Beyond, New York, 1999, p55-60.] As stated above, when recruitment of nerves outside of the original injury site has developed, Neuromodulators, including topicals, transdermals and AEDs can be extremely helpful. Neuromodulators can also be helpful with pain caused by nerve damage. The most powerful effects of opioids are at the Dorsal Horn of the Spinal Cord, but this is also the major location of NMDA receptor induced opioid tolerance. .[Mao, J, Price, DD, Mayer, DJ, Mechanisms of hyperalgesia and morphine tolerance: a current view of their possible interactions, Pain 1995a, 62: 353-364.] New approaches to blocking NMDA receptors as an adjunctive treatment with opioids are being developed.[Dickenson, A and Suzuki, R, Function and dysfunction of opioid receptors in the spinal cord, In Kalso, et al, Opioid Sensitivity of Chronic Non-Cancer Pain, Seattle, IASP Press, 1999, p17-44.] Pathways that normally modify pain from the brain back down to the dorsal horn of the spinal cord via modulation of nor-epinephrine can be helped with use of low dose tricyclic antidepressants.[McQuay, HJ, et al: A systematic review of antidepressants in neuropathic pain, Pain 1996; 68: 217-227.] Alpha-1 antagonists and Alpha-2 agonists.[Yaksh, T, Alpha-2-agonists as analgesics, In Sawynok and Cowens, editors, Novel Aspects of Pain Management: Opioids and Beyond, New York, 1999, p179-202.] GABAergic medications can be helpful with anxiety, sleep disturbance and pain and include use of Benzodiazepines, Zolpidem, and GABA enhancing anticonvulsants.[Bowery N and Marzia M; Gamma-aminobutyric acid and pain, In Sawynok and Cowens, editors, Novel Aspects of Pain Management: Opioids and Beyond, New York, 1999, p249-264.] Depression and anxiety often accompany cases of chronic low back pain and can effectively be treated with Serotonergic antidepressants with low drug to drug interactions, such as Citalopram, Escitalopram, Sertraline, Mirtazepine and Venlafaxine or with the mostly Dopaminergic antidepressant, Buproprion. (Figure 12)
Figure 12 depicts sites of action of medications. Several medications have multiple sites of action. Combining different medications from different classes with various routes of administration can result in effective pharmacotherapy. Drug to drug interactions and side effects are important considerations. Much of this treatment is off-label.
Pharmacological Strategies for Different Routes of Administration
Varying and mixing medications with different routes of administration presents yet another strategy. Oral medication delivery provides systemic absorption and potential penetration of the blood/brain barrier. Most pharmacologic management of pain relies upon these facts. Problems can arise however with central and peripheral nervous system side effects and in this situation using topical treatment as a primary or adjunctive form of pharmacotherapy is quite attractive. The major topical agents are capsaicin, a Substance-P inhibitor and the topical Lidocaine patch, a sodium channel blocker. Transdermal and transbuccal delivery of opioids presents with another set of strategies. The former allows for long acting efficacy of short acting Fentanyl and the latter allows for rapid absorption of Fentanyl for breakthrough pain. Compounded transdermal agents applied to local sites of pain may provide a hybrid action of topical and systemic transdermal effects. Parenteral routes of medication administration tend to be limited to hospitalized patients, with one exception. Implanted intrathecal pumps can be a useful and alternative means of medication administration with delivery of a constant infusion of medications, including opioids, Baclofen, and Clonidine. Experimental protocols for the N-type Calcium Channel Blocker, Ziconotide, are currently in clinical trials by this route of administration. The advantages of intrathecal delivery of opioids appear to be that of less drowsiness and less constipation. Disadvantages are that of aesthetic alteration, pump/surgical complications and mechanical failure.
The reader is referred to the table of commonly used medications at the end of this article. Medication dosing has been left out due to wide variability. (Table 1) The second table refers to less commonly used medications that can be quite effective for treating neuropathic low back pain. (Table 2) The general rule with all medications is to start low and go slow.
Clearly the task of treating chronic neuropathic low back pain is extremely complex and challenging. Even the determination of what constitutes neuropathic low back pain can be clinically daunting. Fortunately there are equally complex strategies for dealing with this vexing problem. Hindering these strategies are the lack of good long-term random controlled trials. Clinical experience with medications shown to be efficacious for these other disorders, already established general analgesics and medications addressing chronic pain mechanisms will dominate the picture of treatment. Future treatments aimed at control of the NMDA receptor, GABAergic enhancement, Neurokinin-1 receptor blockade, peripheral and central antiinflammatory treatment, CNS Calcium Channel blockade and peripheral sodium channel blockade will be developed. Clinical efficacy will need to be proven for these treatments and the landscape of pharmacological options will change over time. For now, and into the foreseeable future, the effect of properly balanced polypharmacology, using drugs with low drug interactions, balancing locally directed treatment with systemic treatment and aiming at multiple neurochemical and neuroanatomical pathways will provide optimal medication response.
Michael H. Moskowitz, MD, MPH