Current and future pharmacological agents for the treatment of back pain

KEYWORDS:

• Acetaminophen
• antidepressants
• benzodiazepines
• gabapentinoids
• low back pain
• medications
• muscle relaxants
• NSAID
• opioids
• pharmacotherapy

1. Epidemiology and scope of low back pain

Low back pain (LBP) is anatomically defined as pain between the lower posterior margin of the rib cage and the horizontal gluteal fold, with chronic pain defined as pain on most days for at least 3 months [1]. Chronic LBP is the leading cause of disability in the world with 60.1 million disability-adjusted life-years in 2015, an increase of over 50% since 1990. The worldwide mean point prevalence of LBP was estimated to be 18% in 2008 and the global point prevalence of activity-limiting LBP was 7.3% in 2015, with over half-a-billion people affected at any one time [2]. Aging, sedentary lifestyles, increasing body mass index, smoking, a high co-prevalence of psychosocial dysfunction, and low socioeconomic status are some of the important causes of this crisis [3]. LBP is a major contributor to health-care costs, and in the US, it was estimated to cost over 100 billion USD annually over a decade ago [4].

2. Classification of low back pain

Duration of symptoms is often used to classify LBP, with acute LBP defined as pain lasting less than 4 weeks, subacute LBP defined as lasting between 4 and 12 weeks, and chronic LBP persisting beyond 12 weeks [5]. LBP tends to resolve with time, but at least one-third of patients report persistent LBP of at least moderate intensity 1 year after an acute episode, and one in five report substantial limitations in activity [6]. Another categorization scheme LBP is mechanism- based, which that classifies etiology as neuropathic and nociceptive. In a recent study characterizing the clinical manifestations of LBP in over 35,000 patients, the proportions with likely neuropathic pain were 13.1%, 16.5%, and 23.3% for individuals with acute, subacute, and chronic LBP, respectively, [7]. In 2017, the category of nociplastic pain was introduced to classify a separate set of pain conditions characterized by altered processing of nociceptive signaling due to central sensitization [8]. Some forms of chronic LBP, particularly phenotypes characterized by a poor correlation between symptoms and pathology, may be nociplastic in nature. Studies have shown that patients with multiple nonimmune-mediated hypersensitivity reactions [9] suggestive of central sensitization respond poorly to interventional treatments for LBP. Perhaps the most common form of LBP is ‘non-specific’ [10], which has been defined as ‘ … pain that may or may not radiate down toward the knees, for which specific etiologies such as infections, neoplasms, metastases, osteoporosis, fractures, rheumatological disorders, neurologic disorders, and other relevant pathologic entities have been ruled out clinically.’ Nonspecific LBP may share components of both nociplastic and nociceptive pain. Pain classification is important because it affects treatment decisions at all levels of care (i.e. ordering diagnostic tests, prescribing medications, whether and what type of procedures to perform). Mechanism-based pain treatment is widely considered to be more effective than disease or condition-based, but identifying pain mechanisms in clinical practice can be challenging [11].

Treatments for LBP include pharmacotherapy, psychological approaches, physical medicine, procedural interventions, and integrative medicine options. Although a multimodal, personalized approach incorporating these treatments has been shown to provide superior outcomes compared to unidimensional treatment [10], this editorial focuses only on pharmacological options because it is often an early and a major component of the care package. Furthermore, the evidence for treatments discussed below is for treating chronic LBP not associated with sciatica unless otherwise specified.

3. Overview of pharmacological options for low back pain

Patients with LBP use medications from various classes, often using more than one class concomitantly. A survey in the United States of America of 700 adults under 70-years-old with chronic LBP revealed that 37% took at least 1 prescription pain medication in the past 30 days. Nineteen percent of these had used opioids, 10% took non-steroidal anti–inflammatory drugs (NSAIDs), 8.5% took muscle relaxants, and 7% used gabapentinoids. Antidepressants and hypnotics were used by 18% and 5%, respectively [12]. This survey showed that analgesic polypharmacy is common, though guidelines on the use of medications for LBP highlight the small benefit from these therapies. Combining pharmacotherapies may have additive effects with one study reporting that the combination of pregabalin and celecoxib was found to afford better pain relief than either agent alone [13], though adverse effects can also be additive. Table 1 summarizes the existing knowledge on benefits and adverse effects from various pharmacotherapeutic classes commonly used for LBP.

Table 1. Summary of therapeutic benefits and adverse effects of pharmacological options for patients with low back pain. Data is in absolute units on the 0–100 scale unless otherwise specified.

Pharmacological therapy	Effect size (95% CI) on a 0–100 scale for pain/SMD/RR	Level of evidence	Comments
Acetaminophen/ Paracetamol	0.0 (–9.7, 9.7) [17]	Low
NSAIDs	−6.97 (−10.74, −3.19) [14]	Low	No increase in AE with NSAIDs compared to placebo
Gabapentinoids	2.2 (−5, 7.7) [23]	Very low	A meta-analysis of 3 studies that compared gabapentin against placebo did not show a difference between gabapentin, pregabalin and placebo [23] but did not include a crossover study that showed superiority of gabapentin to pregabalin [24]
TCA	SMD: 0.10 (−0.51, 0.31) [18]	Low
SSRI	SMD: 0.11 (−0.17, 0.39) [18]	Low
SNRI	RR: 1.41 (1.11, 1.78)* [15]	Moderate
Muscle relaxants including BZD	−21.3, (−29.0, −13.5) [21]	Low	Data positive for acute pain relief; no evidence for long-term outcomes or radicular pain
Opioids	All opioids: −10.1 (−12.8, −7.4) for short-term benefit [27] SMD for Tramadol: −0.55 (−0.66, −0.44) SMD for transdermal buprenorphine: −2.47 (−2.69, −2.25) SMD for strong opioids: −0.43 (−0.52, −0.33) [26]	Moderate	Higher incidence of nausea, dizziness, constipation, vomiting, somnolence and dry mouth with opioids compared to placebo. No longer considered first-line treatment for any chronic pain condition.

*Likelihood of 50% or greater relief of low back pain after 12 weeks.

AE: adverse effects; BZD: benzodiazepines; NSAIDs: non-steroidal anti–inflammatory drugs; RR: relative risk; SMD: standardized mean difference; SNRI: serotonin–norepinephrine reuptake inhibitors; SSRI: selective serotonin reuptake inhibitors; TCA: Tricyclic antidepressants.

3.1. Acetaminophen (paracetamol)

Acetaminophen, an analgesic anti–inflammatory medication with an unclear mechanism of action, is often used by patients with LBP as a first-line medication. The frequent use of acetaminophen in LBP may reflect its wide availability, the ability of patients to access it without prescription, and a relatively favorable risk profile although patients with hepatic dysfunction and those consuming excessive amounts of alcohol may be at risk of complications. There is also a potential for cardiovascular, respiratory, and renal complications with long-term use [16,17]. Systematic reviews on the efficacy of acetaminophen demonstrate a lack of benefit in LBP for analgesic and quality of life-related outcomes [18].

3.2. Non-steroidal anti-inflammatory drugs

NSAIDs exert their analgesic anti–inflammatory effects through inhibition of cyclooxygenase (COX) enzymes. NSAIDs are very often prescribed for LBP with many members of this class available without a prescription. NSAIDs are recommended as first- or second-line options in most clinical guidelines on LBP [5]. Moderate-quality evidence shows a small reduction in intensity of LBP with NSAIDs and some improvement in function [18]. However, potential adverse effects on gastrointestinal, hepatic, cardiovascular, and renal systems, especially in the elderly, must be taken into account. Guidelines recommend the use of concomitant gastroprotective treatment with NSAIDs and the lowest effective dose for the shortest possible period of time [5]. There appears to be no difference in analgesic efficacy between different NSAIDs, or between COX-2 selective versus nonselective NSAIDs. There is mixed evidence for the use of NSAIDs in pain conditions with a predominantly neuropathic component including lumbosacral radicular pain, though these medications are often used to treat these conditions [19,20]. A Cochrane review evaluating NSAIDs for sciatica based on predominantly low-quality studies found non-significant differences between NSAIDs and placebo for pain relief, though NSAIDs were superior for global improvement [20].

3.3. Muscle relaxants including benzodiazepines

Muscle relaxants are often prescribed for LBP of suspected myofascial origin, with some drugs available in combination with acetaminophen. The class includes medications from different chemical origins that act on a variety of receptors. Examples include tizanidine (alpha-2 agonist), orphenadrine (muscarinic cholinergic antagonist, antihistaminic), carisoprodol (gamma-aminobutyric acid (GABA)-ergic), cyclobenzaprine (5-HT₂ receptor antagonist), baclofen (GABA-B agonist), and benzodiazepines. A systematic review on the efficacy of muscle relaxants (not including benzodiazepines) in LBP found only short-term analgesic benefits [21]. The evidence for benzodiazepines for LBP is not robust either, being stronger for acute than chronic pain. For neuropathic pain in general, there is no evidence to support any muscle relaxant with the possible exception of baclofen, which has not been studied for radiculopathy [22]. Muscle relaxants also have the potential for somnolence, dependence, and abuse. In summary, there is no supporting evidence for benefit with long-term use of muscle relaxants for LBP and these medications should be used for short periods of time in patients with severe and frequent muscle spasms accompanying LBP.

3.4. Gabapentinoids

Gabapentin and pregabalin are recommended as first-line treatments for neuropathic pain in most guidelines [19]. Gabapentinoids manifest their analgesic efficacy through action on the α-2 delta-2 subunit of the voltage-dependent calcium channels located on presynaptic receptors of afferent neurons. As neuropathic mechanisms are implicated in the causation and propagation of lumbosacral radicular pain and neurogenic claudication, gabapentinoids are often prescribed to patients. Systematic reviews that analyzed results of clinical trials on the role of gabapentinoids in chronic LBP with or without radicular pain reported no significant benefit when compared to placebo or other analgesics [23]. Furthermore, confusion, dizziness, fatigue, and visual disturbances were reported more often in patients on gabapentinoids [23]. There have also been concerns about misuse, diversion, and dependence with long-term use of these medications. These medications may have a theoretical role in patients whose LBP has a strong neuropathic component [24], but the number needed to treat is high. A medication trial period of up to 4 weeks at therapeutic doses followed by cessation in absence of benefit is reasonable.

3.5. Antidepressants

Tricyclic antidepressants (TCA) and serotonin–norepinephrine reuptake inhibitors (SNRI) are often prescribed for patients with LBP to provide analgesia, and alleviate depression, anxiety, and insomnia. However, there is conflicting evidence for their use. A recent systematic review of pharmacotherapy for LBP found no analgesic benefit with TCA or SSRI over placebo and unclear evidence on adverse effects. However, there was evidence of analgesic benefit and functional improvement with duloxetine compared to placebo, though the drug was associated with a greater risk of withdrawal due to adverse events [18]. For lumbar radicular pain, there is one small crossover trial demonstrating superiority of duloxetine over placebo [25]. There are few studies comparing antidepressants against each other for LBP. Similar to our suggestion regarding gabapentinoids, patients with LBP that has a strong neuropathic character may benefit from a trial of TCA and SNRI with cessation after two to 4 weeks if therapeutic doses do not provide pain relief.

3.6. Opioids

Opioids are the most commonly used prescription medication for chronic LBP, with one in five Americans using them to treat pain and a majority using them chronically [12]. Similar to all other analgesic medications, there is limited evidence in the form of controlled trials supporting opioid use beyond 4 months, though long-term use is associated with an increased risk of abuse and overdose. A systematic review of 15 trials in patients with LBP for at least 3 months who were taking opioids for at least 1 month found evidence of short-term efficacy (moderate for pain and small for function) compared with placebo [26]. Two other recent systematic reviews evaluating the short-term effects of opioids for LBP found moderate-quality evidence for analgesic benefits [18,27]. However, the median rate of adverse events for the eight trials that reported them in one of the reviews was 68.9% in the opioid group vs. 49.1% for placebo treatment [27]. For quality of life, a systematic review that included 19 randomized trials with at least 12 weeks of follow-up for chronic opioid therapy (11 for chronic LBP including 1 that followed patients for over 1 year) found evidence for improvement in physical, but not mental function [28]. A recent systematic review of clinical trials on the use of opioids for acute LBP found a significant association with long-term continued opioid use (Relative Risk: 1.57, 95% CI, 1.06–2.33) [29]. However, most pain societies and health organizations continue to state that opioids can be used safely for chronic, refractory LBP in carefully selected patients provided the dose is limited (e.g. to 90 morphine milliequivalents per day). These recommendations must be tempered by the awareness regarding lack of strong evidence for benefit with long-term opioid therapy and their potential for serious adverse effects.

4. Expert opinion

Choosing appropriate pharmacotherapy for acute and chronic LBP continues to pose a major challenge to health-care providers and sufferers. Identifying distinct phenotypes that respond to targeted treatments, improving diagnosis to enable mechanism-based rather than symptom-based treatment of pain (e.g. muscle relaxants for muscle spasms, antidepressants for LBP with a neuropathic character), treating concomitant conditions that may undermine effectiveness (e.g. sleep and psychological disorders) and exploring stepped care and interdisciplinary models, while monitoring for adverse effects are promising approaches that need to be studied for comparative-effectiveness against standard of care. The paucity of pharmacologic treatments for LBP has spurred a search for more efficacious and safer medications. Although they have not been extensively studied for back pain per se, intravenous infusion tests designed to identify pain mechanisms and predict response to oral analgesic therapy represents an area ripe for investigation [30]. Nerve growth factor inhibitors, cannabinoids, newer opioids, ketamine, and botulinum toxin have been evaluated in a variety of clinical studies for LBP. The evidence regarding the role of these therapies is mixed and of low quality. Table 2 provides a summary of some of the important studies on these agents in the last decade. Increasing evidence on newer options will help to optimize the treatment of this challenging and growing problem. It is also important to consider non-pharmacological options for LBP such as physical therapy and psychological treatments. Health-care providers should advise their patients with low back pain about non-pharmacological analgesic options that are supported by evidence of efficacy and safety, while informing them that many of them may not respond to currently available treatments, in which case functional restoration in the context of persistent pain may improve their quality of life.

Table 2. New pharmacotherapeutic agents evaluated in studies for treating low back pain with or without radicular pain.

1st Author, year (study type – number of participants)	Main results
Nerve growth factor inhibitors (tanezumab, fulranumab, fasinumab)
Katz 2011 (RCT – 216) Kivitz 2013 (RCT – 1347) Sanga 2011 (RCT – 389) Tiseo 2014 (RCT – 157)	Tanezumab, but not Fulranumab or Fasinumab, more efficacious than placebo or naproxen AE: arthralgias, headache, para/hyperesthesia, abnormal peripheral sensation, and dizziness
Cannabinoids (CBD, THC+CBD)
Mondello 2018 (Retrospective – 11) Silveira 2014 (Animal study −19)	The addition of THC-CBD resulted in effective pain management in all 11 people who had failed to respond to SCS for neuropathic FBSS Injection of CBD into inflamed intervertebral discs attenuated degeneration in response to induced injury AEs: dizziness, drowsiness, attention deficits, dry mouth, headaches, migraines, nausea, vomiting, palpitations, forgetfulness, urinary retention, fatigue, balance difficulty, myalgias, weakness, hot flushes
Bisphosphonates: denosumab, minodronate, pamedronate, zoledronate
Fujimoto 2017 (CS – 113) Koivisto 2014 (RCT – 40) Sakai 2014 (CS – 397) Tetsunaga 2017 (CS – 80) Pappagallo 2014 (RCT – 44)	Participants had LBP secondary to osteoporosis (without or with vertebral fractures) or degenerative spondylosis Newer bisphosphonates require less frequent administration and reduce intensity of LBP in the short term, decrease bone metabolism markers, and may increase bone mineral density Acute-phase reactions (fever, flu-like symptoms, arthralgia) seen with bisphosphonates
New opioids: cebranopadol
Christoph 2017 (RCT – 360)	Combines nociceptin/orphanin FQ peptide and opioid peptide receptor agonism Demonstrated analgesic efficacy, with significant and clinically relevant improvements in pain over placebo (comparable to tapentadol), along with both improved sleep and functionality in patients with LBP Low incidence of AE
Ketamine
Amr 2011 (RCT – 200) Patil 2012 (CS – 7)	Epidural S(+) ketamine enhanced analgesic effects of steroids and local anesthetics and reduced disability in patients with LBP and radicular pain Subanesthetic intravenous ketamine (mean dose 0.9 mg kg^−1) in severe refractory pain states reduced pain for over 3 weeks Epidural ketamine may be associated with neurotoxicity
Botulinum toxin
Machado 2016 (RCT – 37) Cogne 2017 (RCT – 19) Ney 2006 (CS – 60) Jabbari 2006 (CS – 75) Foster 2001 (RCT – 31)	Higher doses (>200 units per session with 40–50 units per site) may decrease intensity of LBP and improve quality of life Analgesic effect mediated by chemical denervation of local muscles and inhibition of release of inflammatory mediators and pain neurotransmitters Most studies were sponsored by industry Beneficial clinical response to the first injection predicts response to reinjection Patients on opioids do not respond as well as opioid-naive subjects AE: Mild flu-like reaction lasting a few days

AE: adverse effects; CBD: cannabidiol; CS: case series; FBSS: failed back surgery syndrome; LBP: low back pain; RCT: randomized controlled trial; SCS: spinal cord stimulation; THC: delta-9-tetrahydrocannabinol.

Declaration of interest

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

One referee declares that he was on the NICE accreditation advisory committee until March 2017 for which they received a fee. They are also an investigator on multiple previous and current research grants from the UK National Institute for Health Research (NIHR), Arthritis Research UK and is a co-investigator on grants funded by the Australian NHMRC. As an NIHR senior investigator, they receive research support from NIHR. They have received travel expenses, and honoraria, for speaking at conferences from the professional organizations hosting the conferences. They are an employee and shareholder of Clinvivo Ltd that provides electronic data collection for health services research. They are part of an academic partnership with Serco Ltd. They have accepted an honorarium from the consortium for advanced research training in Africa. They also provide editorial services for the NIHR journal series and serve as a member of the NIHR Journal editors Group for which they receive a fee. They also are a co-investigator on two studies, funded by the NIHR, that receive support in kind from Stryker Ltd. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

This work was supported in part by the Uniformed Services University, Department of Physical Medicine & Rehabilitation, Musculoskeletal Injury Rehabilitation Research for Operational Readiness (MIRROR) (HU00011920011).