OPIOID ANALGESICS

The efficacy and safety of long-term opioid analgesics specifically for CLBP has been reviewed recently, and much of the following discussion regarding efficacy appears from that paper.¹ There are now multiple randomized, controlled trials that consistently demonstrate efficacy and safety of opioid therapy for patients with CLBP.^2–11 The strongest evidence comes from numerous prospective randomized, placebo-controlled trials (PRPCT) that compared a particular opioid to placebo.¹² Katz et al. performed a 12-week enriched enrollment PRPCT to compare pain relief of oxymorphone-ER to placebo in 325 opioid-naive patients with CLBP.² During titration, 120 patients discontinued treatment for reasons that included lack of efficacy and adverse events. In the 205 remaining patients who obtained adequate analgesia and tolerable side effects during titration, pain intensity measured by visual analogue scale (VAS) decreased from 69 mm to 23 mm. These patients were subsequently randomized to continue oxymorphone-ER or placebo. After 12 weeks, the opioid group had statistically better pain numerical rating scores and patient satisfaction scales. There were 34 patients in the oxymorphone group who discontinued treatment for reasons that included lack of efficacy (12 patients) and adverse events (9 patients). In the placebo group, 53 discontinued treatment for similar reasons, lack of efficacy (35 patients) and adverse events. The authors concluded that oxymorphone-ER was safe and effective for opioid-naive patients with CLBP.

Hale et al. performed a 12-week enriched enrollment PRPCT to compare pain relief of oxymorphone-ER to placebo in 250 opioid-experienced patients with CLBP.³ Function was not addressed. During the titration period, 108 patients discontinued treatment due to adverse events (47), lack of efficacy (7), among other reasons. In 143 remaining patients who obtained adequate analgesia and tolerable side effects during titration, there were significant decreases in pain. These patients were subsequently randomized to continue oxymorphone-ER or placebo. After 12 weeks, there were clinically better outcomes in the oxymorphone-ER group for both numerical rating and patient satisfaction scales. There were 20 patients in the oxymorphone group who discontinued treatment due to lack of efficacy (8), adverse events (7), or other reasons. In the placebo group there were 55 who discontinued treatment due to lack of efficacy (39), adverse events (8) including opioid withdrawal (5), or other reasons. These authors also concluded that oxymorphone-ER was safe and effective for opioid-experienced patients with CLBP.

Hale et al. reported an 18-day multicenter randomized, double-blind, active and placebo-controlled comparison of oxymorphone-ER, oxycodone-CR, and placebo in 213 patients with CLBP.⁴ The mean change in pain intensity was significantly greater in both opioid-treated groups compared to placebo, and there were no clinically meaningful differences between opioids. Categorical pain ratings were most impressive. About 35% of the opioid-treated groups described their pain as absent or mild versus 12% in the placebo group. Sixty-one percent of the opioid groups reported moderate to complete pain relief versus 28% of the placebo group. Conversely, 45% of the placebo group described their pain as severe versus 14% in the opioid groups. There were statistically significant improvements in general activity, mood, normal work, and enjoyment of life, but not walking ability. During the titration phase about 15% of opioid-treated patients withdrew due to side effects and 4% due to lack of efficacy. In the treatment phase, 25–33% of the opioid groups withdrew due to adverse effects. There were no instances of addictive behavior. Side effects were common, but only sedation and constipation were more frequent in the opioid groups compared to controls.

Peloso and associates performed a 91-day randomized double-blind, placebo-controlled study comparing the efficacy and safety of flexible dose tramadol 37.5 mg plus acetaminophen (APAP) 325 mg to placebo in 338 patients with CLBP.⁵ The active treatment group had statistically and significantly better improvements in pain (VAS) and function (Roland Disability Questionnaire) compared to placebo patients. About 49% of the active treatment patients had ≥30% reduction in pain and 49% had ≥50% relief. The number needed to treat was 4 for ≥30% relief and 5 for ≥50% relief.

Schnitzer et al. reported a prospective randomized, controlled 4-week trial that compared tramadol to placebo in 254 patients with CLBP who had had been shown to be tramadol responders in the open-label phase of the study.⁶ The tramadol group had significantly greater improvements in pain (VAS) and function (Roland) compared to the control group. Ruoff and associates reported a 91-day PRCT that compared tramadol plus acetaminophen to placebo in patients with CLBP.⁷ The active treatment group had significantly better pain (VAS) and function compared to placebo.

There are also numerous prospective, randomized active-control trials (PRACT) that compared opioids to one another.8–11,^13,14 Rauck et al. performed an 8-week open-label PRACT to compare effectiveness and safety of a once-a-day morphine sulfate-SR (Avinza®) with twice daily oxycodone-CR (Oxycontin®) in 392 patients with moderate to severe CLBP.⁸ Both groups had statistically and clinically significant reductions in pain. Although there were slightly better outcomes in the morphine group, the authors recognize that the study protocol mandated 12-hour dosing of the oxycodone-CR rather than 8-hour dosing that is more often necessary. This may have biased the results slightly in favor of the morphine group. Numerical rating scale (NRS) decreased from a mean of about 6.5 to a mean of 3.4–3.7. The morphine-SR group had somewhat smoother pain control. Function was not addressed. About 32–43% of patients withdrew, most often due to side effects, but also due to inadequate pain relief. There were four instances of abuse or diversion in the oxycodone-CR group.

Allan et al. performed a 13-month unblended, randomized parallel group study to compare titrated doses of transdermal fentanyl (TDF) to morphine-SR in 680 patients with CLBP.⁹ Doses of drugs were titrated according to individual patient response. The TDF and morphine-SR produced similar results. Depending on level of activity, 50–65% of patients described themselves as improved, and 37–53% of patients had at least 50% reduction in pain. There were significant improvements in mean SF-36 scores for physical functioning, bodily pain, role-physical, vitality, social function and role-emotional. About 31–37% of patients withdrew the due to adverse events. Over the 13 months of the study, opioid doses increased only slightly, usually early in treatment, and were attributed to titration to optimal dose rather than tolerance. There were no instances of addiction or abuse behavior.

Hale et al. performed a 10-day randomized, double-blind trial to compare the efficacy and safety of titrated doses of oxycodone-CR and to oxycodone-IR in 47 CLBP patients.¹⁰ There were equal and significant improvements with both formulations. Pain intensity decreased from moderate to severe at baseline to slight at the end of titration with both oxycodone formulations. Eleven patients (23%) withdrew due to side effects.

Jamison et al. performed a 16-week randomized, prospective study to compare naproxen, fixed dose oxycodone, and a titrated dose of oxycodone plus morphine-ER in 36 patients.¹¹ All three groups had improvement in pain, activity level, and emotional distress. The titrated dose opioid group did best with less pain and less emotional distress than the other two groups. Both opioid groups were better than the naproxen group. There were 86% who found opioids beneficial. Three patients in the opioid groups withdrew due to side effects and one patient had an abuse problem.

In addition, there are multiple prospective observational studies that have shown consistent findings of efficacy and safety for opioids in patients with CLBP. Simpson et al. found statistically significant improvement in pain in 50 patients changed to TDF compared to their prior regimens of pain-contingent oral opioids for chronic LBP.¹³ Gammaitoni et al. prospectively studied 33 patients with CLBP treated with titrated doses of oxycodone-IR plus APAP three times daily for 4 weeks.¹⁴ Three patients were not able to tolerate the oxycodone and two others withdrew for other reasons. The mean NRS was reduced from 6.4 to 4.4 and worst pain from 7.7 to 5.6. There were significant improvements in general activities, mood, walking tolerance, and sleep. Side effects were common, but there were no serious adverse effects. There were no instances of addictive behavior or other abuse.

Recently, Martell and associates published a systematic review of opioid treatment for CLBP.¹⁵ Despite the fact that the authors did not include several PRPCTs that showed both efficacy and safety of opioids for CLBP, they still concluded that opioids ‘may be efficacious for short-term pain relief.’^2,6,7,9 They also state that ‘long-term efficacy is unclear.’ However, they did not include the study with the longest duration of 13 months⁹ and gave less weight to other long-term studies that are of lower quality but provide the best available evidence.^16,17

While it is clear that the evidence regarding long-term efficacy and safety of opioid use is not strong, there are several prospective observational and retrospective studies that are consistent in showing long-term efficacy and safety and no studies that demonstrate loss of effectiveness over time unless there is disease progression.^16,17 Schofferman reported a prospective case series of 33 patients with refractory CLBP treated with opioids for 1 year.¹⁶ There were no control or comparison groups. The specific opioid was selected by response during the trial titration phase. Five (15%) patients withdrew because of side effects. In the remaining 28, there were statistically and clinically significant improvements in pain and function at one year. The mean NRS improved from 8.6 to 5.9 and mean Oswestry Low Back Disability Index (OSI) from 64 to 54. There was a biphasic response. In 21 patients there was an improvement of NRS from 8.45 to 4.9, while 7 others had no change. Overall, of the 33 patients who started the study, opioids were beneficial in 21 (64%).

Mahowald et al. reviewed their experience with opioid use over a period of 3 years in an orthopedic spine clinic.¹⁷ Opioids were prescribed for 152 patients, 58 of whom received opioids long term. There was adequate follow-up in 117. Pain was reduced from a mean of 8.3 to 4.5. It is noteworthy that there was no significant dose increase over time and the authors stated that they did not see tolerance in their patients. Side effects were common but well tolerated by most patients. There was a low prevalence of abuse. The authors concluded that there was clinical evidence to support treating CLBP patients with opioids.

There are multiple studies and systematic reviews that examined the efficacy and safety of opioid analgesics for the treatment of chronic musculoskeletal pain, all of which included but were not limited to patients with CLBP. Furlan’s meta-analysis of opioids for chronic pain concluded that opioids were more effective than placebo for pain and functional outcomes in patients with nociceptive pain, including CLBP.¹⁸ With respect to side effects, only nausea and constipation were clinically and statistically significantly greater in the opioid than placebo patients. Study withdrawal rates averaged 33% in the opioid groups and 38% in placebo groups.

The MONTAS study group performed a 2-week prospective, randomized double-blind, placebo-controlled trial of MS-ER in 48 patients, 12 of whom had refractory CLBP.¹⁹ In 17 (35%) patients, pain was reduced from a mean of 7.4 to 2.9 (excellent responders). In 17 others, pain was reduced from 7.8 to 5.6 (partial responders). In 14 (29%) (non-responders) there was either no meaningful relief or intolerable side effects. Markenson et al. performed a 90-day RCT comparing oxycodone-CR to placebo in 107 patients with moderate to severe osteoarthritis, 40–50% of whom had CLBP.²⁰ There were statistically significant differences favoring the oxycodone-CR group versus controls in pain intensity, pain-induced interference with general activity, walking, work, mood, sleep, and enjoyment in life. The improvements in pain were only modest. The discontinuation rate was similar between groups, either due to inadequate pain control or side effects.

In summary, there is sufficient evidence to make a Grade A recommendation that opioid analgesics are safe and effective for the treatment of patients with CLBP, at least in the short term. In all high-quality studies there is clinically and statistically significant improvements in pain and the results are uniformly better for opioid than placebo. There is sufficient but less robust evidence to state that opioids retain their effectiveness over the longer term and no reports of loss of efficacy over time. About 10–20% of patients do not tolerate opioids due to side effects. Of those patients who tolerate opioids, about one-third are excellent responders, one-third fair responders, and one-third non-responders. The evidence for improvement of function with opioids is also good, but not as strong as for improvement in pain.

There is no evidence to show any opioid to be best. Each opioid that has been studied has been shown to be more effective than placebo and clinically effective for change in LBP. However, because there is individual patient variability in opioid responsiveness, it is not uncommon to have to rotate opioids to find the best one for an individual patient. Opioids are generally safe. Toxicity is rare. The incidence of diversion, addictive behavior, or other social problems is low. Side effects are common, but are usually manageable.

OPIOIDS: CELLULAR MECHANISM OF ACTION

Opioids bind with delta, kappa and mu receptors in the central nervous system (CNS) and peripheral tissues to provide analgesic effects. Mu receptors are located in the CNS and provide central analgesia, but also play a role in the development of respiratory depression, physical dependence, and withdrawal symptoms. Most commonly used opioids such as morphine, codeine and fentanyl are mu-agonists.

The majority of kappa receptors are located in the cerebral cortex and substantia gelatinosa of the dorsal horn. Kappa receptors are responsible for analgesia at the level of the spinal cord, the brain, and also kappa-mediated dysphoria. Kappa receptors have less of a role in physical dependence and withdrawal. Delta receptors are concentrated in the substantia gelatinosa of the dorsal horn and have a primary effect upon spinal and supraspinal analgesia.

There is evidence that points to the involvement of N-methylD-aspartate (NMDA) receptors in the development and maintenance of neuropathic pain.²¹ It follows that NMDA receptor antagonists may be helpful in the treatment of neuropathic pain. However, in order to obtain complete analgesia, a combination of an NMDA receptor antagonist and an opioid receptor agonist may be needed. In vitro data have demonstrated some medications (such as methadone), in addition to being opioid receptor agonists, are also weak noncompetitive NMDA receptor antagonists.²¹

CLINICAL CONCERNS REGARDING LONG-TERM OPIOID THERAPY

Chronic pain is undertreated. There are many complex reasons for this, most of which are based on anecdotes and social pressures. Some of the circumstances that contribute to the prevalence of undertreated pain include:

Lack of knowledge of medical standards, current research and clinical guidelines for appropriate pain treatment;

The perception that prescribing adequate amounts of controlled substances will result in unnecessary scrutiny by regulatory authorities;

Misunderstanding of addiction and dependence;

Lack of understanding of regulatory policies and processes.

Some physicians continue to have a bias against the use of long-term opioid analgesics for chronic pain. The most prevalent concerns are organ toxicity, tolerance, dependency, addiction and fear of legal consequences, and at least in theory the development of opioid-induced hyperalgesia. In addition, there is a concern that the side effects may outweigh benefits.

Tolerance

Tolerance is the need for progressively higher doses of an analgesic to produce the same degree of pain relief, which would translate into partial loss of opioid efficacy over time. True tolerance is a biological process at the cellular level based on downregulation of opioid receptors, desensitization of receptors or both.²² True tolerance is not common in the clinical setting. Once the proper balance of pain control, function, and side effects has been reached; there is rarely dose escalation unless there is disease progression. Early in treatment it is common to note ‘pseudotolerance,’ the need for increased dosage due to an increase in function, which in turn is due to better pain control. The development of pseudotolerance requires rebalancing the opioid dose for maximal pain control and function with minimal adverse effects. When tolerance appears to occur well after stable doses have been reached, it is most likely due to diseases progression. However, the other factors that must be considered include lack of compliance, change in other medications with unexpected drug interaction and diversion of medications (Fig. 14.2).²³

Fig. 14.2 Pseudotolerance. NRS is numerical rating score for pain. A lower score indicates less pain. OSI is Oswestry Disability Index for function. A lower score indicates better function.