Introduction
Chronic low back pain is a common, debilitating, and costly health problem (Centers for Disease Control and Prevention, 2001; Collins et al., 2005; Mapel et al., 2004; Vogt et al., 2005). The prevalence of chronic low back pain is higher in women and in whites versus blacks (Andersson, 1999; Mapel et al., 2004). Although the risk of chronic low back pain increases with age (Mapel et al., 2004), back pain remains the most common cause of disability in adults aged <45 years (Andersson, 1999).
Treatment guidelines issued in the United States (Chou et al., 2007) and Europe (Airaksinen et al, 2006) both state that back pain becomes chronic if it persists for >12 weeks. United States guidelines also distinguish between acute (<4 weeks) and subacute (4-<12 weeks) back pain (Chou et al., 2007). The 12-week threshold for classifying back pain as chronic makes sense given data suggesting that 80% to 95% of patients with disabling back pain can return to normal activities within 12 weeks, with less certain outcomes thereafter (Andersson, 1999).
Treatments for chronic low back pain include nonpharmacologic therapies (eg, exercise, lifestyle modification), which are discussed elsewhere in this topic. First-line pharmacotherapies include acetaminophen, nonsteroidal anti-inflammatory drugs (NSAIDs), weak opioids, and strong opioids. Each of these therapies, including opioids, may be initiated during the acute phase of back pain, depending on the severity of pain (Chou et al., 2007). However, the course of chronic low back pain may be protracted (Andersson, 1999), meaning patients may require treatment over years or decades. It is therefore important that pharmacological treatments be effective and as safe as possible both in the short term and in patients requiring long-term treatment. Acetaminophen is a well-tolerated first-line pharmacotherapy but has limited efficacy (Zhang et al., 2010) and high-dose, long-term use is associated with hepatic toxicity (Watkins et al., 2006). NSAIDs have better efficacy than acetaminophen but have well-known risks of gastrointestinal (Boers et al., 2007; Gabriel et al., 1991; Hippisley-Cox et al., 2005), cardiovascular (Antman et al., 2007; Caldwell et al., 2006; Hippisley-Cox & Coupland, 2005; Kearney et al., 2006; Motsko et al., 2006), and other systemic adverse effects that increase with age, dose, and duration of use.
The serotonin norepinephrine reuptake inhibitor (SNRI) duloxetine has received US Food and Drug Administration (FDA) approval for the treatment of musculoskeletal pain such as chronic low back pain (Cymbalta® Delayed-Release Capsules, 2010). Duloxetine has only modest efficacy (Skljarevski et al., 2010a; Skljarevski et al., 2009), is associated with systematic adverse events (Skljarevski et al., 2010a; Skljarevski et al., 2009), and may also interact with other analgesics, most notably certain opioids (Smith, 2009).
Opioids have the greatest efficacy of any oral therapy for relieving pain but also have significant risk of adverse events and potential for pharmacokinetic drug interactions (Malhotra et al., 2001; Tulner et al., 2008) and carry a substantial risk of addiction and abuse. Chronic opioid therapy requires long-term monitoring for treatment compliance.
This topic will review guideline recommendations and clinical evidence for pharmacotherapies available for the management of chronic low back pain.
Review of guidelines for chronic low back pain and clinical evidence for pharmacotherapies
Multimodal approach to therapy
United States guidelines for the management of chronic low back pain recommend that effective pharmacotherapy be administered in conjunction with self-care options (eg, application of heat, continued activity, adoption of physical fitness regimens, use of a medium-firm mattress, lifestyle modifications) (Airaksinen et al., 2006; Chou et al., 2007). Nonpharmacologic therapies are discussed in detail elsewhere in this topic. For patients not responding to medication and self-care, clinicians should employ other nonpharmacologic options, including intensive interdisciplinary rehabilitation, exercise therapy, acupuncture, massage therapy, spinal manipulation, yoga, cognitive behavioral therapy, and progressive relaxation (Airaksinen et al., 2006; Chou et al., 2007).
Acetaminophen
United States Guidelines for the management of chronic low back pain recommend acetaminophen as a first-line therapy for mild to moderate pain (Chou et al., 2007); the recommendations are based on modest efficacy and overall favorable tolerability in patients with osteoarthritis. European guidelines do not recommend acetaminophen (Airaksinen et al., 2006).
Randomized controlled trials of acetaminophen for chronic low back pain are lacking. Both US and European guidelines acknowledge that acetaminophen has been less effective than oral NSAIDs for osteoarthritis pain (Airaksinen et al., 2006; Chou et al., 2007), and effect sizes for acetaminophen calculated by the Osteoarthritis Research Society International are below the threshold for a clinically meaningful analgesia (Zhang et al., 2010). Acetaminophen is not as safe as it was once believed to be, particularly for a chronic condition such as chronic low back pain that may require treatment for many years and may also require dose escalation as the disease progresses over time. At doses >4 g/ d, acetaminophen is known to cause liver enzyme increases >3-fold greater than the upper limit of normal in healthy volunteers (Watkins et al., 2006). The FDA is currently considering lowering the maximum recommended dose to 3.25 g/day (Kuehn, 2009). In the United States, acetaminophen is the leading cause of acute liver failure (Chun et al., 2009).
Overdose of acetaminophen may be accidental (Camidge et al., 2003; Larson et al., 2005), resulting from patients taking too much without realizing there is a dosage ceiling, or from patients taking a product that combines a narcotic with acetaminophen and then augmenting it with additional acetaminophen (Larson et al., 2005). Overdose may also be deliberate (Camidge et al., 2003; Larson et al., 2005). Despite its reputation as a safe drug, acetaminophen is among the drugs implicated most frequently in suicide attempts (Camidge et al., 2003; Larson et al., 2005). Also, thought not typically believed to carry substantial potential for pharmacokinetic interactions, acetaminophen-associated liver failure is significantly associated with alcohol abuse and antidepressant use (Larson et al., 2005).
Nonsteroidal anti-inflammatory drugs
A Cochrane review of 65 randomized controlled trials found evidence that NSAIDs are effective for the management of chronic low back pain pain, but effect sizes were indicative of only a small treatment effect (Roelofs et al., 2008). The evidence presented in this Cochrane review suggested that different NSAIDs, including cyclooxygenase-2 inhibitors, show similar efficacy in patients with chronic low back pain.
Evidence from meta-analyses provide evidence that NSAIDs are moderately more effective than acetaminophen (Lee et al., 2004; Roelofs et al., 2008), and chronic low back pain guidelines also state that NSAIDs are more effective (Airaksinen et al., 2006; Chou et al., 2007). In patients with osteoarthritis, effect sizes for NSAIDs are substantially higher than with acetaminophen but nonetheless correspond to a small treatment effect (Zhang et al., 2010).
Nonsteroidal anti-inflammatory drugs are associated with well-known risks of gastrointestinal (Boers et al., 2007; Gabriel et al., 1991; Hippisley-Cox et al., 2005), cardiovascular (Antman et al., 2007; Caldwell et al., 2006; Hippisley-Cox & Coupland, 2005; Juni et al., 2004; Kearney et al., 2006; Motsko et al., 2006), and renal (Barkin & Buvanendran, 2004; Evans et al., 1995) adverse events. Hepatic adverse events have been reported infrequently (Rostom et al., 2005; Rubenstein & Laine, 2004), and are certainly less common than with acetaminophen. NSAIDs have been associated with potentially clinically meaningful increases in blood pressure (Forman et al., 2005; Johnson et al., 1994; Pavlicevic et al., 2008), probably both through direct effects (Forman et al., 2005; Johnson et al., 1994) and adverse pharmacokinetic interactions with diuretics and angiotensin-converting enzyme inhibitors (Pavlicevic et al., 2008). Drug interactions of NSAIDs with antihypertensives (angiotensin-converting enzyme inhibitors, diuretics), selective serotonin reuptake inhibitors (SSRIs), and corticosteroids (American Geriatrics Society, 2009; Malhotra et al., 2001; Tulner et al., 2008) are common. Ibuprofen can nullify the cardioprotective effects of aspirin (Gengo et al., 2008). Gastrointestinal bleeding events may occur in patients receiving an NSAID with warfarin (Cheetham et al., 2009), an SSRI, or a corticosteroid (American Geriatrics Society, 2009). Proton pump inhibitors administered as gastroprotection in NSAID-treated patients can block the cardioprotective effects of clopidogrel (Ho et al., 2009; Juurlink et al., 2009).
Cyclooxygenase-2 inhibitors have been associated with fewer adverse events than nonselective agents (Roelofs et al., 2008), and in particular have shown improved gastrointestinal tolerability (Singh et al., 2006). In a 12-week, 13,000 patient trial, the occurrence of abdominal pain, dyspepsia, diarrhea, headache, and nausea was reported in 2% to 5% of patients treated with celecoxib and 3% to 6% of those treated with naproxen or diclofenac (Singh et al., 2006). Confirmed gastrointestinal bleeding events were also less frequent with celecoxib compared with nonselective agents, but occurred in 1% of patients compared with 2.1% treated with the nonselective agents.
Given their risk/benefit ratio, guidelines for the treatment of chronic low back pain (Airaksinen et al., 2006; Chou et al., 2007) and osteoarthritis (Zhang et al., 2007; Zhang et al., 2008) recommend that oral NSAIDs be administered for the shortest period possible at the lowest effective dose. Guidelines for the management of chronic pain in the elderly suggest that regular oral NSAID use should be avoided if possible (American Geriatrics Society, 2009).
Topical therapies
United States guidelines for the management of chronic low back pain do not recommend topical therapies. European guidelines recommend only capsaicin plaster for short-term therapy (<3 wk) in patients with chronic low back pain (Airaksinen et al., 2006).
The lidocaine 5% patch
The lidocaine 5% patch has US approval only for the management of postherpetic neuralgia (LIDODERM®, 2010), particularly with allodynia (Attal et al., 2010). However, the lidocaine 5% patch has also shown efficacy in patients with chronic lower back pain (Galer et al., 2004; Gimbel et al., 2005). In a 6-week open-label trial, 71 patients with acute/subacute (n=11), short-term chronic (n=17), or long-term chronic (n=43) nonradicular back pain applied up to 4 patches to the maximal area of pain (Galer et al., 2004). All patient subgroups reported significant improvements in scores on the Neuropathic Pain Scale. In a second 6-week open-label study, 131 patients with subacute (n=21), short-term chronic (n=33), and long-term chronic (n=77) back pain reported significant mean improvements in scores on the Brief Pain Inventory (Gimbel et al., 2005) and significant improvements in the Brief Pain Inventory item for interference with quality of life. Beck Depression Inventory scores also improved significantly. Approximately two thirds of patients and investigators reported being Satisfied or Very Satisfied with treatment.
Application of up to 4 lidocaine 5% patches daily results in peak systemic lidocaine concentrations <200 ng/mL, or >7-fold less than the concentration of >1500 ng/mL required to produce cardiac effects and 20-fold less than the threshold concentration for toxic effects (Gammaitoni et al., 2003). Consistent with low systemic exposure, the most common adverse events reported in a trial of the lidocaine 5% patch in patients with chronic low back pain were mild to moderate skin and subcutaneous tissue reactions (eg, rash, pruritus, dermatitis, erythema, edema). Treatment-related systemic adverse events were relatively uncommon and included dizziness (4%), headache (2%), and nausea (2%-3%) (Gammaitoni et al., 2003).
The capsaicin 8% patch
In a double-blind, randomized trial, 320 patients with chronic low back pain were assigned to apply the capsaicin 8% patch or placebo patch once daily for 3 weeks (Frerick et al., 2003).
The capsaicin 8% patch was associated with a 42% reduction in pain assessed using the Arhus low back rating scale versus a 31% reduction with placebo. Sixty-seven percent of the capsaicin 8% patch group and 49% of the placebo group met response criteria, defined as a >30% reduction in pain. Though statistically significant, these differences compared with placebo are very modest. These results were similar to those obtained with a similar capsaicin 8% patch in a similarly designed trial (Keitel et al., 2001). The most common adverse events were application site warmth or itching, although inflammatory contact eczema, urticaria, small hemorrhagic spots, vesiculation, and dermatitis have been reported in a small number of patients (Frerick et al., 2003; Keitel et al., 2001). Given modest efficacy and generally good tolerability, the capsaicin 8% patch may only be useful as adjunctive therapy or as monotherapy in a small subgroup of patients.
Topical nonsteroidal anti-inflammatory drugs
Topical NSAIDs produce dramatically lower systemic NSAID concentrations compared with oral NSAIDs (Kienzler et al., 2010) and are recommended as a first-line therapy in patients with osteoarthritis of the knees or hands (Zhang et al., 2007; Zhang et al., 2008), particularly the elderly (American Geriatrics Society, 2009). However, no chronic low back pain guidelines recommend topical NSAIDs and chronic low back pain is not an approved indication for either of the two topical NSAID formulations used in the United States: diclofenac sodium 1% gel (Voltaren® Gel 1%, 2007) and diclofenac sodium 1.5% in 45.5% dimethyl sulfoxide solution (Pennsaid®, 2009). Neither formulation has been evaluated in patients with chronic low back pain.
Topical diclofenac diethylamine 1.16% gel and an indomethacin plaster were evaluated in 64 patients with mild to moderate, nonsurgical chronic low back pain (Waikakul et al., 1996). Both formulations provided statistically significant improvements in pain and function; however, topical NSAIDs have not been evaluated in clinical trials since this 1996 study.
Serotonin-norepinephrine reuptake inhibitors
Duloxetine has shown modest efficacy in two 12-week trials (Skljarevski et al., 2010a; Skljarevski et al., 2009) and in a 41-week open-label study (Skljarevski et al., 2010b). Active treatment was superior to placebo in both 12-week trials when a 30% improvement in pain was defined as a positive response (Skljarevski et al., 2010a; Skljarevski et al., 2009) but only in 1 trial when a 50% improvement was defined as a positive response (Skljarevski et al., 2010a). Given modest efficacy, duloxetine may be a helpful adjunctive therapy (discussed below).
Weak opioids
The weak opioids codeine, hydrocodone, and propoxyphene have not been studied extensively in patients with chronic low back pain. A single 1976 study of propoxyphene showed efficacy in patients with chronic low back pain (Baratta, 1976) and a 1998 study of an extended release formulation of codeine combined with acetaminophen demonstrated efficacy similar to tramadol (Muller et al., 1998). Hydrocodone has not been studied in patient with chronic low back pain. Nonetheless, a meta-analysis showed that weak opioids do not have a clear efficacy advantage compared with oral NSAIDs or tricyclic antidepressants (Furlan et al., 2006). Several safety issues limit use of weak opioids. These agents are commonly combined with acetaminophen, imposing a dosage ceiling on their use (Victor et al., 2009). Codeine has been associated with a high rate of constipation. Moreover, though a weak opioid, propoxyphene is associated with more potent respiratory and cardiovascular depression than are strong opioids (Barkin et al., 2006; Ulens et al., 1999).
Combination weak opioid-norepinephrine reuptake inhibitors
A Cochrane review of 3 trials found that tramadol is more effective than placebo for treating pain and functional impairment (Deshpande et al., 2007). It should be observed that 2 of the 3 trials analyzed in the Cochrane review evaluated products combining tramadol with acetaminophen (Peloso et al., 2004; Ruoff et al., 2003). Although pure opioids theoretically have no analgesic dosage ceiling, the presence of acetaminophen imposes a dosage ceiling on the product because acetaminophen is hepatotoxic above a daily dose of 4 g (Watkins et al., 2006). Tramadol is not a pure opioid agonist but rather also has serotonergic effects that limit the recommended daily dose to 300 mg (Ultram ER®, 2009). Serotonergic effects of tramadol increase the risk of serotonin syndrome in patients receiving other serotonergic drugs, and must not be combined with SNRIs, SSRIs, and tricyclic antidepressants (Ultram ER®, 2009).
Two 6-week trials found that tramadol monotherapy was less effective than the selective cyclooxygenase-2 inhibitor celecoxib (O’Donnell et al., 2009). Likewise, tramadol was found to be no more effective than extended-release codeine, even though tramadol has the additional noradrenergic mechanism of action. The most common adverse events reported in the 3 trials of tramadol with or without acetaminophen were typical opioid effects, including nausea (8.7%-13%), headache (4.7%-6.6%), somnolence (9%-12.4%), dizziness (7.5%-10.8%), dry mouth (6%-8.1%), vomiting (6%), and constipation (10.2%-11.2%) (Peloso et al., 2004; Ruoff et al., 2003; Schnitzer et al., 2000).
Tramadol relies on the cytochrome P450 enzyme 2D6 for metabolism to its active metabolite, O-desmethyltramadol, giving it potential for pharmacokinetic interaction with drugs metabolized via this pathway. Five to 15% of the white population is classified as having either rapid or slow cytochrome P450 enzyme 2D6 metabolizers, meaning that some patients will have insufficient or increased sensitivity to both analgesic and adverse effects of the drug (Smith, 2009). Collectively, tramadol appears to offer no efficacy advantage compared with weak opioids despite its secondary norepinephrine reuptake inhibitor activity.
A second weak opioid-SNRI combination, tapentadol, differs from tramadol in several important respects. Tapentadol is metabolized via glucuronidation and therefore lacks the potential for pharmacokinetic interaction and variability response observed with tramadol (Kneip et al., 2008). Its noradrenergic effects are more potent than its serotonergic effects (Tzschentke et al., 2007), theoretically reducing the risk of serotonergic adverse events, although in the United States tapentadol has an FDA "black box" warning against its combined use with SSRIs or SNRIs. Analgesic effects of SNRIs are attributed primarily to the noradrenergic effects (Max et al., 1992); hence, reduced activity with respect to serotonergic function decreases adverse events without compromising analgesic efficacy.
In a prospective, randomized, double-blind, trial, 981 patients with chronic low back pain received tapentadol 100 to 250 mg/d, oxycodone controlled release, or placebo for a 3-week titration period followed by a 12-week maintenance period. Tapentadol and oxycodone produced similar, statistically significant improvements in pain compared with placebo. Tapentadol was associated with a lower occurrence compared with oxycodone of total adverse events, constipation, nausea, and vomiting. A pooled analysis of this study and two 12-week studies of patients with osteoarthritis (n=2010) demonstrated that tapentadol was as effective as oxycodone controlled release in improving mean pain intensity, but more people treated with tapentadol experienced 30% and 50% reductions in pain (Lange et al., 2010). Only tapentadol was associated with significant improvements in quality of life, which may reflect superior tolerability with tapentadol. Tapentadol exhibited a superior gastrointestinal tolerability compared with oxycodone (Lange et al., 2010). A subsequent subanalyses of these trials found that the lower incidence of constipation with tapentadol (13%-18%) compared with oxycodone (27%-36%) should be associated with a lower rate of work absenteeism and productivity loss in tapentadol-treated patients, based on previous research correlating constipation with productivity (Cepeda et al., 2011).
A limitation of studies comparing tapentadol with oxycodone controlled release was the dosage ranges used. Patients assigned to oxycodone received a dose of 20 to 50 mg twice daily (Lange et al., 2010), which is well below the effective dose for many patients. All patients, including opioid-experienced patients, were started on oxycodone 10 mg. The maximum dose of 50 mg is much less than the maximum doses reported in oxycodone trials (Rauck et al., 2006a). Nonetheless, if patients were underdosed with respect to analgesia, they also would be expected to have more adverse events at higher, more effective doses. Thus, at optimal oxycodone doses with respect to analgesia, differences in tolerability favoring tapentadol might be even more pronounced.
Given statistically significant efficacy compared with placebo, a trial of a weak opioid-SNRI combination drug before prescribing a strong opioid may be a valuable option in many patients.
Strong opioids: Efficacy
Despite increasing use in patients with chronic noncancer pain in recent years, opioids have not been studied extensively in patients with chronic low back pain. A 2007 Cochrane review of opioids for chronic low back pain (Deshpande et al., 2007) included only 1 trial of a strong opioid (Jamison et al., 1998). This 1998 trial found that oxycodone immediate release and oxycodone combined with morphine extended release were both significantly superior to naproxen in patients with moderate to severe chronic low back pain (Jamison et al., 1998). Similarly a 2006 meta-analysis of opioids in chronic noncancer pain included the same single study of oxycodone in chronic low back pain and reached the same conclusion (Furlan et al., 2006). A 2011 meta-analysis included 4 additional studies, one of tramadol, 2 of oxymorphone extended release, and 1 of oxycodone with ultra-low-dose naltrexone (Kuijpers et al., 2011).
Oxymorphone extended release was evaluated in two 12-week randomized controlled trials in opioid-naive and opioid-experienced patients with moderate to severe chronic low back pain (Hale et al., 2007; Katz et al., 2007). Results from these studies demonstrated that use of a flexible, individualized titration schedule will allow the majority (60% in the 2 studies combined) of patients to be titrated to an effective, generally well-tolerated opioid dose (Peniston & Gould, 2009). The trial in opioid-nai’ve patients demonstrated that the lowest available doses of oxymorphone extended release can be administered safety in patients who have never received an opioid. In contrast, some long-acting opioids are not appropriate as a starting medication for opioid-naive patients; for example, the fentanyl transdermal patch and hydromorphone OROS (discussed following) (Duragesic®, 2009; EXALGO complete prescribing information, 2010). The trial of oxymorphone extended release in opioid-experienced patients demonstrated that patients who grow tolerant to one opioid formulation can be successfully transitioned to another opioid with ostensibly the same mechanism of action and experience adequate pain relief with acceptable tolerability.
This finding is important given the clinical observation that patients requiring long-term opioid therapy typically require successive trials of different opioids to find one that works, and that even after finding the right opioid, will frequently grow tolerant to the initial opioid and need to be rotated to another opioid to maintain analgesia or avoid adverse events that emerge with dose escalation (Grilo et al., 2002; Mercadante & Bruera, 2006; Quang-Cantagrel et al., 2000). It is important to have multiple effective opioids from which to choose an initial agent and have options for switching as tolerance develops (Slatkin, 2009). It remains to be determined whether after switching from an initial opioid to another, responsiveness to the original opioid will return over time to allow a switch back if tolerance to the second opioid emerges. In both opioid-nai’ve and -experienced patients, oxymorphone extended release analgesia and functional improvement remained significantly superior to placebo for the entire 12-week treatment period without dose escalation (Hale et al., 2007; Katz et al., 2007). Of patients successfully titrated and randomized to oxymorphone extended release, 69% completed 12 weeks of treatment (Peniston & Gould, 2009).
Oxycodone has been formulated with ultra-low-dose naltrexone in order to minimize adverse events (eg, respiratory depression, constipation, physical dependence) (Amass et al., 2000; Chindalore et al., 2005; Webster et al., 2006). In a double-blind trial, 719 patients with moderate to severe chronic low back pain received placebo, oxycodone immediate release 4 times daily, oxycodone with naltrexone 4 times daily, or oxycodone plus naltrexone twice daily for 12 weeks (Webster et al., 2006). All patients initiated treatment with a total daily oxycodone dose of 10 mg and were titrated over a period of 1 to 6 weeks to a dose that provided effective analgesia with acceptable tolerability or to a maximum daily dose of 80 mg. All oxycodone regimens provided significantly superior analgesia compared with placebo. The mean daily oxycodone dose was 12% lower with the oxycodone plus naltrexone regimens than with oxycodone. Compared with oxycodone administered 4 times daily, oxycodone plus naltrexone twice daily was associated with 44% less constipation, 33% less somnolence, and 51% less pruritus. Upon abrupt opioid discontinuation, the severity of withdrawal symptoms assessed using the Short Opiate Withdrawal Scale was 55.8% less with oxycodone plus naltrexone twice daily compared with oxycodone 4 times daily (Webster et al., 2006).
In a randomized, open-label trial, 392 patients with chronic low back pain were randomized to an effective, generally well-tolerated dose of once-daily morphine extended release or twice-daily oxycodone controlled release (Rauck et al., 2006a). Patients then entered an 8-week evaluation phase comparing the 2 treatments. As in the previous trials, use of a flexible titration schedule allowed 67.9% of patients (morphine extended release, 65.0%; oxycodone controlled release, 70.9%) to be titrated successful. Once-daily morphine extended-release capsules provided significantly better analgesia with a lower morphine-equivalent daily dose compared with oxycodone twice daily (Rauck et al., 2006a). Sleep quality was significantly better with once-daily morphine, and rescue ibuprofen use was significantly less compared with twice-daily oxycodone. During a 4-month extension phase, patients receiving once-daily morphine continued to experience lower pain intensity and greater improvements in sleep quality (Rauck et al., 2006b).
A second once-daily opioid formulation, hydromorphone OROS, was also associated with statistically significant improvements in pain, quality of life measures, and sleep quality during a 6-week open-label trial (Wallace et al., 2007) and 6-month open-label follow-up (Wallace & Thipphawong, 2010). A clinical trial of hydromorphone OROS demonstrated that administration with ethanol does not result in premature release of hydromorphone from the capsule, a phenomenon referred to as "dose-dumping" (Sathyan et al., 2008). This is important because a previous formulation of hydromorphone extended release exhibited dose-dumping, resulting in its withdrawal from the US market (FDA asks Purdue Pharma to withdraw Palladone, 2005).
The efficacy and tolerability of fentanyl transdermal patch were demonstrated in open-label trials lasting 1 month (Simpson et al., 1997), 3 months (Lee et al., 2011), and 13 months (Allan et al., 2005), respectively. In the 13-month study (Allan et al., 2005), 673 patients with moderate to severe chronic low back pain ingested morphine sustained release once every 12 hours or applied a single fentanyl patch every 72 hours. Both treatments provided similar, significant analgesia throughout the study. Each opioid formulation was associated with typical opioid-associated adverse events such as constipation, nausea, vomiting, dizziness, diarrhea, and somnolence. Constipation was significantly less frequent with the fentanyl patch (52%) than with morphine sustained release (65%). The 3-month trial enrolled 1576 patients with severe, treatment-refractory chronic low back pain. Fentanyl provided significant improvements in measures of pain, function, and sleep quality. It should be noted, however, that <1% of patients enrolled in this trial had received a strong opioid as their prior medication; hence, the results support the use of transdermal fentanyl in patients with severe refractory pain but do not distinguish fentanyl from other opioids. Of note, the use of fentanyl transdermal patch for opioid-nai’ve patients in this study is contrary to the products’ approved use.(Duragesic®, 2009) In contrast, studies of oxymorphone (Hale et al., 2007), oxycodone controlled release, and morphine extended release (Rauck et al., 2006a), and hydromorphone OROS (Wallace et al., 2007) have demonstrated efficacy in patients with chronic low back pain that was moderate to severe despite previous opioid therapy. Fentanyl may offer advantages over oral opioids for patients with compliance issues, including those with problems chewing or swallowing. Fentanyl clearance appears to be affected very little by renal failure (Dean, 2004) or hepatic cirrhosis (Haberer et al., 1982) impairment, making it an option that may be used with caution in these populations. However, fentanyl patch should not be administered to opioid-nai’ve patients because there is a significant risk of respiratory depression even with the lowest available fentanyl patch dose, 25 ^g/hour (Duragesic®, 2009).
In a randomized, double-blind crossover study, 79 patients with chronic low back pain applied a 7-day buprenorphine transdermal patch or placebo patch for 4 weeks of treatment before crossing over to the alternate therapy for an additional 4 weeks (Gordon et al., 2010).
Patients began treatment with the 5 ^g/h patch and were titrated with each successive application to the maximum tolerated dose (5, 10, or 20 ^g/h). Fifty-four of 73 (74%) patients completed 4 weeks of treatment with buprenorphine patch. Mean reduction in pain (100mm Visual Analog Scale) was 39.5% in the buprenorphine group and 29.8% in the placebo group, which was statistically significant. Improvements in Pain Disability Index scores were significant compared with baseline but not compared with placebo.
