Answers are grounded in this paper and cite their sources.
In recent years, opioid consumption for pain management has increased drastically in the United States and other developed countries. Of the more than 25 million US adults affected by chronic pain, 2 million are diagnosed with an opioid use disorder. Thus, the opioid crisis is now recognized as one of the most pressing public health challenges of our time [1]. Many patients use opioids for chronic nonmalignant pain, though for some such use may be unnecessary, and for many it may not be advisable clinical practice due to side effects and the potential for opioid abuse [2]. Weaning patients off of opioids without compromising adequate pain management can be challenging. Regulatory bodies and professional organizations increasingly encourage or even require use of nonpharmacologic interventions to reduce reliance on opioids [3,4]. However, few studies have evaluated how to optimally incorporate nonpharmacologic interventions into clinical practice to achieve such a goal. Acupuncture is a nonpharmacologic pain management therapy originating from Traditional Chinese Medicine (TCM). It involves insertion of filiform needles into certain points on the body, followed by manual manipulation, heat, or electrical stimulation to elicit therapeutic effects [5]. Neuroscience research has shown that acupuncture increases production of endogenous analgesic neurotransmitters, such as endorphins and adenosine [6–8]. The effect of acupuncture for chronic nonmalignant pain is well established. In a meta-analysis of 29 randomized controlled trials (RCTs; N = 17 922), acupuncture was found to reduce chronic pain, such as back and neck pain, osteoarthritis, chronic headache, and shoulder pain, more than usual care or—to a lesser degree—sham acupuncture [9]. Whether acupuncture can be used to treat chronic pain so that patients are able to reduce their use of opioids is unknown. In this issue, Zheng et al. [10] report data from an acupuncture study designed to address this gap. They conducted a multicentered, randomized, three-arm trial of 108 patients comparing electro-acupuncture (EA), sham electro-acupuncture (SEA), and pain medication management education only (PMM) on chronic pain and opioid use. At the end of the treatment phase, they found that EA led to a 20.5% (P < 0.05) reduction in opioid dosage (primary outcome measure), SEA led to a 13.7% (P < 0.01) reduction, and PMM led to a 4.5% reduction (not significant). However, no significant between-group differences were found. Pain intensity for all three groups did not change over time. No between-group differences in dosage of nonopioid analgesics, pain intensity, function, or opioid-related adverse events were found. Interestingly, the PMM group had a 47% reduction in opioid dosage after crossing over to a course of electro-acupuncture treatment. The strength of this study is its multicenter, randomized, three-arm, participant- and assessor-blinded design. However, several methodological issues reduce its ability to guide clinical practice. These issues also illustrate the difficulties in designing and conducting acupuncture studies. First and foremost is whether the acupuncture regimen used delivered maximal therapeutic effects. The investigators did not observe any significant reduction in pain scores, as seen in other studies of acupuncture treatment for chronic musculoskeletal pain. This raises the possibility of suboptimal “dosing” of acupuncture. The regimen used was “twice a week for four weeks followed by once a week for two weeks then once every two weeks for four weeks.” Mechanistically speaking, if endogenous opioid production is thought to be responsible for acupuncture’s analgesic effect, how long does that effect last after a treatment? Is twice a week frequent enough? Perhaps a regimen that provides constant stimulation, for example, using indwelling auricular needles, would provide better effects in this setting. Second, the trial protocol did not specifically ask the therapists to address the type of chronic pain the participants were experiencing. For example, for participants with chronic neck pain, no local neck points were used, as is typical in TCM practice. These issues may have contributed to the lack of pain reduction and the hesitation from trial participants to reduce their opioid dosage. Another issue is the way opioids were tapered in the study; participants were asked to reduce their opioid dosage “by 30% in week 8, 50% by week 11, and 75% to 100% by week 14 as long as their pain did not get worse.” However, it is unclear how frequently the assessment and dose reduction were implemented. Were they daily? Weekly? If participants were not assessed frequently enough, acupuncture may have led them to have reductions in both pain and need for opioids without having their opioid dose reduced in time. If this was the case, the method for implementing opioid dose reduction was not sensitive enough to detect the primary outcome. Accrual is often difficult in acupuncture and other chronic pain studies. In this study, the investigators had to reduce their sample size from the initial size of 316 participants to 120. Ultimately, they enrolled 108 participants. They also had to change the randomization ratio from 1:1:1 to 2:1:1 to improve accrual. However, the power calculation was based on 1:1:1 randomization to detect an effect size estimate of 0.48 between EA and SEA. Changing to a ratio of 2:1:1 and pooling data from the two ratios reduced the power of the study. A larger difference between EA and SEA would have to exist in order to be detected by the sample. As opioid consumption is not typically a continuous outcome with normal distribution (some outliers of high opioid consumption can skew the data), a small sample size can fail to detect an effect size that may ultimately be clinically relevant. An interesting study finding is that when PMM participants finally received EA, there was a further 47% reduction in opioid dosage. This reduction is significantly greater than the 20.5% reduction in participants who received EA in the first place. Did the anticipation of finally getting acupuncture treatment enhance the effect of EA? Did the acupuncturists provide more effective treatments by having more interaction with the participants once they did not have to worry about masking group assignment? We can only speculate. But this finding opens the door for further investigation. Despite these issues, this is an important study because opioid overuse is a major clinical problem and effective opioid-sparing pain management strategies are very much needed. As we face a tremendous public health crisis in overdependence on opioids for pain management, there is an urgent need for rigorous and practice-changing acupuncture research. The best way to avoid chronic opioid dependence is to use nonpharmacological interventions such as acupuncture that may effectively reduce the need to consider prescribing opioids in the first place. Research ought to examine the integration of acupuncture into the treatment protocols for acute or subacute pain to help reduce the use of opioids initially, which may lead to long-term opioid use. As opioid dependence and chronic pain management are highly complex, trials using acupuncture or pragmatic comparative effectiveness designs may be more productive and timely in guiding public health practice rather than the more narrow testing of the specific efficacy of acupuncture needling. Further, it is important to collaborate with neuroscientists to better use appropriate tools/biomarkers to phenotype various chronic pain conditions and determine if targeted acupuncture stimulation (i.e., electric, manual, or auricular) is the most appropriate method of opioid-sparing personalized pain management for specific types of pain. The opioid crisis in the United States and other countries creates an opportunity for innovative and clinically meaningful acupuncture research that answers the right questions, employs the right methods, and gets to the right points. Funding sources: The authors received support from the National Institutes of Health/National Cancer Institute Cancer Center Support Grant P30 CA008748.
No takes yet. Share an insight, caveat, or question.
Deng et al. (2018) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: