ARTICLE IN BRIEF
Long-term use of opioids led to no functional improvement, but a higher risk of adverse events in patients with polyneuropathy in a large population. Experts agree the findings mean prescribing doctors should proceed with caution in prescribing opioids.
Patients with polyneuropathy were commonly prescribed long-term opioid therapy for 90 or more days, yet it led to no improvement or a slight worsening in functional status and a higher rate of adverse events, including opioid dependence and overdose, compared to controls, according to a population-based study published online on May 22 in JAMA Neurology.
Polyneuropathy, which encompasses nerve damage and pain caused by diabetes, various infections, nutritional deficiencies, and cancer is one of the most commonly treated conditions in both general and specialty clinics, particularly among the elderly. While some evidence from clinical trials supports the short-term efficacy of opioids for pain, little evidence exists of long-term safety and effectiveness. Moreover, opioids have been associated with a growing, nationwide pandemic of addiction, dependency, or overdose.
The current findings, which confirm both that long-term opioid use is prevalent among patients with polyneuropathy and that it carries a risk of these adverse effects, should be taken into consideration by all prescribing physicians, the study authors concluded.
“The burden of opioid use we observed was very large,” said Christopher J. Klein, MD, professor of neurology at the Mayo Clinic in Rochester, MN, and senior author of the study, said in an interview with Neurology Today. “In this large population, almost one-fifth of people were on opioids. And yet they did not have an improved functional status. One of the goals of chronic analgesic therapy is to improve patients' quality of daily life, and we did not see that here.”
For their study, researchers at the Mayo Clinic assessed data from the long-running Rochester Epidemiology Project (REP) database to identify all people living in Olmsted County, MN, between January 1, 2006 and December 31, 2010. They identified a total of 2,892 patients with polyneuropathy and 14,435 controls. They used administrative codes contained in patients' medical records to identify cases of polyneuropathy and all adverse events during the four-year study period.
The REP includes a comprehensive prescription database, which the researchers used to identify opioid prescriptions, including the indication, date prescribed, medication, strength, amount dispensed, number of refills, and the intended duration. However, information on the morphine-equivalent dose was not available.
They found that patients with polyneuropathy were more likely to receive long-term opioid therapy, defined as one or more consecutive opioid prescriptions resulting in at least 90 days of continuous use, compared to controls (18.8 percent vs. 5.4 percent, respectively).
In addition, patients with polyneuropathy who received long-term opioid therapy were more likely to report having pain at the end of the study compared to patients with polyneuropathy who did not receive long-term opioid therapy (adjusted odds ratio [OR], 2.5; 95% CI, 1.9-3.4). Patients who received long-term opioid therapy did not show improvement on any functional status markers and worsened on several. These included an increased likelihood to require an assistive device (adjusted OR, 1.9; 95% CI, 1.4-2.6); to have trouble climbing stairs (adjusted OR, 1.7; 95% CI, 1.2-2.4); and to be unable to work (adjusted OR, 1.3; 95% CI, 0.8-2.0).
In addition, patients with polyneuropathy who used opioids long-term were more likely to experience an adverse outcome of any kind, even after the researchers adjusted for sex, medical comorbidities, and the use of non-opioid analgesics. These included depression (adjusted hazard ratio [HR], 1.53; 95% CI, 1.29-1.82); opioid overdose (adjusted HR, 5.12; 95% CI, 1.63-19.62); opioid dependence (adjusted HR, 2.85; 95% CI, 1.54-5.47); and other chemical dependence (adjusted HR, 1.73; 95% CI, 1.21-2.49). However, the association with mortality was nonsignificant.
“My hope is that clinicians would use this information in the office setting when having a conversation with a patient,” said E. Matthew Hoffman, DO, PhD, a neurologist at the Mayo Clinic and lead author of the study, in a telephone interview with Neurology Today. “For patients not yet on opioids, clinicians should be candid about expectations. And for those on opioids who show no benefit, they should offer further support to help get them off the opioid.”
“This has important implications for policy, too,” Dr. Klein added. “I think we want to continue the conversation to get doctors to change their [prescribing] behavior.”
The researchers noted several limitations to their study. Among them were its retrospective design, their use of administrative codes to ascertain cases of polyneuropathy and adverse outcomes, and lack of confirmation that prescriptions were filled and taken as intended.
In an accompanying editorial, Nora D. Volkow, MD, director of the National Institute on Drug Abuse at the National Institutes of Health, and Walter J. Koroshetz, MD, FAAN, director of the National Institute of Neurological Disorders and Stroke, said the results were “not surprising” given opioids' well-established potential to develop tolerance, dependence, and addiction. They urged the development of new, non-opioid therapies for chronic pain, and in the meantime, they concluded, “structural changes in the health care system, including training of physicians in the screening and management of pain, as well as coverage by insurance of comprehensive pain management programs” are merited.
“Based on these findings, I would recommend to the Academy that they re-examine the guideline that was published in 2011 on the treatment of diabetic neuropathy,” said Gary M. Franklin, MD, FAAN, research professor in the department of environmental and occupational health sciences at the University of Washington in Seattle. “Anything you'd say for diabetic pain you'd say for other neuropathic pain. I think this study is strong enough that those conclusions [that opioids may be used for neuropathic pain] should be revisited, especially because of the balance of poor functional outcomes and potential severe dependence, addiction, or overdose. If function's not better, why put the patient at risk?”
While several experts agreed that the trial's retrospective design and its reliance on administrative claims were limitations, they added that its size was a major strength.
“Having a large trial is very valuable and strengthens the likelihood that the findings are not due to chance alone,” said Roy Freeman, MD, professor of neurology at Harvard University and director of the Center for Autonomic and Peripheral Nerve Disorders at Beth Israel Deaconess Medical Center. “However, it is a retrospective study, and my concern is that these were patients who had worse disease and therefore were predisposed to a bad outcome no matter what. An ideal study would be prospective, where the patients were equally matched. One could see if opioids in fact did increase the likelihood of worsening function and dependency and overdose.”
Dr. Franklin agreed. “This is a great study, but a prospective study would be even stronger. You can do as much looking at randomized, controlled trials as you want, which is the gold standard for efficacy, but you have to look at large populations to really learn about harms from treatments. Because oftentimes harms are severe, but they're relatively rare. If you only have 150 people in a randomized, controlled trial, you may not find a severe adverse event like severe dependence. Plus, it probably takes longer. Most trials of opioids have lasted less than a few months.”
An important question remains unanswered, Dr. Freeman noted. “This study reinforces a notion we already know: that for the vast majority of patients, opioids are unnecessary and result in poor outcomes. But which patients really do need, respond to, and have good outcomes with opioid therapy? That requires a randomized, prospective study with pre-specified hypotheses, but the patients must have the same degree of pain and the same degree of function at the outset,” which may be difficult, he noted.
“Looking at the types of diseases for which these medications were used, for the majority there were quite effective alternatives,” said Rodica Pop-Busui, MD, PhD, professor of internal medicine, metabolism, endocrinology and diabetes at the University of Michigan. She pointed out that “the majority were not provided by pain specialists, who have a much higher understanding of the various interactions with medications and the potential for addiction.” Prescribing opioids should only be done by very experienced physicians who understand their mechanisms and interactions, and physicians who do not have this level of expertise should provide referrals to those who do, she noted.
Dr. Pop-Busui is the co-chair of the American Diabetes Association panel that published the recent updated evidence-based guidelines on the management of diabetic neuropathic pain earlier this year. [Neurology Today reported on the guideline, which included input from neurologists, in the March 16 issue.] “This paper is somewhat in line with the level of evidence we found when we published our position statement, where we tried to look very critically at level of evidence and pain agents,” Dr. Pop-Busui said. “We found that in fact opioids, considering their potential for harm and addiction, are not justified considering we have alternative agents and therapies. I'm very proud that in the diabetes neuropathy field, we took a proactive step against using opioids, because of their high risk for addiction and harm, and the evidence for effective alternatives.”
For now, Dr. Pop-Busui said, prescribing doctors should proceed cautiously and with an eye toward each patient's unique needs. “I think each disease state has to be very carefully analyzed and the evidence involving efficacy in pain reduction has to be balanced very carefully with the potential for addictions and very serious side effects. Don't apply templates to patients. Understand the particularities of their situation and in which situations using opioids – for a limited amount of time, at the lowest possible dose – is justified.”
To enact change will require cooperation among prescribing doctors across the country, Dr. Pop-Busui concluded. “The entire health care community has to take a very strong stand to try to limit the use of opioids and undo potential harm.”
EXPERTS: ON USE OF OPIOIDS FOR POLYNEUROPATHY