Should We Prescribe Opioids for Chronic Pain?

Charles E. Argoff, MD


June 18, 2015

This feature requires the newest version of Flash. You can download it here.

This is Dr Charles Argoff, professor of neurology at Albany Medical College and director of the Comprehensive Pain Center at Albany Medical Center in Albany, New York. Today, I want to discuss whether we, as providers and prescribers, should be prescribing opioids for chronic pain. If so, to whom and how? Let us think about that and look at some very important facts.

The Institute of Medicine reports[1] that approximately 100 million adult Americans experience chronic pain. Many of us see patients in pain day in and day out. In fact, regardless of your specialty but especially in primary care, pain management, and neurology, we see people with chronic pain on a regular basis.

The next consideration is our training. As we learn to become practicing physicians and other providers, are we taught about the importance of managing a person's pain? Yes, we are. Although the training is not complete enough, we certainly are exposed to the need to address palliative issues, that pain relief is very important, and that opioids are one of the treatments that are available for pain relief.

To become certified in the subspecialty of pain management, there is an Accreditation Council for Graduate Medical Education (ACGME) process. This certification is as fully developed as that of other subspecialties. To become eligible to be ACGME-certified in pain management, candidates are required to complete an accredited fellowship and pass a written examination.

What about in practice? We see a variety of individuals with chronic pain and different chronic pain types: pain associated with chronic musculoskeletal conditions, such as osteoarthritis and osteoporosis; and neuropathic pain, such as diabetic neuropathy, postherpetic neuralgia, postlaminectomy pain, and small fiber neuropathy. We see patients with a number of different chronic pain conditions, including conditions that do not neatly fit into neuropathic, musculoskeletal, or non-neuropathic inflammatory categories—such as fibromyalgia, and so on. We do see many people who experience chronic pain associated with other medical conditions.

Neuropathic Pain Management

The management of pain in all of our patients is extremely important part of what we do. Today, however, I want to focus on neuropathic pain as an example. In subsequent blogs, I may talk about other specific areas where opioids may be prescribed, but let us look at neuropathic pain for the remainder of this blog.

The Scope of the Problem

The neuropathic pain states that may or may not involve the use of opioids include diabetic neuropathy, postherpetic neuralgia, and others. We know that chronic pain affects people who have diabetic neuropathy and postherpetic neuralgia, but what about other neurologic conditions that may be associated with pain?

A study[2] of patients with Guillain-Barré syndrome showed that 38% of 156 patients who were evaluated continued to experience moderate to severe chronic pain after 1 year. A survey[3] of nearly 8000 patients with multiple sclerosis found that 49% complained of severe pain and 51% had severe pain at more than four sites.

Chronic pain is one of the most common nonmotor symptoms in patients with Parkinson disease, and is independent of motor symptoms.[4] Symptoms of chronic pain often precede manifestation of the motor disturbances.

Thus, chronic pain is associated with a number of neuropathic states. When we try to define the most appropriate treatment regimen for each person in pain, we can think in terms of buckets, or categories, of treatment. These categories include pharmacotherapy; interventional, invasive approaches; behavioral strategies; psychological support, lifestyle changes; complementary and alternative approaches; and physical medicine and rehabilitative approaches. We should be thinking of multimodal strategies that may include opioids as one appropriate consideration.

Where Do Opioids Fit Into the Pain Management Paradigm?

Let us look at opioid strategies and the numbers needed to treat (NNTs). How many people do I need to treat for one person to achieve 50% relief from a particular agent? Opioid analgesics are superior to other analgesics in having a lower NNT.[5] They are superior to gabapentin; pregabalin; capsaicin; selective serotonin reuptake inhibitors (SSRIs); and serotonin norepinephrine reuptake inhibitors (SNRIs), such as duloxetine, all of which we commonly prescribe for chronic pain.

Opioids are not superior in general to tricyclic antidepressants, and tricyclic antidepressants are superior to many of the more commonly prescribed drugs.[5] But like opioids, tricyclic antidepressants have harm-related issues.

Risk Reduction and Effectiveness Monitoring

As we think about prescribing opioids to someone with chronic pain, we need to play a very active role in reducing the risks of harm. We need to reduce the risks of any treatment we prescribe, whether it is surgery or other medical approaches.

When opioids are being considered as part of a chronic pain treatment plan, we need to have an established diagnosis; we need to perform an appropriate history and physical examination; we need to evaluate the results of relevant diagnostic tests; and we need to complete a risk assessment. Is the person at low, medium, or high risk for abuse and misuse? Do I have the ability to take care of someone who is at high risk in my practice?

We see different types of patients, and we need to be able to manage and monitor them, appropriately. And we need to remember that no one is at no risk. Validated tools, including the Opioid Risk Tool (ORT) and the Screener and Opioid Assessment for Patients With Pain (SOAPP), can be used to help with risk assessment.

We need to monitor patients on a regular basis. Just as my blood pressure treatment trial may not work in lowering blood pressure and the treatment may change, a trial of opioid therapy may not work because it is not best for a particular patient. A drug may not be best because of side effects or lack of benefit. We need to consider this as an initial trial with monitoring on an ongoing basis and to remember that we are prescribing opioids as part of multimodal therapy.

Evidence of Benefit and Risk

Abundant evidence has shown that opioid analgesics can be useful in patients with neuropathic pain. Studies[6,7] have shown, for example, that controlled-release oxycodone is beneficial for patients with painful diabetic neuropathy.[7] Gilron and colleagues[8] found that the combination of morphine and gabapentin allowed a person to use a lower dose of each, with more benefit from the combination and fewer side effects.

There is no shortage of evidence that these medications can be helpful, either alone or in combination. The neuropathic pain special interest group of the International Association for the Study of Pain,[9] the Canadian Pain Society,[10] and the European Federation of Neurological Societies[11] have recommended a place for chronic opioid therapy in the treatment of neuropathic pain.

There are serious risks as well. We know that opioid analgesic overdoses have become a public health epidemic, with opioid deaths between 1999 and 2010 increasing fourfold.[12] Fortunately, measures have been implemented and are beginning to reverse this trend.[12]

When we prescribe an opioid to someone with chronic pain or chronic neuropathic pain, we must have our eyes wide open and realize that we have to use these as safely as possible. As with any other class of medication, opioids have risks and benefits. Some of the risks include death, but they also include physical dependence, tolerance, sedation, confusion, constipation, and other side effects. We have to monitor our patients for those.

Should We Prescribe Opioids?

Should providers prescribe opioid therapy for certain individuals with chronic pain? "Should we prescribe" is not really the right question. What we need to ask ourselves is, how well are we prepared to prescribe opioids to our patients to provide the greatest benefits with minimal risks?

This is an important consideration. We must learn as providers, as we take care of people through our training and our experience, how to select patients for opioid therapy when indicated, and how to manage those patients on opioid therapy as safely and effectively as possible.

Final Thoughts to Address the Controversy

The tendency at this time is for people who are doing extremely well on opioids to be taken off of opioids for unclear reasons, other than provider concerns about the regulatory aspects of opioid prescribing. For a recent referral to me, the provider wrote that the individual who was being referred for chronic neck pain had undergone surgery of the spine in the past, and had received many other different treatments, injections, and medical therapies. This patient had been on a stable dose of opioids for many, many years, but this physician's practice had decided that it was no longer comfortable prescribing opioids for her and that was why the patient was being taken off opioids. In fact, in the referral note, the referring physician emphasized that "having said all this, this patient has done so remarkably well on opioid therapy for so many years." Here was someone who had been proven to be successfully treated with opioids, but was being taken off of them for no clear medical reason.

Critics will say, "There are no long-term data." In fact, a recent position statement[13] from the American Academy of Neurology suggested that there were no long-term data to support long-term opioid prescribing for selected people with chronic pain. That is a mischaracterization, and not what the data suggest. I can think of at least four published studies[14,15,16,17] in which the evidence for long-term use of opioids has been proved. These may be open-label studies, not randomized controlled studies, but I would argue that there is no area of modern medicine where a patient would be expected to be treated with a placebo for a year.

Analgesic studies typically ask a patient to be involved in a placebo-controlled trial for approximately 12 weeks. At the end of that 12-week period, the person may be able to enter an open-label study where they actually receive the medicine. To say that we have no data for long-term use is not the same thing as saying there are no randomized controlled studies for 1 year. That may be true for opioid therapy, but it is also true for many other therapies.

Barriers Are Not the Solution

The data we have more accurately reflect true practice. We do not put our patients on randomized controlled regimens. That needs to be emphasized, and not mischaracterized by people who are just so anti-opiate they cannot see the benefit for many people of this class of therapy. That is tragic unto itself.

Many positive policy changes, such as prescription drug monitoring programs in different states and other screening approaches for risk factors and behaviors for misuse and abuse, are making opioid therapy safer. We can do more. We know that the numbers of deaths due to unintentional overdose are beginning to decline. More can be done.

Misrepresenting opioid efficacy and erecting barriers to meaningful relief of chronic pain for our patients have not helped the situation. I hope that you will join me in recognizing that there is a role for chronic opioid therapy for certain selected patients with chronic pain, and that we will all work together in a constructive way to further use opioid analgesics in an appropriate setting as safely and effectively as possible.

I am Dr Charles Argoff, professor of neurology at Albany Medical College and director of the Albany Medical Center Comprehensive Pain Management Center. Thank you for your time.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: