Whether you consider opioids to be the gold standard for treating pain or a last resort, they do the job immensely well. That hasn’t changed.
What has changed is that for the past 15 years and counting, prescription opioids have been increasingly associated in the political and public consciousness with addiction and overdose deaths—fairly to some degree, considering their involvement in the “first wave” of the overdose crisis, but excessively and harmfully given their relatively small role in today’s overdose landscape. With anti-opioid rhetoric and crackdowns, the federal government has conspicuously failed to solve the addiction-pain conundrum.
Science is tasked with providing the evidence to inform a better balance. And given the political urgency, it doesn’t lack resources.
“We have $1 billion for two years for opioid use disorder and pain,” said Linda L. Porter, PhD, director of the Office of Pain Policy at the National Institute of Neurological Disorders and Stroke (NINDS), referring to the Helping to End Addiction Longterm (HEAL) grants launched by the National Institutes of Health almost two years ago.
The pain side of HEAL includes “a huge breadth of different types of research,” Porter told Filter. “This includes basic science to better understand pain. It includes programs to develop new therapies, including drugs, biologics, devices, and behavioral approaches to help manage pain.” The purpose of HEAL’s pain research in general is to “find out what works for what kinds of pain in what people,” she said.
Porter, a physical therapist by training, knows that opioids are far from the only resource to treat pain. The HEAL grants fund several levels of clinical trial networks allowing NIH to test anti-pain devices, pharmaceuticals and other treatments in humans, and a network to do later-phase trials on comparative effectiveness. “For example, you could take a drug like duloxetine [an SSNRI used to treat depression] and compare it to physical therapy.”
The quest is not so much for measuring pain to find out whether patients feel it. Rather, it is about being able to predict who will feel better with what kind of treatment.
Funding for HEAL was initially for two years, but there isn’t an end-point, said Porter. “We have committed HEAL-related dollars to five years’ worth of studies at this point,” she said. Congress gave HEAL $500 million for 2018 and $500 million for 2019, to be shared between addiction and pain research. “So each year, until Congress takes it away, which we don’t expect that they ever will, we will have this. If they do take it away, it would make them look worse.”
So what is some of this money being poured into?
One HEAL program, for example, focuses on low back pain, with the aim of getting better diagnostics and treatments, said Porter. The burden of lower back pain has been increasing, resulting in poor health and mortality behind only ischemic heart disease and chronic obstructive pulmonary disease, according to the National Institute of Neurological Disorders and Stroke. About 20 percent of people with acute back pain (lasting a few days to a few weeks) go on to develop chronic back pain (defined as lasting at least 12 weeks) for over a year.
Another subject is the phenomenon of opioid-induced hyperalgesia, whereby some patients on long-term opioids experience more, not less, pain as a result. The strongest evidence for this taking place is in long-term opioid patients with migraines who go to the emergency room with an acute attack. The opioids don’t work, so the patient gets more, said Porter.
Pain is notoriously hard to measure objectively. But the quest is not so much for measuring pain to find out whether patients feel it—most sensible researchers agree that if people say they feel pain, they do.
Rather, it is about being able to predict who will feel better with what kind of treatment. And if the pain is chronic, that treatment will most likely not be opioids.
The Pain Researcher
Sean Mackey, MD, PhD, Redlich professor and chief of the Stanford Division of Pain Medicine and professor of Anesthesiology, Perioperative and Pain Medicine at the Stanford University Medical Center, is nothing if not a pain expert.
He conducts much NIH-funded research on pain and addiction, but mainly on pain. Yet he says that he likes to steer clear of opioids when talking to reporters “because when you talk about opioids, it sucks all the oxygen out of the room.”
He told Filter that pain is unequivocally subjective, defined as an “unpleasant emotional or sensory experience.” That doesn’t mean it isn’t real, he stressed. “Think of other experiences—happiness, sadness, love, anger, fear—they’re all subjective.”
So he would never question a patient’s claim to be in pain. “If you tell me you’re happy, I’m not going to question it. If you say you’re in love, I’m not going to question it.”
Pain does differ from emotions, however, because it involves the body, a tactile sensation, said Mackey. “That’s what makes pain unique, and confusing.”
“If you say your pain is a 7, I’m not going to question it. But it will be different from someone else’s 7.”
Traditionally, pain is measured with a pain scale—how bad it is from 1 to 10, as reported by the patient. “If you say your pain is a 7,” said Mackey, “I’m not going to question it. But it will be different from someone else’s 7. We have learned that for a given stimulation, one person may rate their pain different than someone else. This is a highly personal experience.”
Pain scales therefore present many challenges, said Mackey. “So we’re trying to capture a broader experience across physical, psychological and social domains.”
This field of study is aimed at making pain’s relationships with mood and physical functioning part of our composite understanding of what pain is.
Pain biomarkers—characteristics of patients with specific pain conditions—are a key area that HEAL is investigating, but one that is so far hypothetical.
“What we think is that there will be unique biomarkers for sickle cell pain, for example, or for neuropathy,” Porter said. So HEAL is funding studies looking at the characteristics of patients with different kinds of pain conditions. These characteristics may be psychosocial as well as genetic, however.
Biomarkers, Porter stressed, aren’t viewed by NIH as a way to measure pain. “When we say biomarker, we mean a lot of things: It can be a gene of a physiological attribute, or a psychological trait associated with a certain pain condition like fibromyalgia, or it could be a pharmacological marker,” she said.
The biomarkers might be predictive, she said. For example, “Some people have chronic pain after knee surgery, some don’t,” and understanding why would clearly be valuable.
“The field has to develop composite biomarkers of pain. There isn’t one sole biomarker.”
Investigations into certain single potential biomarkers have been tried, however, said Mackey, who has submitted a grant proposal to further investigate multiple biomarkers. “But EMG [measurements of skeletal muscles’ electrical activity], pupil size, heart rate—all failed miserably at this. The field has to develop composite biomarkers of pain. There isn’t one sole biomarker.”
Biomarkers can nevertheless be useful, said Porter—not in determining how much pain someone has, but in predicting how they will respond to a certain treatment, such as a medication.
Beyond physiological indications, a psychological characteristic like “catastrophizing”—in which patients think too much about their pain, ruminate about it, lie in bed focusing on it—may turn out to correlate with a specific pain condition, said Porter. “And if you add other characteristics to the catastrophizing, this can make up a formula, and that formula can tell you a lot about a person and their likelihood to respond to different treatments.”
Nociception is the nervous system’s response to a harmful or potentially harmful stimulus, such as heat or mechanical damage to the body. Pain, based on a perception, is different.
One of the earlier HEAL programs—dubbed the “pain-o-meter”—involved looking for a way to measure pain, or to quantify what a patient means by saying something hurts.
The call for researchers to apply for pain-o-meter funding went out two years ago, and was for a “very specific and limited solicitation looking for a device, or a combination of means, to measure different characteristics that could be correlated to pain,” said Porter. “The solicitation was very narrow.”
NIH funded a total of 10 awards, mostly to small companies looking to do this. “Some focused on acute pain—experimentally induced acute pain.”
Some of the pain-o-meter investigators have looked at changes in facial expression.
The measures being looked at to measure induced pain are usually a combination: for example, pupillary response combined with brain changes seen through EEG measures (detecting electrical activity in the brain) combined with heart rate.
“But there’s nothing that solidly tells us” whether pain exists, said Porter. “An MRI will show you that there are changes, that activity changes occur when you induce pain—we know a lot about that, we know MRIs will show us that. We also know that MRIs will show us long-term changes in structures of the brain if someone has pain for a long time.”
Of course, better understanding underlying chronic pain is of central interest to NIH, and the trans-NIH initiative goes way beyond the pain-o-meter.
Some of the pain-o-meter investigators have looked at changes in facial expression, added Porter. “Maybe in the future, a doctor in a primary care setting can use a combination of things that are easy to do in the office, such as an EEG linked up to a change in facial expression,” she said. “We’re not there yet.”
The whole point of this, of course, and the reason such resources have been devoted to it, is to help physicians only prescribe opioid medication when real pain exists. But there are heavy ethical implications to these judgment calls. And how “real” pain is depends not on the injury or condition, but on the brain.
Until around 2011, there was no way of getting a “physiological readout” of pain, said Mackey. But then, Mackey and colleagues were able to use brain imaging to develop a physiological signature for pain. “We can use brain imaging and machine learning to decode brain patterns, to see if a person is in pain,” he said.
Their technique has now been extended from acute to chronic pain. However, this is just in the laboratory—it’s not a clinical tool. At least, not yet. And Mackey notes that those brain imaging changes would be seen with other emotions and experiences too—not just pain.
“NIH is interested in developing this objective pain-o-meter,” Mackey concluded. “But it’s a bit nuanced.” And he dismisses the idea of using such a device to tell whether someone is in pain or not. “The ability to objectively determine whether someone is experiencing pain is very boring, because I can simply ask them,” he said. “I don’t need a machine to tell me they’re in pain. So it’s just not a very interesting question.”
Rather, Mackey is interested in encapsulating more of the biopsychosocial experience of pain, recognizing that it impacts people differently.
“You may see a wide variability in the amount of disability somebody has to a given degree of impairment,” he said. For example, a star athlete or a CEO running a Fortune 500 company might be highly motivated to stay functional. “They have jobs, they love what they’re doing.” Someone with a low-paid, uninspiring job or no job may feel pain more. “This doesn’t lend itself to something nice and simple.”
Here’s what Mackey finds really exciting: using an MRI brain signature to determine whether a patient will respond well to a particular treatment. This, if it turns out well, is precision medicine.
There has been a major push for precision medicine at NIMH following the introduction more than 10 years ago of the Research Domain Criteria (RDoC)—a new way of looking at treating mental illness.
Psychiatric medications so obviously don’t work for everyone, and psychiatrists are often in the position of trying drug and then another, until one works. The RDoC situation is analogous to pain because in psychiatry, as in pain, medicine is working with subjective experience, said Mackey.
“The head and heart of NIH is in the right place here. They’re not trying to call out people who are faking it or lying.”
“It’s not a matter of trying to figure out if this person is in pain right now,” he explained. “If you argue about that, all you do is get into this quagmire of arguing whether you’re trying to take away autonomy.”
So about the placebo factor, if a patient thinks something will work to fix a condition which is subjective? “Placebo is huge,” he conceded. “But it just doesn’t explain this in its entirety.”
So the big interest in research is not in diagnosing pain or its causes, but in being able to predict a treatment response. “The head and heart of NIH is in the right place here,” said Mackey. “They’re not trying to call out people who are faking it or lying.”
George F. Koob, PhD, is the director of the National Institute on Alcohol Abuse and Alcoholism (NIAAA), and also a basic researcher at the National Institute on Drug Abuse (NIDA). One of NIDA’s mandates is treatment for pain, and Koob’s NIDA lab is using rodents in pain testing.
The researchers there use Von Frey hair tests to measure pain. In this test, the rodent’s paw is poked with a “hair” (a nylon tube) to test its threshold for lifting its paw away. “There are different measures of force and sharpness of the hairs, so we can get a quantitative measure of the latency to remove their paws,” Koob told Filter.
Opioids, when given acutely, increase the latency—the rodents don’t feel the pain as much and are slower to life their paws away. But when opioids have been given chronically to the rodents, they lift their paws more quickly than if they have not been on any opioids, demonstrating hyperalgesia. “The Von Frey test doesn’t hurt the animals, and gives a good baseline for hyperalgesia,” Koob said.
Despite opioids’ deficiences for chronic pain, acute pain demands them, Koob said.
Why are researchers so worried about hyperalgesia? Because long-term opioids given to humans for chronic pain may actually increase, not decrease, their sensitivity to pain. “This is why we talk about opioids causing pain,” Koob said, adding that an equivalent phenomenon is seen with alcohol. “It’s because the chronic administration of the drug produces a process in the brain that results in the pain coming back even more.”
Despite opioids’ deficiences for chronic pain, acute pain demands them, Koob said. “If you break a bone, you are going to beg for opioids, and you should get them, because otherwise you can go into shock.”
Still, there are limitations in using animal models for human chronic pain because of the cognitive component in humans, he said. “Animal models for chronic pain are usually something that causes an inflammatory reaction, injected into a muscle. Industry has used those models for developing opioids to treat pain in general.” But what’s usually used in human pain research is self-reports.
Sensitivity to pain can be as simple as how you respond to the blood pressure cuff being inflated on your arm. Using self-reports, researchers have studied the “pressure pain” of the cuff, in both opioid- and alcohol-dependent subjects. For alcohol-dependent subjects, hyperalgesia was documented based on thermal or electrical stimuli.
Another measure is the cold-water test, involving dipping the hand in cold water. “If you’re under the influence of opioids, you can hold your hand in the water longer,” said Koob. “But if hyperalgesia is present (due to long-term use of opioids), you can hold your hand in the water for less time.”
For humans, there are also “emotionality tests,” said Koob. “You can measure aspects of emotional pain, and we know that both opioids and alcohol can dull emotional pain.” He added that physical pain can cause emotional pain.
Today, there’s a reluctance for chronic pain to be treated with opioids,” Koob noted. And he agrees that science’s hunt for pain measures is not connected to people faking their pain. “People don’t lie” about pain, he said. “If they say they’re in pain, they’re in pain.”
But the real challenge is this: “How do you sort out emotional pain—such as depression or loss of a loved one—from physical pain?” There is “crosstalk in the amygdala,” said Koob. “That’s the brain region that I’m in love with, and studying. There’s stress circuitry there. You can evoke pain relief in the amygdala with one circuit, and make pain worse in another.”
The brain, not the affected body part, is the key to pain.
He went on to explain: “The splenic pain pathway goes into the dorsal root ganglia of your spinal cord, thalamus, and cortex, and eventually intersects with the frontal cortex and the amygdala. We have known this historically. But there’s another pain pathway that goes directly into the amygdala through the parabracheal pathway, and is much faster.”
Even if you aren’t an anatomy specialist, you get the idea: The brain, not the affected body part, is the key to pain. And there is a great deal that we just don’t know about the brain.
All of which illustrates how, even with billions in funding, pain research will be hard pushed to unravel the complexity of this phenomenon. And while this doesn’t come down to the researchers, it is incumbent on policymakers and clinicians to exercise extreme caution in using research to justify any further barriers to pain patients receiving opioids.