Key points

Introduction

For people with chronic pain, the prospect of doing exercise may seem like an overwhelming and impossible task. And yet, exercise therapy is frequently prescribed for patients with a wide variety of chronic pain problems. Exercise provides multiple benefits for patients including improvements in strength, flexibility, and endurance; decrease in cardiovascular and metabolic syndrome risk; improved bone health; improved cognition and mood; and often most importantly for the patient, improved pain control ( Box 1 ). It therefore might seem that patients should be eager to participate in an exercise program. However, patients with chronic pain frequently present with significant levels of fear-avoidance behaviors and are often resistant to participating in an exercise program. Prescribing an appropriate exercise program for a person with chronic pain requires an understanding of the person’s biopsychosocial circumstances, an armamentarium of different exercise techniques, and good communication with the therapy team.

Introduction

For people with chronic pain, the prospect of doing exercise may seem like an overwhelming and impossible task. And yet, exercise therapy is frequently prescribed for patients with a wide variety of chronic pain problems. Exercise provides multiple benefits for patients including improvements in strength, flexibility, and endurance; decrease in cardiovascular and metabolic syndrome risk; improved bone health; improved cognition and mood; and often most importantly for the patient, improved pain control ( Box 1 ). It therefore might seem that patients should be eager to participate in an exercise program. However, patients with chronic pain frequently present with significant levels of fear-avoidance behaviors and are often resistant to participating in an exercise program. Prescribing an appropriate exercise program for a person with chronic pain requires an understanding of the person’s biopsychosocial circumstances, an armamentarium of different exercise techniques, and good communication with the therapy team.

Patient evaluation overview

Exercise has been found to be beneficial for patients with a wide variety of chronic pain diagnoses including arthritis, fibromyalgia, complex regional pain syndrome (CRPS), chronic neck pain, and chronic low back pain. No matter what the specific diagnosis, the patient evaluation must include an assessment of the biopsychosocial circumstances of the patient. A physical examination does not provide sufficient information. It is necessary to understand the patient’s psychological state and beliefs about pain, health, and wellness. It is important to consider the goals of the exercise program ( Box 2 ). Goals might be specific correction of impairments (eg, better range of motion), a reduction of disability (eg, walking without a cane), or improvement in participation (eg, return to work). Acute pain and chronic pain present different challenges. In the case of acute injury, assessment includes analysis of functional biomechanical deficits including muscular weakness, inflexibility, scar tissue, muscle strength imbalance, poor coordination, and decreased endurance. Treatment addresses these specific biomechanical deficits through a specifically tailored exercise program. Pain is often a useful guide for determining the intensity and frequency of exercise activities.

A patient with chronic pain may have all of the functional biomechanical deficits and in addition may be dealing with excessive deconditioning, fear-avoidance, depression, centrally mediated pain, loss of role in family and society, and entrenched disability. There has typically been a progressive loss of function and diminishing quality of life. Exercise intervention needs to address the biomechanical issues but must do so in the context of these other complicating conditions. Pain is typically a poor guide for determining the frequency and intensity of exercise. Education of the patient plays an important role in facilitating their participation in an exercise program. This education needs to be about the specific biomechanical issues, about the functioning of the nervous system, and about the psychosocial conditions that contribute to the person’s pain and disability.

How does exercise affect pain?

It is not always clear what it is about an exercise program that brings benefit to the patient. Exercise impacts the musculoskeletal system, the cardiovascular system, and the brain. The effects on the brain include impact on sensory processing, and improved motor coordination and cognitive and emotional functioning. For the chronic pain patient the impact on the brain may be the most important for improving the patient’s function and sense of well-being. A variety of research studies have demonstrated the impact of exercise on the brain. For example, exercise is an effective treatment of depression, which frequently accompanies chronic pain. Research has also shown that specifically targeted exercise programs can improve chronic pain and function, but changes in pain do not correlate well with improvement in physical measures, suggesting that it may be something other than the change in musculoskeletal function that is mediating the pain relief.

Pain perception is brought about by a complex interaction between peripheral nociceptive input and modulatory processes at the spinal and supraspinal level. The nervous system has an endogenous pain modulatory system that has inhibitory and facilitatory functions. Pain modulation occurs in cortical, hypothalamic, midbrain, and brainstem structures, and in the spinal cord. The peripheral nervous system also modulates pain perception. Multiple neurotransmitters are involved in pain perception and pain modulation. Studies have been done in animals and humans to try to elucidate the impact of exercise on pain perception and on this endogenous pain modulatory system.

Animal studies have addressed the effects of exercise in models of acute pain and chronic neuropathic pain. Stagg and colleagues evaluated the effect of repeated aerobic exercise on a neuropathic pain model in the rat. They demonstrated that sensory hypersensitivity in a sciatic nerve ligation model of neuropathic pain was reversed by regular moderate aerobic exercise (forced treadmill running). These effects were reversed by naloxone suggesting that endogenous opioids play a role in modulating pain perception in response to exercise. Chen and colleagues studied a rat model of acute postoperative mechanical hypersensitivity caused by skin/muscle incision and retraction. They looked at the effect of 5 day per week treadmill running for 4 weeks on behavioral markers of pain and on levels of substance P and cytokines in the dorsal root ganglion. They demonstrated that mechanical hypersensitivity was alleviated by the treadmill running and was associated with suppression of excess substance P and inflammatory cytokines in the dorsal root ganglion. Mazzardo-Martins and colleagues showed an antinociceptive effect of a swimming exercise in a chemical behavioral model of nociception in mice. This antinociceptive effect was blocked by naloxone and by an inhibitor of serotonin synthesis suggesting that this effect was mediated by the opioid and serotonergic systems. Martins and colleagues studied a mouse model of CRPS-1 and evaluated the effect of high-intensity swimming on mechanical allodynia. There was marked mechanical allodynia in this model of CRPS-1. High-intensity exercise significantly reduced the allodynia shortly after exercise (30 minutes) and at a delayed evaluation (24 hours). They further demonstrated the role of the opioid systems and the adenosine systems in mediating this response.

Multiple studies in humans have looked at the relationship between physical activity and pain sensitivity. Improved pain control is associated with exercise for a variety of painful conditions including chronic low back pain, fibromyalgia, osteoarthritis, neuropathic pain, and CRPS. Various studies of healthy individuals have demonstrated a relationship between a person’s participation in general physical activity and aerobic exercise and his or her tolerance to pain. Ellingson and colleagues studied a group of healthy women using self-reported and accelerometer measures to determine their degree of sedentary versus active behavior and evaluated the relationship between levels of physical activity and average intensity and unpleasantness of a heat pain stimulus. They found that those participants who met physical activity recommendations (moderate intensity aerobic activity for minimum of 150 minutes per week or vigorous aerobic activity for minimum of 75 minutes per week) had significantly lower pain unpleasantness ratings and pain intensity ratings to noxious thermal stimuli than the participants who did not meet this activity level. Delfa de la Morena and colleagues studied a more localized hypoalgesic response to exercise. They evaluated the effect of repeated eccentric exercise at high intensity on pressure pain thresholds of the wrist extensors in healthy subjects. They demonstrated that an initial bout of high-intensity eccentric exercise resulted in delayed-onset muscle soreness and lowered pressure pain thresholds 24 hours postexercise consistent with hyperalgesia. Subjects underwent a second bout of eccentric exercise after 7 days. The subjects had delayed-onset muscle soreness after this second bout of exercise, but did not show lowered pain pressure thresholds indicating that there had been adaptations that limited the local hyperalgesia. The authors speculated that a combination of neural, mechanical, and cellular adaptations were responsible for this change. Naugle and colleagues sought to determine an intensity threshold for aerobic exercise-induced hypoalgesia. They studied the immediate impact on healthy young men and women of moderate (50% heart rate reserve) and vigorous (70% heart rate reserve) aerobic exercise on pressure and heat pain modulation. Both moderate and vigorous exercise was associated with reduced pain intensity to heat stimulus, and there was a dose response effect with more reduction in pain intensity after vigorous exercise. Vigorous exercise also led to increased pressure pain thresholds.

Researchers have sought to determine how exercise impacts the pain modulatory processes in the central nervous system. Naugle and Riley studied healthy adults and looked at the relationship between self-reported physical activity and pain facilitatory and inhibitory function. They used tests of temporal summation to evaluate pain facilitation and conditioned pain modulation to evaluate endogenous pain inhibition. They found that levels of total and vigorous physical activity predicted pain facilitation and inhibition. Meeus and colleagues also used temporal summation and conditioned pain modulation to evaluate exercise-induced analgesia in female patients with rheumatoid arthritis, chronic fatigue syndrome, and fibromyalgia and sedentary healthy control subjects. Pain assessment was done before and after submaximal exercise on a bicycle ergometer at 75% age-predicted maximal heart rate for up to 15 minutes. The patients with rheumatoid arthritis demonstrated exercise-induced analgesia expressed by reduced temporal summation of experimental pain. Paracetamol reinforced this response to exercise. The patients with chronic fatigue and fibromyalgia had a less clear response with different responses at different experimental pain locations, suggesting that their endogenous pain modulating system is dysfunctional. Ellingson and colleagues tried to determine if conditioned pain modulation was responsible for exercise-induced hypoalgesia. They studied healthy subjects and evaluated heat pain intensity and unpleasantness at rest, after nonpainful aerobic cycling exercise, and after painful aerobic cycling exercise. Both exercise conditions resulted in decreased pain responses, although with a larger magnitude response in the painful, more intense exercise condition. These studies suggest that conditioned pain modulation plays a role in exercise-induced analgesia, but that other mechanisms also contribute.

Studies of motor cortex stimulation allow the evaluation of the hypoalgesic effects of activation of the motor cortex without the concomitant action of the musculoskeletal and cardiovascular systems. All exercise involves motor cortex activation and several studies have tried to evaluate the role that the motor cortex plays in exercise-induced analgesia. Yezierski and coworkers studied the effect of electrical stimulation of the motor cortex in primates showing inhibition of the spinal neuronal responses to noxious pressure and pinch stimuli. A study of transcranial magnetic stimulation of the motor cortex in patients with chronic neuropathic pain using PET imaging for evaluation showed that stimulation induced activity during and after stimulation in multiple brain regions including sensory pathways known to be involved in pain processing. Pain relief correlated with regional blood flow changes. A recent randomized, sham-controlled trial of repetitive transcranial magnetic stimulation of the motor cortex in chronic myofascial pain showed significant reduction in daily pain scores and enhanced corticospinal inhibition. These studies all suggest that activation of the motor cortex contributes to activation of endogenous pain modulatory systems.

There is evidence, however, that endogenous pain modulatory systems may be dysfunctional in at least some chronic pain states. Knauf and Koltyn studied the effect of isometric exercise on pain perception in patients with diabetes with and without diabetic neuropathy. They assessed exercise-induced muscle pain and used an experimental thermal pain protocol to assess temporal summation. Patients with diabetes with and without diabetic peripheral neuropathy performed isometric exercise (grip) at 25% maximum voluntary contraction for 3 minutes. Exercise-induced muscle pain was significantly higher in the patients with peripheral neuropathy and they also rated the exercise as more effortful. Temporal summation was assessed in both groups before exercise and was found to be present in both groups with increasing pain ratings with repeated stimulation. After exercise the patients without peripheral neuropathy showed no temporal summation indicative of the presence of exercise-induced hypoalgesia. The patients with peripheral neuropathy continued to have temporal summation and did not seem to gain any pain-relieving benefit from the exercise. Naugle and colleagues summarized the hypoalgesic effects of exercise in healthy and chronic pain populations in a meta-analytic review. The studies reviewed evaluated the impact of acute exercise on experimentally induced noxious stimulation. Different exercise modalities were studied including aerobic, isometric, and dynamic resistance exercise. Both healthy populations and various chronic pain populations were studied. In the healthy populations the evidence supports that all three types of exercise decrease the perception of experimentally induced pain. In the chronic pain populations the results were more mixed depending on the particular chronic pain condition being studied. For chronic low back pain, findings were similar to healthy participants. For patients with regional chronic pain, exercise of muscles outside the painful region reduced pressure sensitivity of the painful area, but exercise of the painful muscles tended to increase pain sensitivity in those muscles. In fibromyalgia, low-to-moderate intensity activity seems to be effective for eliciting exercise-induced hypoalgesia. However, in some studies of individuals with widespread pain, moderate-to-vigorous activity led to hyperalgesia. Nijs and colleagues also reviewed the literature and came to similar conclusions.

In summary, the literature supports that acute bouts of exercise consistently activate the endogenous pain modulatory systems in healthy adults and have mixed effects on patients with chronic pain depending on the specific diagnosis and type of exercise. The mechanisms involved in exercise-induced hypoalgesia are multifactorial. Likely important contributors to the effect include the release of endogenous opioids and activation of spinal and supraspinal pain inhibitory pathways.

Exercise effects on specific chronic pain problems

The role of exercise prescribed in a therapeutic manner for the management of a variety of chronic pain conditions has been widely studied. These studies frequently assess function and quality of life, and pain. The effects of exercise for a few of these conditions are summarized next.

Nonspecific low back pain

The effects of exercise on chronic, nonspecific low back pain have been widely studied. There are many approaches to exercise for back pain including specific exercise protocols targeting specific impairments, general exercise protocols, mind/body techniques of motor control, and multidisciplinary pain management and functional restoration. A 2010 review of 37 randomized controlled trials found that exercise therapy for low back pain, compared with usual care, decreased pain intensity and improved disability and long-term function. There was no evidence that one particular type of exercise was better than another. Mannion and colleagues noted a similar difficulty in demonstrating significant differences in outcomes from different exercise approaches to chronic nonspecific low back pain. Smith and Grimmer-Somers summarized evidence of the long-term effectiveness of exercise therapy for chronic low back pain. In reviewing 15 trials using various exercise approaches, they found evidence of ongoing pain reduction and decreased recurrence rates up to 6 months after treatment. Some of the evidence for different exercise approaches for chronic low back pain is summarized next.