Thursday, December 6, 2012

Exercise & Chronic Pain Relief, Meta-Analysis

ExercisePhysical exercise is often recommended as a component of treatment and rehabilitation for many types of chronic pain conditions. However, a better understanding of how exercise influences pain perception, to what extent, and in which patients is needed. A recently reported systematic literature review and data meta-analysis attempted to better define the effects of brief exercise on pain perception in adults with chronic pain.

A team at the Pain Research and Intervention Center for Excellence, University of Florida, Gainesville, conducted an extensive literature search for studies using repeated measures designs to examine the effect of acute (ie, brief) isometric, aerobic, or dynamic resistance exercise on pain threshold and pain intensity measures [Naugle et al. 2012]. Pain threshold denotes the point at which a noxious stimulation is first perceived as painful, and pain intensity is measured once the threshold is reached.

All studies measured pre- and post-exercise responses to experimental pain-invoking stimuli involving either heat, cold, pressure, or electrical current. The 3 types of exercise used were: 1) aerobic — eg, stationary cycling, running, or step exercise; 2) dynamic resistance — eg, strength training involving muscle contractions over moving joints; or 3) isometric — eg, strength training using static contractions without joint movement.

Writing in the December 2012 edition of the Journal of Pain, Kelly Naugle and colleagues report locating 50 potential studies in their search, of which 25 met inclusion criteria. These encompassed 622 participants (437 healthy, 185 with chronic pain); 12 studies used isometric training, 11 aerobic exercises, and 2 dynamic resistance.

The results suggest that all 3 types of exercise reduce perception of experimentally-induced pain in healthy participants, with effects ranging from small to large depending on pain-induction method and exercise protocol. The mean sizes for beneficial pain threshold [thr] and intensity [int] effects [expressed as Cohen’s d] for aerobic exercise were moderate (d[thr] = 0.41, d[int] = 0.59), while the mean effect sizes were large for isometric exercise (d[thr] = 1.02, d[int] = 0.72) and dynamic resistance (d[thr] = 0.83, d[int] = 0.75).

Therefore, among healthy persons, all 3 types of acute exercise appear to reduce perceptions of experimentally-induced pain, with the largest effect sizes resulting from isometric and dynamic resistance exercise. [Understanding effect sizes was previously discussed in a Pain-Topics UPDATE here.]

In patients with chronic pain the results were more limited and not as straightforward, and only studies involving either aerobic or isometric exercise were discovered for analysis. The magnitude and direction of the effect sizes for these two types of exercise were highly variable and appeared to depend on the chronic pain condition being studied as well as the intensity of the exercise. For example…

  • Studies examining chronic low back pain found that aerobic exercise strongly reduced pain intensity (d [int] = 1.50) but had little effect on pain threshold (d[thr] = 0.08). Other forms of exercise were not studied.

  • Some, but not all, of the studies variably suggested that isometric exercise of nonpainful muscles for individuals with regional chronic pain conditions (eg, shoulder myalgia) produces an overall increase in the pain threshold and may be a helpful method to temporarily relieve pain in aching muscles (d[thr] range = –0.67 to 1.62).

  • Results in patients with fibromyalgia suggested that pain-relieving effects may only be elicited in response to low-to-moderate intensity aerobic exercise; whereas, isometric exercise was often detrimental in terms of pain threshold and intensity.

  • In persons with widespread pain associated with chronic fatigue syndrome, pain-relieving effects of exercise were practically nonexistent, with moderate-to-vigorous exercise often having a hyperalgesic (pain increasing) effect.

The researchers conclude that, while trends could be identified, the optimal dose of exercise that is needed to produce pain-reducing (hypoalgesia) effects could not be systematically determined due to the deficient amount of data available. And, although the research suggests that acute exercise has a hypoalgesic effect on experimentally-induced pain in healthy adults, and both hypoalgesic and hyperalgesic effects in adults with chronic pain, there are important questions that need to be addressed in future research.

COMMENTARY: This review and meta-analysis is important because, as much as anything, it helps to describe the current state of evidence on the effects of exercise for chronic pain conditions. In that regard it is interesting, but it also demonstrates the unsatisfactory body of research on this topic and is not enlightening from a clinical perspective that seeks guidance for better patient care.

The results of this study by Naugle and colleagues suggest that exercise generally might have hypoalgesic effects — potentially increasing the pain threshold and reducing its intensity — but not all types of exercise or for all types of pain. Yet, as is so often the case in the pain research field, there were significant limitations of available data and subsequent analyses that prohibit reaching externally valid conclusions. For example, focusing only on the studies pertaining to patients with chronic pain….

  • Study sizes were small — group sizes ranged from merely 8 subjects to 26 (in one study) — so almost all trials were underpowered to detect statistically significant effects.

  • Far more women than men were enrolled, and participant ages and other demographics were not taken into account as variables potentially affecting outcomes.

  • The researchers did not assess or take into account heterogeneity across studies, although this was probably significantly large due to small sample sizes, the various pain conditions examined, and the different pain stimuli that were applied. It is possible that these factors should have precluded conducting a meta-analysis [as was discussed in an earlier UPDATE here.]

  • Naugle et al. also did not report 95% confidence intervals for the calculated effect sizes, which were likely extremely wide and imprecise, or P-values; so the statistical significance of the various effect sizes is undetermined.

  • Only aerobic and isometric exercises were studied in chronic pain and, at that, not all pain conditions were examined using both types of exercise or measuring both pain intensity and pain threshold. So, the portrait of exercise effects is still very sketchy and incomplete.

  • The researchers concede that few studies for meta-analysis compared pain perceptions associated with exercise to a control condition of any sort (eg, resting control or a comparator treatment). Essentially, the body of evidence consisted of observational studies, which have many limitations and sources of bias for producing reliable and valid outcomes in a meta-analysis.

By using experimental pain stimuli and standardized brief exercise protocols the various studies attempted to objectively assess and quantify ameliorating effects of exercise. However, the small sample sizes defeated the determination of statistically valid outcomes, and the approach says nothing about the dose of exercise (duration or intensity) that might produce optimally beneficial clinical effects.

Probably most disappointing, this line of research does not measure effects of brief exercise on subjects’ preexisting clinical pain; so, the external validity of exercise, of any type or duration, applied to real-world patients with chronic pain is lacking. Assuredly, an appropriate exercise regimen has been beneficial to many patients with all types of pain conditions, but the current state of research does little to enlighten our understanding of those effects.

REFERENCE: Naugle KM, Fillingim RB, Riley JL. A Meta-Analytic Review of the Hypoalgesic Effects of Exercise. J Pain. 2012(Dec);13(12):1139-1150 [abstract here].

eNotifications Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.