Comment on: ‘Longitudinal Patterns of Analgesic and Central Acting Drug Use and Associated Effectiveness in Fibromyalgia’



Charles J. Vierck


In this issue you will find a paper by Wolfe et al., entitled ‘Longitudinal Patterns of Analgesic and Central Acting Drug Use and Associated Effectiveness in Fibromyalgia’ (Wolfe et al., 2013). Patterns of drug use by 3123 fibromyalgia patients are documented over an 11-year period (2000–2010), with assessments of pain, fatigue and functional status at 6-month intervals. Over this period, non-steroidal anti-inflammatory and tricyclic antidepressant drug use declined, while opioid, serotonin and norepinephrine reuptake inhibitors (SNRI; duloxetine and milnacipran) and anticonvulsant (primarily pregabalin) drug use increased. Despite a prevalence by 2010 of treatment with an opioid (46.7% of patients; 12.5% using strong opioids), an SNRI (20.4%) or an anticonvulsant (26.4%), pain scores averaged ≥ 6 (1–10 rating scale) for each of these classes of pharmacological agents. A focus on changes in clinical status from the inception of new centrally acting drug (NCAD) intake (i.e., SNRIs and anticonvulsants) for 508 patients revealed a statistically significant, but clinically meaningless decrease in pain (from 6.21 to 6.04 on the 10-point rating scale). Ratings ≥ 6.0 represent a high level of clinical pain, and a decrease of 2.0 points or greater would have indicated that a clinically meaningful attenuation of pain had occurred (Farrar et al., 2003). Also, NCADs did not significantly affect fatigue or health and disability over the study period. These results extend and elaborate upon a previous report on 1555 fibromyalgia patients over the same time period but without separate analyses of treatment modalities (Walitt et al., 2011). Overall, ratings of pain, fatigue, sleep disturbance, health and disability for these patients waxed and waned without appreciable improvement over the study period.

These investigations of patients across 11 years offer an important perspective on pharmacological approaches to treatment of fibromyalgia pain. They conflict with shorter duration clinical trials against placebo administration. For example, significant attenuation of different combinations of fibromyalgia pain, fatigue, sleep disturbances and health-related quality of life has been demonstrated for amitriptyline, duloxetine, milnacipran and pregabalin, relative to placebo (Häuser et al., 2010, 2011). It is understandable that placebo comparisons are utilized for clinical trials of drug efficacy, but often, administration of the active drug can be detected by subjects who have expectations of therapeutic benefits. Also, these studies necessarily evaluate short-term effects, and reports of reduced pain can occur when a drug improves mood, but does not directly affect nociceptive processing.

In contrast to clinical trials against placebo administration, the 11-year study of patient reports on pain levels provides strong evidence that NCADs minimally affect fibromyalgia pain or lose antinociceptive actions over time (e.g., as a result of neuroadaptive changes in receptor distribution or sensitivity). Wolfe et al. state that ‘… our results not only raise questions as to whether the new fibromyalgia therapies are truly effective, but also how one should determine effectiveness in clinical practice. …’ As a cautionary point, they suggest that patients in their study may not have followed the strict guidelines required of a randomized control trial. However, pharmacotherapy should be judged in the real world of self-administration, motivated by a powerful incentive such as relief of pain. Because NCADs are directed at central processing of pain and/or pain affect, administration of the drug must continue as long as a source of abnormal input remains, activating central nervous system circuits for pain processing. Under these conditions of chronic drug intake for chronic pain, NCADS did not provide a ‘… clinically meaningful benefit’, as indicated by ratings of pain, fatigue or disability. The implications of these findings are clear.

Conflict of interest

None declared.