Introduction: Long-term treatment of chronic muscu-loskeletal pain with opioids often causes a cluster of unpleasant side effects such as constipation, dizziness and cognitive impairment and is likely to lead to tolerance and to hyperalgesia, which is clinically important but not yet well researched. In this study we investigated the development of hyperalgesia after long-term treatment with opioids in patients with chronic low back pain (cLBP). The goal of this prospective longitudinal study was to investigate the long-term (> 1.5 years) effects of opioid analgetics on thermal sensation and pain thresholds and to follow the changes in pain sensitivity for 6 months during opioid withdrawal.

Methods: Using quantitative sensory testing (QST), we compared thermal sensation and pain thresholds on the palm of the hand and the low back bilaterally among three groups: patients with cLBP and long-term treatment with opioids (group 1, n=35); opioid-naive patients with chronic low back pain (group 2, n=34) and subjects with neither pain nor opioid intake (group 3, n=27). The effects of age, sex, pain duration, duration and dose of opioid intake, comorbidity (depression) and self-reported pain intensity assessed by QST were investigated.

All patients were allocated to a 3-week multidisciplinary functional restoration programme that emphasized biopsychosocial factors and included continuous tapering of opioid dose. During the study all patients kept records of the medication they used.

Results: Group 1 patients showed significantly delayed reaction to cold and warm stimuli on the back, compared with both group 2 and group 3. Pain thresholds for cold and heat on the hand were similar in group 1 and 2 but significantly reduced in these groups compared with group 3. Age, sex, pain duration, duration and dose of opioid intake, and self-reported pain intensity, but not depression, correlated significantly with QST results.

Discussion: The present study demonstrated that long-term opioid use significantly delayed thermal QST responses but had no measureable analgesic effects in patients with chronic low back pain. While the pain thresholds in groups 1 and 2 did not differ before opioid withdrawal, both groups 1 and 2 were more sensitive to pain than group 3 (healthy controls). This finding confirms that chronic low back pain itself might cause increased pain sensitivity, which seems not to be counteracted by opioid medication. Rather, treatment in the multidisciplinary pain therapy programme had positive effects on pain thresholds in opioid-naive patients but not in patients after opioid withdrawal. The opioid-naive patients of group 2 showed normalized pain thresholds 6 months after therapy, while the former opioid-positive patients of group 1 still had significantly decreased pain thresholds despite 6 months’ abstinence.