Cerebrospinal fluid levels of opioid peptides in fibromyalgia and chronic low back pain

Abstract

Background: The mechanism(s) of nociceptive dysfunction and potential roles of opioid neurotransmitters are unresolved in the chronic pain syndromes of fibromyalgia and chronic low back pain.

Methods: History and physical examinations, tender point examinations, and questionnaires were used to identify 14 fibromyalgia, 10 chronic low back pain and 6 normal control subjects. Lumbar punctures were performed. Met-enkephalin-Arg6-Phe7 (MEAP) and nociceptin immunoreactive materials were measured in the cerebrospinal fluid by radioimmunoassays.

Results: Fibromyalgia (117.6 pg/ml; 85.9 to 149.4; mean, 95% C.I.; p = 0.009) and low back pain (92.3 pg/ml; 56.9 to 127.7; p = 0.049) groups had significantly higher MEAP than the normal control group (35.7 pg/ml; 15.0 to 56.5). MEAP was inversely correlated to systemic pain thresholds. Nociceptin was not different between groups. Systemic Complaints questionnaire responses were significantly ranked as fibromyalgia > back pain > normal. SF-36 domains demonstrated severe disability for the low back pain group, intermediate results in fibromyalgia, and high function in the normal group.

Conclusions: Fibromyalgia was distinguished by higher cerebrospinal fluid MEAP, systemic complaints, and manual tender points; intermediate SF-36 scores; and lower pain thresholds compared to the low back pain and normal groups. MEAP and systemic pain thresholds were inversely correlated in low back pain subjects. Central nervous system opioid dysfunction may contribute to pain in fibromyalgia.

Figure 1

Systemic symptoms questionnaire results. FM (white bars), LBP (grey bars) and Normal (black bars) results are shown for each domain. The x-axis shows the systems domains with the total number of questions in parentheses. The y-axis shows the number of positive responses within each domain. The error bars are the 95% confidence intervals. FM scores were higher than Normal for ENT and IBS (p ≤ 0.05), Chest, Headache, Neurological and Bladder (p ≤ 0.01), Fatigue and Musculoskeletal (p ≤ 0.001). FM scores were higher than LBP for Neurological (p ≤ 0.05), Chest (p ≤ 0.01) and Fatigue (p ≤ 0.001). Normal and LBP scores were not different.

Figure 2

SF-36 domain results. Scores were ranked Normal (white bars with down-going 95%CI) > FM (black bars with up-going 95%CI) > LBP (grey bars with up-going 95%CI). The Normal group had significantly higher scores than FM for EP (p ≤ 0.05), SF (p ≤ 0.01), PF, PP, E/F, P and GP, but not MH or CH. All Normal means were highly significantly greater (p ≤ 0.001) than all LBP scores. FM scores were higher than LBP for EP, GP, and CH (p ≤ 0.05), SF (p ≤ 0.01), and MH and P (p ≤ 0.001). FM and LBP were not different for PF, PP or E/F.

Figure 3

Pressure pain thresholds (kg/cm²) determined by dolorimetry. Pain thresholds for FM (triangles) were significantly lower than LBP (squares; p = 0.009) and Normal (circles; p = 0.002) groups. The bars indicated means and 95% confidence intervals.

Figure 4

Tender points. The numbers of tender points detected by dolorimetry with ≤ 4 kg/cm² pressure were higher for FM (triangles, n = 14), than LBP (squares, n = 9, p = 0.03) and Normal (circles, n = 6, p = 0.008) groups. The bars indicated means and 95%CI's.

Figure 5

Met-Enkephalin-Arg⁶-Phe⁷ (MEAP) concentrations (pg/ml) in cerebrospinal fluid (CSF) from normal (circles), low back pain (LBP, diamonds) and fibromyalgia (FM, triangles) subjects. The bars indicate geometric means and 95% confidence intervals. The groups were significantly different by ANOVA (p = 0.0014). MEAP in the FM (p < 0.01) and LBP (p < 0.05) groups were significantly higher than Normal (2-tailed unpaired Student's t-tests).

Figure 6

Semi-logarithmic relationships between pain thresholds and MEAP. FM subjects (black triangles) had pain thresholds below 2.3 kg/cm² (heavy, black error bars). Normal subjects (grey circles, cabled grey error bars) had lower MEAP and higher pain thresholds compared to FM. LBP subjects (open diamonds) had results that covered the full ranges for both variables (upper right error bars). The data for all subjects were linearly correlated (Pearson's correlation coefficient R = -0.38, p < 0.05) with an explained variance (R²) of 0.15. The line appears curved on this semi-logarithmic plot.