Myofascial pain syndrome

Pathogenesis and pathophysiology
Article section 4 of 14.  Previous  Next

By Charles Argoff MD and Howard Smith MD

Although the pathophysiologic mechanisms that produce the symptoms of myofascial pain syndrome remain largely speculative, the complex nociceptive processes and pathways that subserve our perception of myofascial pain do not. In the last decade, indirect evidence has arisen from the investigation of nociceptive and neuropathic pain mechanisms, allowing inferences to be made regarding parallel processes believed operant in the myofascial model of pain. In this regard, human and animal studies suggest that the pathogenesis of 3 defining features of the myofascial pain syndrome are related to central nociceptive processing within the spinal cord. These neurophysiologic events, which are elicited by trigger point stimulation, include referred pain, reflex changes in motor activity (local twitch response), and reflex autonomic phenomena in or around the pain reference zone (Simons 1993). Further evidence for a primary role of the CNS was confirmed when nociceptive processes were demonstrated to be qualitatively altered in patients with chronic myofascial pain. This study, which compared 40 patients with chronic myofascial pain to 40 normal volunteers, concluded that their nociceptive pain may be mediated by low-threshold mechanosensitive afferents projecting to sensitized dorsal horn neurons (Bendtsen et al 1996).

As evidence continues to suggest that the CNS plays a role in these processes, no human biopsy study has ever unambiguously identified the trigger points that presumably underlie the clinical expression of myofascial pain syndrome. The peripheral mechanism and initiating event of the myofascial trigger point, its associated sensory hyperactivity (variable tenderness), and the taut band are, as well, largely unknown. The concept that the trigger point represents a transient state of neuromuscular dysfunction is a recent introduction. Support for this evolving concept comes from several clinical observations including: (1) the restoration of affected muscle tissue to normal with a variety of therapeutic techniques; (2) the identification of latent trigger points in clinically asymptomatic individuals; and (3) the inconsistent histopathological results obtained in biopsies to date. Simons has expanded on this concept and proposed that any hypothesis that attempts to define these peripheral initiating events and mechanisms of myofascial pain will also need to account for 4 clinical features: (1) the sustained tension maintained within taut bands, which occurs in the absence of quantifiable EMG motor unit activity; (2) the manner in which multiple and highly variable factors including local muscle overload could precipitate a physiologic event initiating the myofascial trigger point; (3) the presence of substances that sensitize nociceptors in the region of the trigger point and cause their overactivity; and (4) the fact that any technique (ie, TP injections, dry needling, massage) that restores the muscle to its full normal stretch length is therapeutic (Simons 1991). The most frequently referred to model for myofascial pain syndrome is based on the local energy crisis theory originally described by Simons. This hypothesis, which has evolved since its initial description, has yet to be confirmed.

Gerstner and colleagues utilized single-voxel proton magnetic resonance spectroscopy ((1)H-MRS) before and after pressure-pain testing to assess glutamate (Glu), glutamine (Gln), N-acetylaspartate (NAA), and choline (Cho) levels in the right and left posterior insulae of 11 individuals with myofascial temporomandibular disorders and 11 matched control individuals (Gerstner et al 2012). Glu levels were significantly lower in all individuals after pain testing. Among those with temporomandibular disorders, left-insular Gln levels were related to reported pain, left posterior insular NAA and Cho levels were significantly higher at baseline than in control individuals, and NAA levels were significantly correlated with pain-symptom duration, suggesting adaptive changes. The results suggest that significant central cellular and molecular changes can occur in individuals with temporomandibular disorders (Gerstner et al 2012).

Progress in nociception research has led to the identification of numerous nociceptive processes that may contribute to the biological basis of myofascial pain. Acute or chronic tissue trauma can result in a cascade of events including hyperalgesia from nociceptor sensitization and associated somatic (skeletal muscle spasm) and autonomic reflexes (ie, tachycardia, vasoconstriction, pupillary dilation, and sweating). Many nociceptive primary afferent fibers possess dual capabilities, including orthodromic response signaling to the CNS and antidromic release of algogenic substances peripherally (neurogenic inflammation). Prolonged nociceptor activity can be followed by neuronal plasticity at every level of sensorimotor processing and can result in changes that may then become the substrate for chronic pain and autonomic dysfunction. The distribution of nociceptors within skeletal muscle is interestingly confined to the fascia between muscle fibers, with the highest concentration of nociceptors near the muscle endplate region where myofascial trigger points are usually found (Hong 1996). It has been proposed that multiple sensitive loci exist in a trigger point region, and that each locus contains 1 or more sensitized nociceptive nerve endings. The colocalization of these sensitive loci to established trigger point regions near the muscle endplate support the concept that multiple and highly variable causes of local muscle injury are mediated through a common pathogenic pathway, resulting in the development of a predefined trigger point.

Cheshire suggested that the muscle spindle plays a role in both dystonia and myofascial pain (Cheshire et al 1994). The concept that dystonia may represent a disorder of the sensory nervous system (Hallett 1995) has raised some intriguing parallels to myofascial pain syndrome. In recognizing the clear differences between these 2 neurologic disorders, a number of similarities do exist, including the involuntary and sustained muscle contractions that can be triggered by soft tissue trauma or repetitive overuse (Jankovic 1994). Like myofascial pain syndrome, dystonia may be painful despite normal muscle histology and can be relieved by specific sensory maneuvers. The effective use of low dose botulinum toxin in both myofascial trigger points and dystonia, though therapeutically more effective in the latter, is suggested to result from blockage of the neuromuscular junction in the gamma motor neuron leading to a reduction in spindle afferent activity. This potential mechanism of action of botulinum toxin is in part based on the often dramatic muscle relaxation following a Botox injection, which is disproportionate to the actual weakness that it causes. Though EMG studies of myofascial trigger points have conflicted in the past, observations of sustained spontaneous EMG activity within the 1 to 2 mm nidus of myofascial trigger points suggest a possible target for Botox. Whether this electromyographic finding of spontaneous activity, which disappeared with withdrawal of the needle by as little as 1 mm, comprised only normal motor end plate potentials or, as these investigators hypothesized, sustained contraction of intrafusal muscle fibers, remains to be resolved (Hubbard and Berkoff 1993). Furthermore, debate exists whether trigger points are sites of abnormal muscle contraction, or rather consist of electrically silent muscle contracture. The shorter duration of benefit seen with Botox in myofascial trigger points (5 to 6 weeks) as compared to 3 months for dystonia, suggests that factors responsible for the trigger points were still present when the Botox wore off and were not maintained by muscle contraction (Cheshire et al 1994). Hubbard and Berkoff theorize that these perpetuating factors generate spontaneous EMG activity from sympathetically maintained intrafusal muscle fiber contraction, causing an involuntary low-grade but symptomatic muscle tension (Hubbard and Berkoff 1993). This sympathetic activity would also explain the autonomic symptoms associated with trigger points and provide a mechanism by which local injury and nociception cause local tension and by which emotional factors cause widespread pain and tension. The activity of nociceptive afferent fibers from the trigger point to the spinal cord could then induce a feedback loop that maintains trigger point activity, possibly by nociceptor sensitization or local autonomic dysregulation and vasoconstriction.

Shah and Gilliams described biochemical differences found between active and latent myofascial trigger points (MTrPs), as well as in comparison with healthy muscle tissue (Shah and Gilliams 2008).

Ge and colleagues showed that glutamate and isotonic saline injections into the latent MTrPs induced higher peak pain intensity than into the non-MTrPs (both P < 0.05) (Ge et al 2008). Glutamate injection induced higher peak pain intensity than isotonic saline injection into either latent MTrPs or non-MTrPs (both P < 0.05) (Ge et al 2008). Muscle camps were observed in 92.86% of the subjects following glutamate injection into the latent MTrPs but not into the non-MTrPs (P < 0.001). No muscle cramps were recorded following isotonic saline injection into either the latent MTrPs or the non-MTrPs (Ge et al 2008). These results suggest that latent MTrPs could be involved in the genesis of muscle cramps. Focal increase in nociceptive sensitivity at MTrPs may constitute one of the mechanisms underlying muscle cramps (Ge et al 2008).

Kimura and colleagues demonstrated that glutamate injection into MTrPs decreased skin temperature and blood flow in the peripheral area (Kimura et al 2009). The magnitudes of the reduction were comparable to those induced by the breath-hold maneuver, which has been used to induce sympathetic vasoconstrictor response (Kimura et al 2009). Kimura and colleagues concluded that the combination of glutamate injection into latent MTrPs together with the breath-hold maneuver did not result in further decrease in skin temperature and blood flow, indicating that sympathetic vasoconstrictor activity is fully activated by nociceptive stimulation of MTrPs (Kimura et al 2009).

Zhang and colleagues demonstrated an attenuated skin blood-flow response after painful stimulation of latent MTrPs compared with non-MTrPs, suggesting increased sympathetic vasoconstriction activity at latent MTrPs (Zhang et al 2009). Additionally, laser Doppler flowmetry was more sensitive than thermography in detection of the changes in skin blood flow after intramuscular nociceptive stimulation (Zhang et al 2009).

Li and colleagues randomly injected a bolus of either hypertonic saline (6%, 0.1 mL, each), glutamate (0.1 mL, 0.5 M, each), or isotonic saline (0.9%, 0.1 mL, each) into a latent MTrP and a non-MTrP located in the right or left gastrocnemius medialis muscles (Li et al 2009). Injections of either hypertonic saline, glutamate, or isotonic saline into the latent MTrPs induced a higher VAS(peak) and larger VAS([area under the curve] auc) than the non-MTrPs (all, P<0.05) (Li et al 2009). Also, the MTrPs with referred pain after painful injections were found to show higher VAS(peak) and larger VAS(auc) than those without referred pain (both, P<0.001), confirming the existence of hypersensitivity at latent MTrPs and providing evidence that there exists hypersensitivity at latent MTrPs, and the occurrence of referred muscle pain is associated with higher pain sensitivity at latent MTrPs (Li et al 2009).

When compared to non-MTrPs, maximal VAS and the area under VAS curve were significantly higher and larger during sustained nociceptive mechanical stimulation (SNMS) of latent MTrPs (both, P < .05); there was a significant decrease in pedunculopontine tegmental nucleus 10 minutes, 20 minutes, and 30 minutes post-SNMS of latent MTrPs (all, P < .05). Muscle cramps following SNMS of latent MTrPs were positively associated with VASauc (r = .72, P = .009) and referred pain area (r = .60, P = .03). Painful stimulation of latent MTrPs can initiate widespread central sensitization. Muscle cramps contribute to the induction of local and referred pain (Xu et al 2010).

In This Article

Historical note and nomenclature
Clinical manifestations
Pathogenesis and pathophysiology
Differential diagnosis
Diagnostic workup
Prognosis and complications
References cited