It has long-been suggested that SM exerts beneficial effects by affecting the nervous system [13
]; however, to date few studies have examined these claims in humans with LBP. As such, utilizing advanced neurophysiologic assessment techniques to investigate the effects of a commonly used-- but poorly understood-- treatment for LBP is particularly innovative. The most novel findings of the present study are: i) that a single spinal manipulation does not systematically alter corticospinal or the short-latency stretch reflex excitability of the erector spinae muscles in patients with chronic LBP or asymptomatic controls (at least when assessed ~ 10-min following manipulation), and ii) that only when spinal manipulation induces an audible joint sound the erector spine short-latency stretch reflex is attenuated. Below we will discuss these findings in the context of understanding the physiological effects of spinal manipulation.
A recent review of chronic LBP provides evidence for two prominent pain theories [44
]. One of these pain theories, the pain-spasm-pain model of chronic LBP, suggests that pain leads to muscular hyperactivity (spasm), which in turn causes pain. One of the neural pathways of the pain-spasm-pain cycle posits that a hyperactive spinal stretch reflex forms the basis of the cycle (Figure ). Specifically, it has been suggested that feedback of nociceptive afferents on the gamma-motorneurons will increase the sensitivity of the muscle spindles to stretch, which results in excitatory input to the alpha-motorneurons that will subsequently increase muscle activation (for review see [44
]). While several studies suggest there is no increase in spindle sensitivity with low back pain or noxious stimulation of paraspinal tissues [25
], many authors have still postulated that SM functions via
the pain-spasm-pain model by reducing the underlying nociceptive stimulus and consequently attenuating the stretch reflex, with the end organ effect being an overall reduction in muscle activity [45
]. Indeed, several studies have noted reduced paraspinal voluntary EMG amplitude following SM of individuals with LBP [51
], and we recently reported that a combination of manual therapies (incorporating both SM and soft-tissue techniques) normalizes side-to-side differences in the activation patterns of trunk muscles of individuals with sub-acute LBP, as determined by muscle functional magnetic resonance imaging [13
Figure 6 A commonly proposed neural pathway suggested to form the basis of a pain-spasm-pain cycle. Specifically, it has been suggested that feedback of nociceptive afferents (N) on the gamma-motorneurons (γ) will increase the sensitivity of the muscle (more ...)
In the present study we did not quantify changes in muscle activity following SM, but rather assessed the effects of SM on the evoked short-latency stretch reflex amplitude. Although our observation of no pre- versus post-manipulation difference in patients with chronic LBP or asymptomatic controls suggested that SM did not systematically alter the short-latency stretch reflex, we did observe a significant decrease in the short-latency stretch reflex when data were analyzed based on whether SM resulted in an audible joint sound. Many clinicians routinely consider the success of a thrust manipulation technique based on the presence or absence of an audible response. While some evidence suggests that an audible response is not associated with improved clinical outcomes [35
], there are differences in joint laxity and motion when an audible pop is associated with the manipulation [37
]. This may reflect the successful and rapid separation of the joint surfaces resulting in cavitation and an audible response. It has been hypothesized that the rapid stretch of the periarticular muscles and connective tissue associated with SM causes the reduction in spinal reflexes [17
]; however, to our knowledge no previous studies have reported differential physiologic effects dependent upon whether SM results in an audible response. Thus, our finding that SM alters the short-latency stretch reflex--a critical component of the pain-spasm-pain model of LBP (Figure )--only when an audible response occurs is novel. As stated before, the short-latency stretch reflex occurs in response to a muscle being rapidly stretched, which excites the Ia afferent fibers within the muscle spindles [55
]. This observation suggests that when SM results in an audible response it mechanistically acts by down-regulating the sensitivity of the muscle spindles and/or the various other segmental sites of the Ia stretch reflex pathway. It is also possible that the change in reflex activity associated with subjects having an audible release during SM may relate to gapping in the joint surfaces, as it was recently shown that vertebral segments that cavitated during SM gapped (separated) more than those that did not [38
]. This greater joint gapping could result in the break-up of small adhesions present even in normal joints, or due to increased muscle or connective tissue tension surrounding those joints, before SM. Consequently, SM that results in an audible response may conceivably function to restore greater motion to a vertebral segment (as opposed to SM that does not result in an audible response), and this biomechanical effect could result in subsequent changes in reflex activity as we observed.
We did not observe changes in MEP amplitude following SM in patients with LBP, asymptomatic controls, or when data were grouped according to whether an audible response was observed. When a single pulse transcranial magnetic stimulation stimuli is applied to the motor cortex at an intensity above motor threshold, high-frequency indirect waves (I waves) are elicited in the corticospinal tract [56
], which are modifiable by many mechanisms (i.e., glutatmate, GABA, acetylcholine, etc.) [39
] that influence the amplitude of the MEP. Thus, our finding of no change in the MEP indicates that a single SM treatment in patients with chronic LBP does not alter global excitability of the corticospinal tract, at least when assessed ~ 10-min following the manipulative intervention. To date only one other study has examined the effects of SM on corticospinal excitability of the low back muscles using transcranial magnetic stimulation [15
]. In this study Dishman and colleagues examined the effects of a single SM treatment on MEP amplitude in asymptomatic
young adults, and observed a transient increase in the MEP following SM. The MEP facilitation was short-lived however-- as MEP amplitude was increased 10-secs following SM but had returned to baseline levels less than 20-seconds after SM. Thus, in the present we would have missed any short-term, transient effects that occurred as a result of SM. Additionally, in our work as well as that conducted by Dishman et al. it is possible that segmental changes in the nervous systems excitability (e.g., cortical level changes) may have been confounded by no change in or opposite changes in excitability at a different segmental level (e.g., spinal level changes) as the MEP amplitude elicited using single-pulse
transcranial magnetic stimulation can be influenced at both the cortical and spinal levels. To more fully explore the effects of SM on cortico-cortical excitability it is suggested that future investigations utilize paired-pulse transcranial magnetic stimulation to measure intracortical facilitation and inhibition.
There are several limitations of the present study that should be mentioned. First, it should be noted that the present work was conducted in patients with mild-to-moderate chronic LBP and asymptomatic controls, and that these individuals only received a single high-velocity low-amplitude SM thrust with outcome measures assessed shortly after the manipulative treatment. As such, it is possible that i) SM may result in different physiologic responses in other populations (e.g., sub-acute LBP), ii) that a course of SM treatment may have a more pronounced effect (e.g., three weeks of SM twice per week), and/or iii) that greater or lesser effects may have been observed at various time points following manipulation. Additionally, we chose to study chronic LBP patients (as opposed to acute or sub-acute LBP patients) due to the staggering economic costs that are associated with chronic LBP and the fact that many patients with chronic LBP seek manipulation therapy as a treatment option for their LBP [5
]. However, it is possible that the neurophysiologic responses may be different if other groups of LBP patients had been studied as patients with LBP symptom duration for < 16 days are reported to be more likely to respond favorably to SM [57
]. Further, we cannot rule out the potential for a placebo effect to influence our observed reduction in the stretch reflex in individuals who exhibited an audible response to SM. However, with this stated, it seems unlikely that this finding was driven by a placebo effect as one would likely expect to observe a concomitant change in corticospinal excitability-- as a placebo effect would likely be assumed to have systemic effects (as opposed to having a local, selective effect on the stretch reflex only).