There are four main theories of biomechanical changes elicited by spinal manipulation. These are (1) release of entrapped synovial folds or meniscoids; (2) restoration of buckled motion segments; (3) reduction of articular or periarticular adhesions; and (4) normalisation of ‘hypertonic’ muscle by reflexogenic effect (Evans and Breen 2006). However, the relevance of these theories on clinical outcomes remains uncertain. This is due to the fact that although a number of studies have quantified motion with spinal manipulation, biomechanical effects were found to be transient in nature (Colloca, Keller and Gunzburg 2004; Colloca et al. 2006; Coppieters and Butler 2008; Funabashi et al. 2016), and no plausible evidence has yet been found in support of a lasting positional change (Bialosky, George and Bishop 2008a). So far, only the muscular reflexogenic theory has some plausible evidence in support of its mechanical explanation (Clark et al. 2011; Colloca and Keller 2001; Currie et al. 2016); nevertheless, the clinical assertion that hypertonic muscles are influenced by an increased stretch reflex gain is not yet proven (Zedka et al. 1999). Furthermore, a common explanation widely propagated for the success of spinal manipulation is that it corrects changes in biomechanical dynamics, specifically position and movement faults, detected on examination. However, a majority of the current literature does not validate this explanation. This is because palpation has not been found to be a reliable method to identify areas requiring spinal manipulation (Seffinger et al. 2004; Walker et al. 2015), and the thrust applied during a therapy cannot be specific to an intended location (Frantzis et al. 2015) and varies between therapists (Cambridge et al. 2012).
The success of spinal manipulation in treating musculoskeletal disorders despite theoretical inconsistencies in its supposed biomechanical mechanisms indicates the possibility of concurrent additional mechanisms. Biomechanical changes evoked as a result of spinal manipulation may induce neurophysiological responses by influencing the inflow of sensory input to the central nervous system (Pickar 2002). Moreover, the mechanical force applied during spinal manipulation could either stimulate or silence mechanosensitive and nociceptive afferent fibres in paraspinal tissues, including skin, muscles, disc or discs, facet, tendons and ligaments (Currie et al. 2016; Randoll et al. 2017). These inputs have been thought to stimulate pain-processing mechanisms and other physiological systems connected to the nervous system (Bialosky et al. 2008a, 2009; Clark et al. 2011; Maigne and Vautravers 2003; Pickar 2002). In support of this hypothesis, Pickar and Bolton (2012) developed the notion that neural responses arising from the nervous system due to mechanical stimulus might be because of alterations in peripheral sensory input from paraspinal tissues.
Taken together, it can be said that changes in spinal biomechanics trigger the chain of neurophysiological responses responsible for the therapeutic outcomes associated with spinal manipulation, and there is a potential for combined biomechanical and neurophysiological effects following spinal manipulation. However, the possible interaction of these effects has frequently been overlooked in the current literature. The possibility of a combined effect is important to consider as biomechanical characteristics of a given spinal manipulation are shown to have a unique dose–response relationship with biomechanical, neuromuscular and neurophysiological responses (Cambridge et al. 2012; Downie, Vemulpad and Bull 2010; Nougarou et al. 2016). For example, paraspinal electromyographic (EMG) responses have an apparent dependence on the force/time characteristics of the mechanical thrust applied during spinal manipulation (Colloca et al. 2006). Therefore, future clinical studies should be conducted to investigate the relationship between variations in mechanical parameters (e.g., preload, peak force and thrust) and physiological responses and the relevance of varying parameters with biological and therapeutic outcomes.
Neurophysiological effects of spinal manipulation
Many authors have long postulated that spinal manipulation exerts its therapeutic effects by means of a number of neurophysiological mechanisms working on their own or in combination (Bialosky et al. 2008a, 2009; Pickar 2002). These mechanisms involve complex interactions between the peripheral nervous system and the central nervous system, and have been thought to be set in motion when spinal manipulation activates paraspinal sensory afferents (Pickar and Bolton 2012). The activation of sensory neurons is presumed to occur either during the manoeuvre itself and/or because of changes in spinal biomechanics. These paraspinal sensory inputs are assumed to alter neural integration either by directly influencing reflex activity and/or by affecting central neural integration within motor, nociceptive and possibly autonomic neuronal pools (Pickar 2002). However, since current biomechanical studies of spinal manipulation are unable to observe the changes occurring in the brain following the therapy – for example, how sensory afferent neurons produce neurophysiological effects by interacting with those in the central nervous system – the validity and relevance of theorised neurophysiological mechanisms in relation to therapeutic outcomes remains unclear. Implications for specific neural mechanisms of manipulation are suggested from associated neurophysiological responses, which have been observed in mechanistic studies.
Over the past decades, a number of specific and non-specific neural effects of spinal manipulation have been reported, including increased afferent discharge (Pickar and Bolton 2012), central motor excitability (Pickar 2002), alterations in pain processing (Lelic et al. 2016), reduction of temporal summation (Randoll et al. 2017), stimulation of the autonomic nervous system (Sampath et al. 2015), lessening of pain perception (Bialosky et al. 2008b) and many more. These neural responses collectively implicate mechanisms mediated by the nervous system. Figure 1.1 presents a new theoretical model that illustrates the proposed neurophysiological effects of spinal manipulation based on the findings of current mechanistic literature. This model is heavily inspired by the comprehensive model presented by Bialosky and colleagues (2009), which was drawn interpreting the literature of several forms of manual therapy including nerve-based, mobilisation, manipulation and message therapies; hence, its relevance to spinal manipulation alone is unclear. The theoretical model we propose herein is diagrammed including only the literature on HVLA thrust manipulation.
Figure 1.1. Neurophysiological and neurochemical effects of spinal manipulation
Neuromuscular effects
Muscle activation
The muscular reflexogenic response is an important theory that is frequently used to explain the mechanism of spinal manipulation. The muscles of the human body have some reflex responses, by means of their reflex arcs, to protect themselves from potentially harmful force (Evans 2002). In manual therapy literature, the reflexogenic effect is often explained using one of the prominent theories of pain, the pain–spasm–pain cycle (Travell, Rinzler and Herman 1942), which suggests that pain causes muscular hyperactivity (spasm) and muscle spasm reflexly produces pain, establishing a self-perpetuating cycle. Although this pain model lacks unequivocal support from the literature (van Dieën, Selen and Cholewicki 2003), there is enough evidence in support of the fact that low back pain (LBP) patients experience significantly higher levels of paraspinal muscle activity than normal healthy individuals during static postures (Geisser et al. 2004; Hodges and Moseley 2003; Lewis et al. 2012). Spinal manipulation is thought to disrupt the pain–spasm–pain cycle by reducing muscle activity through reflex pathways. Pickar (2002) postulated that the mechanical stimulus applied during manipulation on paraspinal tissues might influence the sensory receptors to cause muscle inhibition, and suggested that afferent stimuli would target this inhibition as a reflex response. Herzog (2000) proposed that the neuromuscular response to spinal manipulation could involve two reflex pathways – the capsule mechanoreceptor pathway and the muscle spindle pathway – and these pathways might differentiate by muscle activity onset delay.
EMG signals are commonly used to quantify changes in muscle activation following