J Manipulative Physiol Ther 2006 (Jan); 29 (1): 14–21 ~ FULL TEXT
Julita A. Teodorczyk-Injeyan, PhD, H. Stephen Injeyan, PhD, DC, Richard Ruegg, PhD, DC
Division of Research,
Canadian Memorial Chiropractic College,
Toronto, Ontario, Canada
OBJECTIVE: To examine the effect of a single spinal manipulation therapy (SMT) on the in vitro production of inflammatory cytokines, tumor necrosis factor alpha, and interleukin (IL) 1beta, in relation to the systemic (in vivo) levels of neurotransmitter substance P (SP).
METHODS: Sixty-four asymptomatic subjects were assigned to SMT, sham manipulation, or venipuncture control group. SMT subjects received a single adjustment in the thoracic spine. Blood and serum samples were obtained from subjects before and then at 20 minutes and 2 hours after intervention. Whole-blood cultures were activated with lipopolysaccharide (LPS) for 24 hours. Cytokine production in culture supernatants and serum SP levels were assessed by specific immunoassays.
RESULTS: Over the study period, a significant proportion (P ≤ .05) of sham and control subjects demonstrated progressive increases in the synthesis of tumor necrosis factor alpha and IL-1beta. Conversely, in a comparable proportion of cultures from SMT-derived subjects, the production of both cytokines decreased gradually. Normalization of the observed alterations to reflect the changes relative to self-baselines demonstrated that, within 2 hours after intervention, the production of both cytokines increased significantly (P < .001 to .05) in both controls. In contrast, a significant (P < .001 to .05) reduction of proinflammatory cytokine secretion was observed in cultures from SMT-receiving subjects. In all study groups, serum levels of SP remained unaltered within 2 hours after intervention.
CONCLUSIONS: SMT-treated subjects show a time-dependent attenuation of LPS-induced production of the inflammatory cytokines unrelated to systemic levels of SP. This suggests SMT-related down-regulation of inflammatory-type responses via a central yet unknown mechanism.
From the FULL TEXT Article:
Although the exact biologic mechanism(s) of systemic effects of spinal manipulative therapy (SMT) remain unclear, the current model of somatoautonomic or spinovisceral reflex provides a critical link between spinal manipulation, the autonomic nervous system, and various visceral functions.  Animal experiments have demonstrated that somatoautonomic reflex effects are apparent in multiple nonmusculoskeletal systemic responses, including those of the cardiovascular, digestive, urinary, endocrine, and the immune system. [2, 3] On the other hand, similar studies in humans are still limited particularly with respect to the effect of SMT on the integrated biologic activities of the nervous and immune systems.
Neural immunoregulation that sustains immune homeostasis is based on the reciprocal communication between the immune and the nervous systems. It is executed primarily by biologic actions of numerous soluble mediators such as cytokines and neurotransmitters. [4, 5] The close association of autonomic nerve terminals with macrophages and lymphocytes also facilitates a chemically mediated transmission between nerves and immune cells.  These observations suggest strongly that spinovisceral reflex effects might include alterations in the functional activity of cells in the immune and/or inflammatory responses.
Inflammation, which is a normal response to disturbed homeostasis caused by infection, injury, or trauma, is associated with augmented production of numerous proinflammatory and immunoregulatory cytokines and neurotransmitters. Among the principal mediators of inflammation, tumor necrosis factor α (TNF-α), interleukin (IL) 1, and the neurotransmitter, substance P (SP), play pivotal roles in the regulation of both local and systemic inflammatory responses. [7-9] Recent studies have shown that the production of such mediators is highly increased in patients with discogenic back pain, implicating their biologic activity in the pathogenesis of this condition. 
The relationship between SMT and the synthesis of biologic regulators of physiological responses by inflammatory cells remains unexplored. Earlier studies by Brennan et al [11, 12] also suggested that peripheral blood lymphocytes from SMT-treated normal subjects may produce higher concentrations of biologically active TNF-α possibly induced by SP. The aim of the present study was, therefore, to examine if a single spinal manipulation has an effect on the lipopolysaccharide (LPS)–induced in vitro production of TNF-α and IL-1β. Because the production of these cytokines may be regulated by biologic actions of SP,  the effect of SMT on the systemic (in vivo) release of SP was also examined.
The present study supports the hypothesis that the spinovisceral reflex effect can encompass functional activity of the immune system. We believe this to be the first report to demonstrate that a single manipulative thrust to an aberrant vertebral motion segment in the upper thoracic spine of asymptomatic subjects results in down-regulation of the capacity of human leukocytes for the production of proinflammatory cytokines induced by LPS. The study design used in the present report allowed us to create a model in which the LPS-induced inflammatory response in vitro, in control subjects submitted to multiple venipunctures, became augmented. The progressive augmentation of the production of both TNF-a and IL-1ß observed in cell preparations from VC and SHM subjects was not unexpected. It has been shown that even minor surgical intervention and local blood clotting may lead to increased expression of messenger RNA for inflammatory cytokines such as TNF-a and IL-6 and thereby result in increased secretion of both cytokines.  Furthermore, immunological and inflammatory changes may be induced by a number of factors such as crush, cut, contusion, biopsy, or stress, and the biologic function of circulating blood cells can be altered by the impact of events in local tissues. [18, 19] Several of these factors, without doubt, contributed to increased levels of constitutive secretion of TNF in LPS-free controls. Interestingly, such increases were not observed in SMT-treated subjects. The SMT group was exposed to the same number of venipuncture procedures as the control groups. Thus, attenuation of TNF-a and IL-1ß in cultures from SMT-receiving normal subjects suggests that spinal manipulation effectively ameliorated the subsequent physiological responses of the peripheral blood cells to inflammatory stimuli.
Furthermore, cultures from SMT-treated individuals displayed reduced sensitivity to otherwise highly effective concentrations of LPS (Table 2, Table 3). LPS-induced activation of mammalian cells involves the interaction of LPS with serum LPS-binding protein (LBP) and recognition of LBP-LPS complexes by a membrane-bound receptor, CD14, on human monocytes.  In addition to CD14, LPS interacts with transmembrane receptor(s) responsible for activating signal transduction.  It has been well documented that the LPS-induced inflammatory cascade manifested by the production of TNF-a, followed by subsequent induction of IL-1, may be initiated by CD14/LBP-dependent or independent binding determined by endotoxin concentration.  Thus, diminished response to higher concentrations of LPS observed after SMT may suggest that one of the activation pathways was affected adversely. In addition, the present study used whole-blood cultures. High (1-10µg/mL) concentrations of LPS induce proinflammatory cytokine production by both mononuclear cells and polymorphonuclear neutrophils.  Attenuated proinflammatory cytokine synthesis observed after SMT may therefore reflect altered activity of polymorphonuclear neutrophils. As reported earlier by Brennan et al, [11, 12] the biologic responses of these cells may be indeed altered by spinal manipulation.
However, results of the present report do not confirm the earlier observations of Brennan et al  concerning the effect of SMT on the synthesis of TNF-a by peripheral blood cells and its relation to the plasma level of SP. In contrast to data presented in this report, Brennan et al [11, 12] observed enhanced production of TNF-a in mononuclear cell cultures from subjects receiving spinal manipulation accompanied by elevation in plasma levels of SP. It should be noted that the study design and experimental protocol described by Brennan et al  cannot be compared directly to ours. The former study used fractionated mononuclear blood cell cultures obtained at 15 minutes after treatment and activated briefly by a single dose of LPS.  This approach did not allocate for the effect of SMT on delayed and sustained LPS activation which occurs 6 to 12 hours after exposure and reflects the capacity of a given cell population for the cumulative synthesis of inflammatory mediators. 
The present report used a whole-blood culture system in which cytokine-releasing cells were maintained in the donor's (autologous) serum. Thus, cytokine-producing cells were cultivated in the presence of all serum-associated, potentially regulatory polypeptides, including SP. Generally, this neurotransmitter is implicated in the induction of inflammatory mediators' production.  Although the synthesis of TNF-a and IL-1ß was attenuated in a significant proportion of the SMT-receiving subjects and the opposite effect was observed in both controls, no apparent fluctuations in serum concentrations of SP were observed in any of the studied groups (Fig 1). It is, therefore, unlikely that SP could have contributed to the observed modulation of cytokine production.
In the present study, the manipulative thrust was directed to the thoracic spinal segment determined to exhibit aberrant segmental movement. An audible cavitation was used as our guide for successful manipulation. However, as determined by Ross et al,  such cavitation does not necessarily occur specifically at the manipulated segment. Furthermore, multiple cavitations might occur, although a single audible cavitation is perceived by the clinician. Thus, no claim can be made of the specificity of the effect of SMT relative to the level of spinal segment involved.
The nature of anti-inflammatory effect observed after spinal manipulation is presently unknown. Recent recognition of a neural immune circuit as instrumental in the control of inflammation [26, 27] may suggest that activation of endogenous anti-inflammatory pathways should be considered as a systemic consequence of SMT. Recent investigations have shown that the vagus nerve may control central inflammatory responses. [28, 29] Surgical or chemical vagotomy can render animals sensitive to endotoxin shock.  Conversely, systemic inflammatory responses to endotoxin are attenuated after the direct electrical or pharmacological stimulation of the vagus nerve.  Investigations in vitro have shown that LPS-induced human macrophage production of TNF-a, IL-1ß, and IL-6 was inhibited after the exposure of these cells to the principal neurotransmitter of parasympathetic nervous system, acetylcholine (Ach). [32, 33] Of interest is that Ach does not inhibit the synthesis of IL-10, an anti-inflammatory cytokine produced simultaneously with TNF, IL-1, and IL-6 by LPS-activated macrophages.  The Ach-mediated reactions, described now as “cholinergic anti-inflammatory pathway,” [32, 33] occur after vagus nerve stimulation and the subsequent interaction of Ach with the nicotinic receptor a7 subunit expressed by many cells of the immune system. 
Direct experimental evidence that the spinovisceral reflex effect on the immune system may be mediated by vagus nerve activation is currently not available. However, the vagus nerve provides innervation to the principal body organs as well as the reticuloendothelial system. It has been postulated that, in humans, the vagus output can be regulated through conditioning, acupuncture, and other therapies targeting the cholinergic anti-inflammatory pathway.  It is now feasible that such therapies could include SMT, and experimental work in this area should yield the supporting evidence.
The clinical significance of observations presented in this study are also related to studies on the pathophysiology of discogenic low back pain and sciatica. Investigations have revealed that local inflammatory response and the associated production of proinflammatory mediators, such as TNF-a, may represent a major factor in the genesis of pain and functional changes in the neural activity of spinal nerve roots.10 Hypoalgesic effects of SMT have been already reported.  Furthermore, a preliminary report by Mohammadian et al  has demonstrated that a single spinal manipulation treatment may lessen capsaicin-induced local hyperalgesia and allodynia, suggesting that a central anti-inflammatory mechanism might be indeed activated by manipulative therapy.
Recent clinical studies have shown that TNF-a blockade by anti–TNF-a monoclonal antibodies was highly effective in reducing sciatic pain.  Based on these and related results, development of drugs targeting the production and/or action of proinflammatory cytokines is now suggested as critical for pain management in patients with low back pain and sciatica.  In contrast to pharmacological interventions, SMT is likely to present a noninvasive and efficacious alternative to such therapies. Future studies are now necessary to address the issue of the effect of SMT on the magnitude and duration of inflammatory responses in chiropractic patients.
SMT-treated subjects show a time-dependent attenuation of LPS-induced production of the inflammatory cytokines unrelated to systemic levels of SP. This suggests SMT-related down-regulation of inflammatory-type responses via a central yet unknown mechanism.
This was a novel experimental design to explore somatovisceral effects of SMT
through experimental assessment of the inflammatory-type responses
using cytokine secretion as an outcome measure.
The findings suggest that SMT may be a potential therapeutic modality to
reduce inflammatory response.