Inflammation and cardiovascular autonomic dysfunction in rheumatoid arthritis: a bidirectional pathway leading to cardiovascular disease

Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease that affects the synovial joints, causing pain and loss of function. Interestingly, this inflammatory pattern also causes damage to the heart and vasculature, increasing the deleterious consequences of the disease. Indeed, cardiovascular disease (CVD) represents the primary cause of morbidity/mortality in RA. The increased rates of CVD in RA can be partially explained by the presence of traditional risk factors, such as diabetes, dyslipidaemia, hypertension and physical inactivity. However, recent studies have identified other factors that contribute to the pathophysiology of CVD in RA. Among them, impairment in autonomic cardiovascular regulation, otherwise known as cardiovascular autonomic dysfunction, has received growing attention over recent years (Adlan et al. 2017).

Using mostly indirect methods, previous studies have presented equivocal information regarding the presence of autonomic dysfunction in RA patients. In order to address such limitations, Adlan et al. (2017) published a recent study in The Journal of Physiology, whereby they performed a comprehensive assessment of autonomic function in RA, including assessments of heart rate, cardiac and sympathetic baroreflex functions using the modified Oxford technique, and recordings of muscle sympathetic nerve activity using microneurography. In order to avoid the potential confounding effects of hypertension, the main outcomes of the study were compared between RA patients with and without hypertension, as well as between normotensive and hypertensive individuals without RA. Using this elegant design, they observed an increased heart rate, a reduction in cardiac baroreflex sensitivity, increased sympathetic activity and a preserved sympathetic baroreflex sensitivity in RA patients. Interestingly, these responses occurred independently of the presence of hypertension, and were correlated with inflammation and pain, highlighting the link between autonomic dysfunction and the typical symptoms of RA.

The cross‐sectional design of this study precludes causal inferences between inflammatory status and autonomic dysfunction in RA; however, information from previous studies does provide support for this concept (Tracey, 2002; Suzuki & Nakai, 2016). More specifically, influential reviews have discussed the bidirectional relationship between the immune system/inflammation and the central nervous system (Tracey, 2002; Suzuki & Nakai, 2016). These authors propose that the central nervous system can regulate immune system function via two plausible mechanisms: receptor‐specific pro‐ or anti‐ inflammatory effects of noradrenaline (norepinephrine) released by sympathetic postganglionic fibres on primary and/or secondary lymphoid organs (Suzuki & Nakai, 2016), or by the direct anti‐inflammatory effects of acetylcholine on splenic macrophages (the source of acetylcholine in the spleen is still unclear), which suppress the release of tumour necrosis factor (TNF) and other cytokines (Tracey, 2002). Conversely, the immune system can also modulate autonomic activity via the activation of afferent neural inputs from the lymphoid organs to the brain centres responsible for autonomic regulation (Tracey, 2002; Suzuki & Nakai, 2016), through the direct effects of inflammatory markers accessing autonomic brain centres devoid of a blood–brain barrier (Tracey, 2002), or via the direct effects of cytokines produced locally in the brain by glia (Tracey, 2002). This reflex arc, which has been termed the ‘inflammatory reflex’ by Tracey et al. (2002), is an integrative response aiming to provide a fast, balanced and reversible defensive response to the presence of pathogens, injury and disease. However, under conditions of altered homeostasis, the impairment of one axis of this reflex arc may seriously compromise the other, and it is plausible that this neural–immune imbalance will occur in several inflammatory conditions, including RA, leading to the development and progression of cardiovascular disease.

Studies evaluating patients with other diseases have proposed models which include autonomic dysfunction and inflammation as integral components of the cardiovascular disease continuum, as shown in Fig. 1. It is possible that the reduced vagal tone may denote a compromised anti‐inflammatory cholinergic pathway (Tracey, 2002). Likewise, depending on the pathological condition, and/or the adrenergic receptor type, an increased noradrenaline release from sympathetic neurons might cause detrimental effects on lymphoid organs, increasing the production of pro‐inflammatory cytokines, in turn increasing the general inflammatory state of the body (Suzuki & Nakai, 2016). Alternatively, it is possible to suggest that macrophage activation and the consequential increase in centrally and/or peripherally driven inflammation may contribute to an increased excitability of the premotor sympathetic neurons and to inhibition of cardiac vagal preganglionic neurons in the medulla, culminating in an increased sympathetic outflow to the heart and vasculature, and decreased parasympathetic outflow to the heart (Adlan et al. 2017).

Figure 1. Neural–immune imbalance as integral components of the cardiovascular disease continuum.

The aggravation of this vicious cycle will promote further prejudice to the cardiovascular system. ACh, acetylcholine; hs‐CRP, high sensitivity C‐reactive protein; IL‐6, interleukin‐6; NE, noradrenaline; TNF‐α, tumour necrosis factor α. [Colour figure can be viewed at wileyonlinelibrary.com]

Given the characteristics of RA disease, it is possible to suggest that inflammation may be the trigger for causing abnormalities in autonomic function in this disease state. However, more important than stating which axis triggers the abnormal response, it is pivotal to note that the impairment of one of them will potentially culminate in activation of the other, producing a cyclic initiation of both autonomic dysfunction and inflammation for patients with RA. The exacerbation of this neural–inflammatory imbalance will consequentially promote maladaptive and detrimental effects in the heart and vasculature, eventually leading to the development of the cardiovascular disease, as depicted in Fig. 1. In light of this, and as acknowledged by Adlan et al. (2017), it is important to investigate the effects of strategies which target both inflammation and autonomic nervous system function in RA patients, mainly focusing on cardiovascular disease outcomes.

Recently, the cholinergic anti‐inflammatory pathway has been investigated as a therapeutic alternative to several diseases, including RA (Syngle et al. 2015; Koopman et al. 2016). Accordingly, it has been elegantly demonstrated that stimulation of vagal nerves using a surgically implantable neuromodulation device inhibits TNF production and reduces disease activity in patients with RA for up to 84 days (Koopman et al. 2016). Similarly, there is also evidence indicating the benefits of disease‐modifying anti‐rheumatic drugs in inflammatory and autonomic profiles of RA patients (Syngle et al. 2015). However, whilst most of these findings do not associate the observed improvements with cardiovascular disease outcomes, other studies have demonstrated the beneficial impact of such interventions on cardiovascular risk markers, reinforcing the importance of the role of the autonomic nervous system as a potential pathway for improvement. Finally, there is a lack of information regarding the effects of non‐pharmacological and non‐surgical therapies on inflammation and cardiovascular autonomic dysfunction in patients with RA. Given the well‐documented benefits of exercise training on CVD profile and cardiovascular autonomic function in several populations, this intervention may be a potential alternative therapy for targeting the neural–inflammatory imbalance in RA patients. In line with this hypothesis, Janse van Rensburg et al. (2012) observed an increase in vagal‐related indices of heart rate variability following 3‐months supervised exercise training in females with RA. However, these findings may have been influenced by the presence of a significant initial bias, i.e. the experimental group presented lower heart rate variability than the control group at baseline. In addition, heart rate variability was the only measure of autonomic function performed in this study, and further information regarding the effects of exercise training on other autonomic parameters, such as muscle sympathetic nerve activity and baroreflex function, remain relatively unclear. Since cardiovascular autonomic function is subject to a complex regulation, future studies should endeavour to investigate the impacts of exercise training and other non‐pharmacological interventions on a wider variety of autonomic indices, specifically focusing on the occurrence and progression of cardiovascular diseases and subsequent cardiovascular events in RA patients.

Additional information

Competing interests

None declared.