Heart rate variability in patients with low back pain: a systematic review

Introduction

Low back pain has the highest disability burden worldwide [1]. In most cases, low back pain is associated with a broad range of factors including genetic, physical, psychological, and social [2]. The current evidence shows that chronic pain conditions are associated with morphological and functional changes in the entire nervous system including the peripheral receptors, the spinal cord, and the brain [3, 4]. Longitudinal studies composed of patients with chronic low back pain showed changes in corticostriatal circuitry and mesolimbic brain structures including an increase in the activity of cognitive-emotional brain areas (i.e., prefrontal cortex, amygdala, insula, and anterior cingulate cortex) [5], an increase in connectivity between the prefrontal cortex and nucleus accumbens [6], and reduction of white matter tract integrity [7]. The brain circuitry identified as being crucial to pain chronification in patients with low back pain is also involved in appetitive and aversive learning [8].

A comprehensive view of the interaction between brain areas and visceral structures was proposed by the neurovisceral integration model [9]. This model describes a complex neural system that integrates signals from inside and outside the body and adaptively regulates physiological cognitive and behavioral responses, perception, and action including sensory and motor changes [10]. In this model the amygdala, which has its control mediated by the prefrontal cortex and outputs to the autonomic nervous system (ANS), can be influenced during threat and uncertainty [11], [12], [13]. One method to assess the interconnection of the central nervous system and the heart is the use of heart rate variability (HRV) [9, 12, 14, 15], a non-invasive method used to evaluate ANS modulation of the cardiac sinus node and which describes the oscillations between consecutive electrocardiogram RR intervals [15].

The neurovisceral integration model can be considered as a flexible neural network associated with self-regulation and adaptability [9, 16]. Despite the complexity of this neural system, it is possible that physiological measures, such as HRV, can serve as indices of the degree to which this system provides flexible, adaptive regulation in the presence of challenges (i.e., environmental, and also pain) [17], [18], [19], [20], [21]. It has been suggested that higher levels of HRV at rest indicate a context of highly adaptive emotional responses, low HRV is associated with health impairments including cardiovascular disease, mood disorders and increased morbidity [22]. In fact, reduction of HRV has been shown in different chronic pain conditions such as low back pain [23], neck pain [24], fibromyalgia [25, 26], complex regional pain syndrome [27], whiplash-associated disorders [28], and phantom limb pain [29].

Pain is inherently salient, aiming to protect our body from actual or potential tissue damage and people in pain may present physiological and behavioral protective responses [30]. Therefore, pain is an intrinsic threat and can alter the sympatho-vagal balance. It is possible that low HRV, as an index of activity in a set of neural structures involved in physiologic, affective, and cognitive regulation, may serve as a useful marker for a range of physical and psychological disorders including chronic pain. Given this substantial amount of data, the aim of this review was to systematically obtain data on resting HRV in people with chronic low back pain.

Materials and methods

Results

Study characteristics

The bibliographic search identified 2,873 potential articles. After the duplicates were removed, 2,187 studies remained. Of the references included for evaluation, 76 studies were selected by title and their abstracts were read; after selection of the abstracts, 12 articles remained for the complete reading. After selection by eligibility according to the inclusion criteria, two studies were selected for this review. Another seven studies were considered for inclusion, obtained from lists of references and citations of relevant articles (Figure 1).

Figure 1:

Selection of studies for inclusion in the systematic review.

Discussion

This study aimed to identify the resting HRV in patients with chronic low back pain. The main findings of the present study suggest a predominance of sympathetic nervous system activity measured by HRV parameters in patients with chronic low back pain in comparison to healthy participants. We found a predominance of sympathetic tone with a lower activation of parasympathetic tone characterized by increased BF (n.u.) and a lower E:I ratio in participants with chronic low back pain, which suggests low vagal tone or may an increased sympathetic flow in patients with pain.

Changes in resting HRV characterized by lower vagal tone have also been reported in other chronic pain conditions such as rheumatoid arthritis [35], fibromyalgia [36], shoulder and neck pain [37], irritable bowel syndrome [38], whiplash-associated disorders [39], chronic prostatitis [40], chronic pelvic pain [41], complex regional pain syndrome [27], head/orofacial pain and chronic migraine [42, 43], and chronic fatigue syndrome [44]. Reyes del Paso et al. [45] in a study composed of 54 patients with fibromyalgia, identified lower mean values in all domains of HRV frequency, indicating a general decrease in HRV, when compared to healthy controls on baseline. Hallman and Lyskov [24] aimed to investigate autonomic nervous system regulation, physical activity, and daily stress perception in 55 patients with chronic neck and shoulder pain. The authors found shorter RR intervals and HRV reduction when compared to the control group. These findings characterize a lower activation of parasympathetic tone in the chronic pain group.

Tracy el al. [46]. aimed to identify changes in HRV in a wide range of chronic pain conditions including fibromyalgia, temporomandibular disorders, chronic neck and shoulder pain, chronic low back pain, and others. The authors reported that chronic pain conditions were associated with a reduction of HRV, in particular in HF domain and time domain measurements. Regarding HF, the authors observed a moderate decrease in patients with chronic pain when compared to healthy controls, and the effect size was considered small (ES=−0.39). In the time domain, the most common HRV measures were RMSSD (root mean square of successive RR interval differences), SDNN (Standard Deviation of the average NN intervals) and RRI (wave interval RR). RMSSD was shown to be strongly associated with parasympathetic cardiac influence. In the review carried out by Tracy et al. [46], there were no significant differences in RMSSD between individuals with chronic pain and controls. Their results also showed that chronic pain is associated with a decrease in SDNN. This finding was attributed to a lower activation of parasympathetic tone. The main difference between this review and the study of Tracy et al. [46] was that we sought to identify HRV changes exclusively in patients with low back pain when compared to healthy volunteers. However, studies that met the inclusion criteria included only patients with chronic low back pain.

The findings of this review are in accordance with the results of other studies composed of people with chronic low back pain. For example, Telles et al. [47] investigated 62 patients with chronic low back pain and found a predominance of LF-HRV parameter (sympathetic activity) in their participants. Regarding time domain measures, the study showed that participants with chronic low back pain presented lower values for pNN50 (mean number of times per hour in which the change in consecutive normal sinus), SDNNi (mean of the standard deviations of all the NN intervals), and RMSSD parameters at baseline. These measures are strongly associated with vagal tone, suggesting that individuals with low back pain have lower parasympathetic activity [48]. Roy et al. [49] evaluated HRV in a sample of 20 participants with low back pain. It was observed that participants had a predominance of LF activity at baseline, thus characterizing higher sympathetic activity and consequently lower parasympathetic activity. In the study by Zhang et al. [50] with a total of 36 individuals with acute low back pain, the authors observed a predominance of LF in relation to HF. These studies were not included in the present review because they did not have a control group with healthy participants, which makes it difficult to interpret baseline data since, to date, there are no normal values reported in the literature.

Resting HRV measurement may be an important clinical measure in chronic pain conditions. Gockel et al. [23] in a group of 46 patients with low back pain, sought to identify the association of HRV parameters with pain intensity and disability. The authors found that HRV was associated not only with greater pain intensity but also with high levels of disability, presenting evidence that HRV was lower among participants with moderate disability compared to those with minimal disability. The values of RMSSD and HF parameters were lower in participants with a higher degree of disability, suggesting that the parasympathetic heart rate modulation was significantly lower in this sample. Koenig et al. [51] recruited 61 patients, a group with chronic post-traumatic neck pain and healthy controls, to verify if the reduction in HRV measurements was associated with high levels of catastrophizing. The authors observed that individuals with chronic neck pain have lower HF compared to healthy controls and report higher levels of catastrophizing. Maladaptive emotional responses such as catastrophizing, for example, may represent poorly adaptive emotional regulation and predict disease and mortality [28].

This study is not without limitations. Although the studies have good methodological quality, it is not possible to generalize the results due to the small number of studies included, the small sample sizes, the different protocols used to measure HRV and the different parameters. We did not include studies reporting participants with clinical history of cardiovascular, respiratory, infectious, metabolic or autoimmune diseases. We decided to include studies even though the they did not reported a detailed description of the sample characteristics. We are aware that sample characteristics, including demographic and clinical, might have an impact on HRV. However, description of study participants is an oft-under-reported aspect of HRV research and recommendation on how to report sample characteristic is novel [52, 53]. We tried to decrease the impact of this limitation on our results contacting the authors to obtain more details and including studies with age and sex matched healthy volunteers in the control group. Another limitation is the absence of studies with acute or subacute low back pain. Finally, it was not possible to perform a meta-analysis, as the studies used different protocols and measures. It is recommended that further studies investigate the domains of HRV in patients with low back pain and its association with pain intensity, disability and psychological and behavioral changes. We also consider that there is an important gap in the literature in relation to prospective studies to identify whether HRV can be considered a biomarker for pain chronification, pain intensity or pain-related disability.

Conclusion

There is limited evidence in the literature suggesting that patients with chronic low back pain have a significant reduction in HRV, with lower parasympathetic activation and consequently sympathetic predominance, when compared to healthy controls. The results of this systematic review should be interpreted with caution due to the restricted number of included studies, small sample sizes and different protocols used to measure HRV.