Low back pain (LBP) is a common condition that is currently considered to be the first cause of years lived with disability worldwide, impacting most adults at least once in their lifetime (Hartvigsen et al., 2018). LBP has been the focus of recent reviews calling for action upon it as a critical public health issue with huge economic implications, mostly related to the loss of workdays and healthcare expenditure (Hartvigsen et al., 2018; Vlaeyen et al., 2018).

Most cases of LBP are described as being non-specific, implying that the etiology of the patient’s complaints is unknown and cannot be attributed to a single tissue or pathology (Vlaeyen et al., 2018). A growing body of research is being directed toward identifying biomarkers and biopsychosocial risk factors that influence the course of the disease (Vlaeyen et al., 2018). A variety of inflammatory cytokines profiles and biochemical markers have been linked to different stages of non-specific LBP (Li et al., 2016; Wang et al., 2016; Klyne et al., 2017; James et al., 2018b; Teodorczyk-Injeyan et al., 2019; Lim et al., 2020; Morris et al., 2020). These cytokine profiles were predominantly quantified through serum analysis (Li et al., 2016; Wang et al., 2016; Klyne et al., 2017), although accumulating evidence suggests that it may be possible to detect some of these biomarkers in urine with a high degree of correlation with circulating levels (Moldawer, 1997; Sirera et al., 2003; Prasad et al., 2016). Thus, urinalysis could provide a cheaper, more feasible alternative to blood samples that is much easier to collect in a private or smaller clinical setting.

Recent studies have reported an important role of Tumor Necrosis Factor-alpha (TNF-α) in the pathogenesis and in possible treatment strategies of chronic LBP (CLBP) (Wang et al., 2008, 2016; Sainoh et al., 2016; Lim et al., 2020). TNF-α is a potent pro-inflammatory cytokine that was closely associated with intervertebral disk degeneration, though it may also induce a downstream cascade of other cytokines involved in CLBP such as interleukin-6 or interleukin-1β (Risbud and Shapiro, 2014; Lim et al., 2020). Observational studies have found elevated serum levels of TNF-α in patients with severe sciatica and CLBP compared to patients with milder symptoms or healthy controls (Wang et al., 2008, 2016). The in vitro production of TNF-α was found to be increased (and positively correlated to that of interleukin 1β) in CLBP when compared to healthy control and acute LBP individuals (Teodorczyk-Injeyan et al., 2019). Animal studies have also reported an increased expression of TNF-α in multiple tissues during early chronic stages of intervertebral disk degeneration (James et al., 2018b).

Chiropractic care offers a conservative option for the management of CLBP. To address this condition, chiropractors use a multimodal approach, including manual therapy along with exercise and patient education as part of a toolkit of routine interventions (Walker et al., 2011; Beliveau et al., 2017). However, spinal manipulative therapy (SMT) is the main intervention utilized in chiropractic practice (Beliveau et al., 2017). SMT has been found to be as effective for CLBP as other recommended interventions, such as exercise therapy (Rubinstein et al., 2019; Gevers-Montoro et al., 2021b) and is currently recommended in the latest clinical practice guidelines (Qaseem et al., 2017; Wong et al., 2017). For both manipulative and exercise therapy, multiple potential mechanisms for pain relief have been proposed (Gevers-Montoro et al., 2021a; Wun et al., 2021). The specific mechanisms are still unknown, but human and animal studies have shown a correlation between SMT and a decrease in serum levels and in vitro production of pro-inflammatory mediators, including TNF-α (Teodorczyk-Injeyan et al., 2006, 2018, 2021; Roy et al., 2010; Song et al., 2016). Similarly, investigations on the mechanisms of exercise therapy for CLBP have also found that a reduction in levels of the same cytokine network may be involved in pain relief associated to exercise therapy (Cheng et al., 2015; Leung et al., 2016; James et al., 2018a).

Therefore, our objectives were to examine the possible presence of TNF-α in urine samples of CLBP patients, and to assess whether levels in this cytokine would vary significantly after a period of multimodal chiropractic care mainly based in SMT, in parallel to pain intensity and disability. We hypothesized that the urinary levels of TNF-α would be elevated in patients with CLBP compared to healthy participants. Additionally, we posited that urinary levels of this cytokine would change following a period of chiropractic care.

Twenty-four patients (14 men and 10 women) CLBP met the inclusion and exclusion criteria, accepted to participate in the study by providing PROMs and urine samples, and concluded the treatment period. All participants included in the study were new patients to the Clinic and had never received chiropractic care prior to the study. Four additional patients were initially recruited but after further scrutiny, 2 of them did meet one of the exclusion criteria (suspected inflammatory spondyloarthropathy, and diagnosis of depression) and another 2 did not complete the treatment period, thus, their data were excluded from the analyses. Treatment consisted of a mean of 8.4 individual visits (ranging from 3 to 13 visits), delivered during a mean of 39.4 days (ranging from 19 to 76 days). As far as the control group is concerned, 5 healthy subjects (3 men and 2 women) were recruited for the study. Table 2 provides a summary of the participants demographic and baseline information.

To the best of our knowledge, this is the first study to measure TNF-α levels in urine samples of CLBP patients. In this cohort of 24 patients with CLBP, urinary concentrations of this cytokine were significantly elevated when compared to a reference value from healthy controls. Urinary TNF-α levels were lower after chiropractic care compared with the baseline values. This was accompanied by lower levels of pain and disability after chiropractic care compared with the levels before. However, clinical variables and TNF-α levels were not correlated. While this study cannot make any causal connection between changes in TNF-α and treatment, future studies with a control group should be able to determine whether chiropractic care may relieve pain and decrease disability in patients with CLBP through modulation of the pro-inflammatory cytokine network.

Although urinary levels of inflammatory cytokines have already been measured and highly correlate to serum levels, they have not been used specifically as biomarkers for chronic pain conditions (Moldawer, 1997; Sirera et al., 2003; Prasad et al., 2016). In this study, we detected elevated concentrations of TNF-α in urine samples of CLBP patients (both before and after treatment), when compared to our reference value, suggesting that urine could be a reliable milieu in which biomarkers for CLBP can be studied. After the chiropractic care period, TNF-α concentrations were significantly reduced, with a medium effect size (d = 0.55). Similarly, Wang et al. found a significant difference in the proportion of participants with elevated levels of serum TNF-α positive when comparing healthy subjects and CLBP patients before treatment (Wang et al., 2008). In this study, a multimodal treatment program of conservative care (not including SMT) diminished the initial proportion of TNF-α positive patients, although not reaching levels found in a healthy cohort (Wang et al., 2008).

Data on a potential influence of SMT on proinflammatory cytokines including TNF-α is still scarce. Teodorczyk-Injeyan et al. (2006) collected blood samples of asymptomatic subjects randomly assigned to either SMT, sham SMT or venipuncture control groups. Blood cultures were stimulated with lipopolysaccharide to induce in vitro production of proinflammatory cytokines for 24 h. The study showed a time-dependent significant attenuation of the in vitro production of inflammatory cytokines (including TNF-α) in the SMT group (Teodorczyk-Injeyan et al., 2006). More recently, the same group reported significantly higher in vitro TNF-α production in acute and chronic LBP patients compared to asymptomatic controls, and a significant reduction after 2 weeks of chiropractic care when compared to changes in the same controls (Teodorczyk-Injeyan et al., 2021). This effect of SMT has not been consistently reported throughout the literature. Degenhardt et al. (2017) did not find significant differences in cytokine levels (including TNF-α) when comparing CLBP patients to healthy controls, before or after receiving manual therapy. Probably, defining CLBP after 6 weeks and limiting treatment to one session may have influenced the results. Similarly, Song et al. (2016) were not able to identify any detectable changes in TNF-α, IL-1β and IL-10 plasma levels of rats with experimental neuropathic and postoperative pain receiving SMT sessions. Instead, elevated levels of the anti-inflammatory cytokine IL-10 were observed locally in the spinal cord. Further research is needed in this area to clarify the link between SMT and concentrations of both serum and urinary inflammatory markers.

The physiological mechanism through which SMT can result in a reduction of pro-inflammatory cytokines is still unknown. Recent studies have shown that activation of pain circuits regulates neuroinflammation, both centrally and in the periphery (Ji et al., 2016, 2018). It seems plausible that SMT-induced modulation of nociceptive input (Gevers-Montoro et al., 2021a) could result in a down-regulation of neuroinflammation. Such neuroimmune interactions have been extensively researched in the context of chronic pain, suggesting a bi-directional communication between nociceptive neurons and microglia necessary for the release of pro-inflammatory mediators such as TNF-α (Ji et al., 2016, 2018). Furthermore, central release of TNF-α seems to be required to induce long-term potentiation and central sensitization in the dorsal horn of the spinal cord, inducing persistent pain (Kawasaki et al., 2008; Ji et al., 2016, 2018). Similarly, in both animal models and patients with rheumatoid arthritis or discogenic LBP, TNF-α blocking drugs were shown not only to reverse central pain responses but also to improve disability (Hess et al., 2011; Sainoh et al., 2016).

On the other hand, it has also been suggested that SMT could have a non-specific effect on sympathetic nervous system (SNS) function (Wirth et al., 2019), which could in turn modulate immune responses (Teodorczyk-Injeyan et al., 2006). Involvement of the SNS in the modulation of TNF-α production and release by macrophages has been observed in animal models (Szelenyi et al., 2000), and clinical applications of this interaction have shown promising outcomes (Poyhonen-Alho et al., 2008). Evidence for regulation of cytokine levels by the SNS in chronic pain conditions is still on early stages (David Clark et al., 2018). Further research is needed in order to determine the mechanisms mediating the reduction in TNF-α levels apparently induced by SMT or other conservative interventions. Resolution of inflammation through an inhibition of pro-inflammatory cytokine activity is currently considered a new therapeutic frontier in the field of chronic pain (Ji et al., 2018).

Our findings are in line with previous data suggesting that SMT is an effective approach to reduce pain levels and improve function in CLBP patients (Rubinstein et al., 2019). The levels of pain intensity and disability were both significantly reduced after the treatment period, although these effects were not compared with a placebo. Nevertheless, the changes observed had very large effect sizes (d = 2.33 and 1.17 for pain and disability respectively), which could also be considered clinically significant. Indeed, pain intensity ratings showed a 68% reduction in pain intensity from baseline, largely exceeding the proposed threshold of 30% (Ostelo et al., 2008), while disability scores were 14.6 points lower (out of 100) after treatment, also superior to the estimated minimal detectable change for the ODI of 11.75 points (Johnsen et al., 2013). The present study identified moderate significant correlations between pain intensity levels and disability scores percentage of change, which is consistent with the ODI being an adequate tool to measure LBP-related disability.

On the other hand, no strong correlation was observed between TNF-α levels and NRS-11 and ODI scores, suggesting that the levels of pain and disability are not linearly related to the cytokine profile in our sample or that cytokine levels are affected by other factors. Changes after chiropractic treatment on one variable did not predict possible alterations of the others. This lack of correlation was also reported by Wang et al. (2008) and could be explained by an insufficient sample size. It has also been speculated that TNF-α may be associated with the lack of recovery from LBP, which may explain the lack of correlation with pain and disability (Morris et al., 2020). Nevertheless, other authors reported correlations between pain intensity and pro-inflammatory cytokines (including TNF-α) in plasma (Koch et al., 2007; Sikorska et al., 2019). Interestingly, TNF-α levels were found to be significantly augmented in patients with severe pain when compared to patients with milder pain intensity (Koch et al., 2007; Wang et al., 2016).

To our knowledge, this is the first study to suggest that urinary levels of TNF-α can be found elevated in patients with CLBP when compared to healthy individuals. The results obtained in this preliminary study suggest that a non-standardized period of chiropractic care could have a significant impact on pain and disability levels, as well as on urinary concentrations of pro-inflammatory cytokine TNF-α in CLBP patients. These potential benefits should be further investigated in a proper randomized controlled trial design, with at least one control group that will not receive chiropractic care. Therefore, results from the current study should be interpreted with caution. Further research to elucidate the mechanisms and role of different conservative interventions for CLBP is an essential upcoming step.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Madrid College of Chiropractic Research Ethics, Real Centro Universitario Escorial María Cristina. The patients/participants provided their written informed consent to participate in this study.

CG-M: conceptualization, methodology, investigation, formal analysis, and writing – original draft preparation. MR-S and LL: investigation, formal analysis and manuscript revision. FC-B: investigation, resources, and manuscript revision. DN: conceptualization and writing- review and editing. LÁ-G: conceptualization, supervision, and writing- review and editing. MP: supervision, funding acquisition, and writing- review and editing. AO-DM: conceptualization, methodology, funding acquisition, supervision, writing- review, and editing and approval of final version. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.