TYPE Systematic Review PUBLISHED 02 September 2022 DOI 10.3389/fimmu.2022.945513 Immune-inflammatory and OPEN ACCESS hypothalamic-pituitary-adrenal EDITED BY Toshikazu Kondo, Wakayama Medical University, Japan axis biomarkers are altered in REVIEWED BY Jacob Raber, patients with non-specific low Oregon Health and Science University, United States Urs Nater, back pain: A systematic review University of Vienna, Austria *CORRESPONDENCE Juan P. Sanabria-Mazo 1,2,3†, Ariadna Colomer-Carbonell 1,2,3†, Xavier Borràs xavier.borras@uab.cat Meritxell Carmona-Cervelló 3, Albert Feliu-Soler 3*, Albert Feliu-Soler Xavier Borràs 3*, Mar Grasa 4,5, Montserrat Esteve 4,5, albert.feliu@uab.cat Juan V. Luciano Michael Maes 6, Sı́lvia Edo 3‡, Antoni Sanz 3‡ juanvicente.luciano@uab.cat † and Juan V. Luciano 1,2,3* These authors have contributed equally to this work and share 1 Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain, 2 Teaching, Research and first authorship Innovation Unit, Parc Sanitari Sant Joan de Déu, St. Boi de Llobregat, Spain, 3 Faculty of Psychology, ‡ Autonomous University of Barcelona, Cerdanyola del Vallès, Spain, 4 Faculty of Biology, University of These authors share senior authorship Barcelona, Barcelona, Spain, 5 Biomedical Research Centre in Physiopathology of Obesity and SPECIALTY SECTION Nutrition (CIBERobn), Institute of Health Carlos III, Madrid, Spain, 6 Department of Psychiatry, This article was submitted to Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand Cytokines and Soluble Mediators in Immunity, a section of the journal Frontiers in Immunology This systematic review aimed to investigate immune-inflammatory and RECEIVED 16 May 2022 hypothalamic-pituitary-adrenal (HPA) axis biomarkers in individuals with ACCEPTED 08 August 2022 PUBLISHED 02 September 2022 non-specific low back pain (NSLBP) compared to healthy control. The search was performed in five databases until 4 November 2021. Two CITATION Sanabria-Mazo JP, Colomer- reviewers independently conducted screenings, data extraction, risk of bias, Carbonell A, Carmona-Cervelló M, and methodological quality assessment of 14 unique studies. All studies Feliu-Soler A, Borràs X, Grasa M, Esteve M, Maes M, Edo S, Sanz A and reported the source of the fluid analyzed: nine studies used serum, two Luciano JV (2022) Immune- used plasma, one used serum and plasma, and two studies used salivary inflammatory and hypothalamic- cortisol. We found preliminary and limited evidence (only one study for each pituitary-adrenal axis biomarkers are altered in patients with non-specific biomarker) of increased levels in growth differentiation factor 15 (GDF-15), low back pain: A systematic review. interleukin-23 (IL-23), transforming growth factor–beta (TGF-b), and soluble Front. Immunol. 13:945513. doi: 10.3389/fimmu.2022.945513 tumor necrosis factor receptor 1 (sTNF-R1) in NSLBP. Inconsistent and limited evidence was identified for interleukin-10 (IL-10). Although C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor–alpha (TNF-a) levels appear to increase in NSLBP, only one study per each biomarker reported statistically significant differences. Interleukin-1 beta (IL-1b), interleukin-17 (IL-17), interferon gamma (IFN-g), and high-sensitivity CRP (hsCRP) showed no significant differences. Regarding cortisol, one study showed a significant increase and another a significant decrease. More robust evidence between GDF-15, IL-23, TGF-b, and sTNF-R1 with NSLBP is needed. Moreover, contrary to the findings reported in previous studies, when comparing results exclusively with healthy control, insufficient robust evidence for IL-6, TNF-a, and CRP was found in NSLBP. In addition, cortisol response (HPA-related biomarker) showed a dysregulated functioning in Frontiers in Immunology 01 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 NSLBP, with incongruent evidence regarding its directionality. Therefore, our effort is to find adjusted evidence to conclude which immune-inflammatory and HPA axis biomarkers are altered in NSLBP and how much their levels are affected. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/ display_record.php?ID=CRD42020176153, identifier CRD42020176153. KEYWORDS non-specific low back pain, immune-inflammatory biomarkers, cytokines, hypothalamic-pituitary-adrenal axis, cortisol Introduction levels of IRS biomarkers increase neuropathic and inflammatory pain (15). Furthermore, stressful events (16), poor sleep (17), Non-specific low back pain (NSLBP) is an umbrella term obesity (18), anxiety (19), and depression (20)—elements which that refers to the presence of tension, soreness, or stiffness in the are usually over-represented in patients experiencing chronic lower back region that cannot be attributed to any recognizable pain—are accompanied by alterations of the IRS. pathology or be considered as a symptom of another disease (1, Pain is associated with a dysregulation of the stress-response 2). Depending on the pain duration episode, NSLBP can be system (16, 21). Being exposed to a painful condition is a classified as acute (less than 6 weeks), subacute (between 6 and potential stressor that may result in a triggered cortisol 12 weeks), and chronic (12 weeks or more) pain (3). Currently, response (11, 22). Under normal conditions, cortisol binds to NSLBP is recognized by the World Health Organization as the the glucocorticoid receptor (GR), present on immune cells, and greatest contributor to global disability worldwide (4). The acts as anti-inflammatory, but sustained increased levels of lifetime prevalence is between 70% and 90%, and its point cortisol block cortisol binding with GR as a compensatory prevalence is about 10% (5). The risk of suffering this health down-regulation system and may disrupt the negative condition increases in elder populations, women, smokers, and feedback mechanism by which cortisol normally inhibits the people with obesity (6). The constant pain experience affects the continued release of corticotropin-releasing hormone by the physical and psychological patients’ wellbeing (7), being hypothalamus, events leading to a pro-inflammatory response considered a public health problem with a great economic (23). Moreover, HPA axis dysfunctions including alterations in burden at the societal level (8). cortisol (24–26) and adrenocorticotropic hormone (20) are Although the etiopathogenesis of NSLBP involves several associated with greater pain levels under chronic and mechanisms, there are multiple physical and psychosocial factors neuropathic conditions (15, 27). that play a role in the onset and recurrence of NSLBP (9). Research While some studies have explored the presence of altered has also highlighted that central sensitization—an enhancement in immune-inflammatory biomarker levels in NSLBP, it remains the neural signaling within central nociceptive pathways due to uncertain how the functioning of the IRS and HPA in NSLBP neuroplastic mechanisms—may play a role in the development and differs from the healthy population. Nonetheless, given the maintenance of this pathology (10). In recent years, there has been methodological approach of previous systematic reviews, burgeoning interest in the interplay of the immune system and nowadays, some immune-inflammatory biomarkers that could hypothalamic-pituitary-adrenal (HPA) axis function under chronic potentially be associated with NSLBP severity can be targeted. and non-chronic pain conditions (11, 12). Cytokines related to the Firstly, the systematic review by van den Berg et al. (28) provided immune-inflammatory response system (IRS) (13), such as evidence for a positive association between C-reactive protein interleukin-1 beta (IL-1b) or interleukin-6 (IL-6), have been (CRP) and IL-6 levels and severity of NSLBP, as well as an involved in central sensitization (14, 15). Systemic inflammation association between tumor necrosis factor–alpha (TNF-a) and has been shown to be a key factor in NSLBP, as well as elevated the presence of NSLBP. Secondly, the systematic review by Morris et al. (29) reported elevated CRP in non-specific acute low back pain (acute NSLBP) and increased TNF-a in non- Abbreviations: HC, Healthy control; ICS, Immune confounders scale; IRS, specific chronic low back pain (chronic NSLBP). Finally, the Immune-inflammatory response system; MQ, Methodological quality; systematic review by Lim et al. (30) informed of elevated CRP, NSLBP, Non-specific low back pain; RoB, Risk of bias. TNF-a, and IL-6 levels in NSLBP. Frontiers in Immunology 02 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 The aforementioned systematic reviews used a less restrictive The specific Boolean searches were adjusted according to the selection of articles, including those with heterogeneous Peer Review of Electronic Search Strategies (PRESS) guideline comparison groups [not exclusively healthy controls (HC)] or statement (35). The format of each database is presented in even without a comparison group. The current research tries to Appendix A. The following limits and filters were activated in all disentangle which immune-inflammatory biomarkers are databases if possible: publication date (from 1 January 2005 to 4 altered in NSLBP by reviewing articles using a pure HC November 2021), type of publication (only studies of interest), comparison group. This effort underlies the necessity to clarify species (humans), and languages (English and Spanish). The possible misleading findings previously reported in the NSLBP publication date was defined considering the results obtained in field such as CRP, TNF-a, and IL-6 associated with previous systematic reviews. Moreover, it was continuously NSLBP severity (28–30) as well as CRP and TNF-a with acute updated through the immediate alerts of the selected NSLBP and chronic NSLBP, respectively (29). Crucially, the use databases. The reference list of included papers was examined of a specific tool for assessing the methodological quality (MQ) by a reverse citation search for further analysis. In addition, of aimed studies at exploring immune-inflammatory biomarker reviews, as well as gray literature (a search carried out in Google differences was incorporated for the first time to guarantee Scholar), were used to verify the reference list in order to ensure adjusted and precise results. Finally, unlike any previous that all eligible studies were included. systematic review that focused on NSLBP, the current research also evaluated preliminary evidence of alterations in peripheral levels of HPA-related biomarkers. The interplay role between Eligibility criteria HPA-related biomarkers and immune-inflammatory biomarkers was not explored in previous systematic reviews. The selection of the eligibility criteria was performed This systematic review aimed to investigate immune- considering the recommendations of the PICOS strategy (36). inflammatory and HPA biomarkers in individuals with NSLBP The inclusion criteria were (a) adults (≥18 years of age) with compared to HC. The identification of signature biomarkers primary NSLBP; (b) screening of acute (up to 6 weeks), subacute might help to deepen our understanding of the biological (from 6 to 12 weeks), or chronic (more than 12 weeks) NSLBP pathways underpinning NSLBP. according to medical history (3); and (c) baseline data from experimental, quasi-experimental, case-control, and observational studies, with a HC (defined by the absence of any significant disease, including chronic pain). Materials and methods The exclusion criteria were (a) secondary pain or low back pain associated with a range of serious conditions (e.g., fracture, Protocol and registration systemic or inflammatory diseases, cauda equina syndrome, active infection, scoliosis, post-surgical, modic changes, This systematic review was performed according to the radiculopathy, or cancer); (b) the presence of a severe Preferred Reporting Items for Systematic Reviews and Meta- psychiatric disorder (e.g., bipolar disorder or schizophrenia), Analyses statement (PRISMA) (31). The review protocol was substance dependence/abuse, and neuro-inflammatory/ registered at PROSPERO (registration number: CRD42020176153). neurodegenerative disorders (e.g., stroke, multiple sclerosis, or Alzheimer’s dementia), and immune-inflammatory disorders (e.g., rheumatoid arthritis, inflammatory bowel disease, or Data sources and searches atherosclerosis); (c) no measurement of immune-inflammatory biomarkers or cortisol as an outcome; (d) non-original Systematic searches were conducted in PubMed, Medline®, (editorials, guidelines, systematic reviews, protocols, etc.), Web of Science, PsycINFO, and Scopus. The search strategy clinical trials, case reports, and qualitative studies; (e) animal identified studies that included combinations of the main terms or post-mortem studies; and (f) no full-text available. [(back pain) OR (lumbar vertebrae) OR (sciatica) OR (radiculopathy)] AND the specific terms of biomarkers [(inflamm*) OR (cytokines) OR (tumor necrosis factor) OR Data management and study selection (transforming growth factor–beta) OR (C-reactive protein) OR (cortisol) OR (hypothalamic pituitary adrenal)] found as In the first phase, search results from all databases were keywords in all fields, titles, or Medical Subject Headings exported to Mendeley. In the second phase, duplicate articles in (MesH). The search terms were selected according to a the databases were removed automatically by Mendeley and validation by a panel of experts and a review of the search manually by the first reviewer (JS-M). In the third phase, the file strategies used in previous systematic reviews on immune- was blind-screened in Rayyan (37) based on their titles and inflammatory biomarkers (28–30, 32) and HPA axis (22, 33, 34). abstracts by the first reviewer (JS-M) and the second reviewer Frontiers in Immunology 03 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 (MC-C). In the fourth phase, the reviewers exported the Data analyses references selected in Rayyan QCRI to a form developed by the research team and blind-screened the full texts to check All first and corresponding authors of selected articles with study eligibility. In the fifth phase, extraction of data from missing data for the synthesis of this systematic review were selected full text, risk of bias (RoB), and MQ of included contacted by email in two rounds. studies was carried out by the reviewers, being key information from each study entered into a standardized data extraction form. Results Selection and inclusion of studies Data extraction and study coding Figure 1 displays the flowchart of the study selection process. The first reviewer (JS-M) summarized all the relevant The initial search conducted in the five databases yielded up to information from the studies in a spreadsheet, whereas the 1,983 potentially relevant studies. In the first phase, after the second reviewer (MC-C) checked the extracted data in order to removal of duplicates, 1,204 titles and abstracts were screened, of avoid errors in information recording. The general information which 32 studies were selected for full-text revision with a 99% extracted from each study was authors, objective, type of research, preliminary consensus. We excluded studies not focused on year of publication, location, diagnostic criteria, duration of NSLBP (n = 578), not the type of targeted study (n = 304), not illness, clinical data, sample size in the study groups (NSLBP human studies (n = 182), not including a HC group (n = 71), and and HC), age, body mass index (BMI), gender (% of women), and not adults (n = 37). type of assay used to quantify biomarker levels. The biomarker In the second phase, after the full-text screening, 16 studies information extracted from each study was sample size (n), mean were excluded with an 87% of preliminary consensus. We (M), standard deviation (SD), effect size (using Hedges g), and excluded studies with participants with a different diagnosis statistical significance between groups (p-value). from NSLBP (n = 11), not including any immune-inflammatory biomarkers or cortisol levels (n = 4), and without a full-text available (n = 1). Finally, 16 articles were included in this Risk of bias and methodological quality systematic review, reporting on 14 unique studies. When studies derived from the same sample were identified [e.g., RoB was assessed using an adapted version of the National Wang et al. (19) and Wang et al. (20) and Klyne et al. (39), Heart, Lung, and Blood Institute tool (38). The total score was Klyne et al. (40), and Klyne et al. (41)], the one with the largest divided into low (8–10), medium (4–7), and high (≤3) RoB based sample was selected [(Wang et al. (19) and Klyne et al. (40)]. on a scale from −10 to 10. The answer options were scored with a There was no need for a third reviewer to resolve a disagreement 1 when the criterion was met, with a 0 when it was unclear or in any of the screenings. cannot be determined, and with −1 when it was not reported. MQ was assessed according to the Immune Confounders Scale (ICS; Appendix B) (32) adapted to also take into account Characteristics of the included studies potential biomarkers of the HPA axis. This checklist comprises two parts: (a) 10 questions for the assessment of the quality Table 1 shows the characteristics of the 14 included studies. biomarkers’ evidence, considering the most important analytical The sample size of the studies groups ranged from 11 to 556 in aspects (e.g., sample size, matched groups, and reporting NSLBP and 7 to 1,572 in HC, and the mean age ranged from 21 to detection limit), and (b) 20 questions for the consideration of 75 years old and 20 to 75 years old in NSLBP and HC, respectively. possible critical confounders that could induce heterogeneity in A total of 10 studies assessed individuals with chronic NSLBP, one the results of the studies analyzed. The first part, in which with acute NSLBP, and three with NSLBP (without stipulating the questions were scored with a 1 if evidence of compliance was duration). Illness duration was reported in five studies, ranging reported, is calculated on a scale from 0 to 10 with higher scores from 20 to 88 months, and BMI was reported in nine studies, indicating better MQ (0–4: low; 5–7: medium; and 8–10: high ranging from 19 to 48 kg/m2. The included studies were quality). The second part of the ICS is on a scale from 0 to 29, conducted in eight countries: Germany (n = 6), USA (n = 2), where 0 indicates the total control of all confounding variables Canada (n = 1), Australia (n = 1), Brazil (n = 1), Norway (n = 1), (e.g., matching study groups, meeting all exclusion criteria, or Israel (n = 1), and China (n = 1). Study designs included six quasi- statistically adjusting for background variables) and 29 indicates experimental, five case-control, two observational, and the total absence of control. one experimental. Frontiers in Immunology 04 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 FIGURE 1 Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flowchart from record identification to study inclusion. We found that 100% of the included studies indicated the established confounder variables. There was a 99% agreement in type of sample (plasma, serum, or saliva), 57% were free of pain the RoB assessment and 91% in the MQ assessment. A third medication, 36% were free of mood medication including reviewer was not needed for reaching a consensus. The RoB antidepressants and mood stabilizers, 50% had matched assessment is presented in Table 2, whereas the MQ assessment participants by age and gender, and 43% specified the time of is reported in Appendix C. extraction. Approximately 93% reported the diagnostic criteria for NSLBP, 36% assessed the pain intensity level, 29% calculated the severity of depressive symptomatology, and 29% measured Relationship between biomarker levels the back functional status. Regarding outcomes, approximately and NSLBP 43% assessed more than one immune-inflammatory biomarker and 14% assessed cortisol. Eighty-six percent of the studies We found 12 studies assessing immune-inflammatory reported the name of the manufacturer of the test kits used to biomarkers (n = 24) and two HPA biomarkers as assessed assay the biomarkers. The characteristics of the individual with cortisol (n = 2). Table 3 summarizes findings for all the studies are detailed in Table 1. biomarkers explored in this systematic review. Risk of bias assessment and Immune-inflammatory biomarkers methodological quality We found that growth differentiation factor 15 (GDF-15), IL-23, and soluble tumor necrosis factor receptor 1 (sTNF-R1) Applying the National Heart, Lung, and Blood Institute tool levels were significantly higher in individuals with NSLBP (38), we found that 29% of the studies presented low and 71% compared to HC. One study out of one (100%) for GDF-15 moderate RoB. Regarding the quality of the studies, 64% (50), IL-23 (45), and sTNF-R1 (46) showed significantly higher presented medium and 36% presented low MQ. Additionally, levels in NSLBP than HC, with strong (g = 0.87), moderate (g = 42% of the studies did not control for at least 10 of the 20 0.70), and low (g = 0.26) effect sizes, respectively. The RoB from Frontiers in Immunology 05 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 TABLE 1 Characteristics of the included studies (n = 14). Author Country Design Diagnosis n Age (SD or Female, % BMI (SD or Pain Assessed (year) Range) Range) duration, biomarkers months NSLBP HC NSLBP HC NSLBP HC NSLBP HC Immune-inflammatory biomarkers Gebhardt et al. Germany CC NSCLBP 41 1572 42.2 (27- NR 65.8 NR 27.7 (19- NR 24.3 hsCRP (2006) (42) 57) (20- 48) 64) Wang et al. Germany CC NSCLBP 120 120 46.6 45.4 43.3 43.3 26.1 (18- 27.1 NR TNF-a (2008) (43) (10.9) (11.4) 35) (19- 48) Wang et al. Germany CC NSCLBP 58 29 44.7 (24- 40.8 58.6 58.6 24.2 (18- 27.1 20.4 TNF-a (2010) (20) 68) (23- 33) (19- 66) 48) Roy et al. Canada QE NSCLBP 11 10 45.6 (8.9) 47.5 36.36 40.0 28 (3.7) 25.3 NR hsCRP, IL-6 (2010) (27) (16.2) (3.6) Heffner et al. USA CC NSCLBP 25 25 30.8 30.8 60.0 60.0 NR NR NR IL-6 (2011) (44) (11.4) (11.4) Luchting et al. Germany QE NSCLBP 37 25 44.5 (21- 43.0 62.2 52.0 NR NR 70.1 IL-6, IL-10, (2014) (45) 73) (24- IL-17, IL-23 54) Queiroz et al. Brazil QE NSLBP 71 71 71.4 71.5 100 100 30 (4.8) 27.5 NR sTNF-R1, IL-6 (2015) (46) (5.06) (4.87) (4.4) Luchting et al. Germany QE NSCLBP 19 19 47.0 (13) 40.0 79.0 58.0 23.9 (3.1) 23.6 71.8 IL-1b (2016) (47) (11) (2.9) Li et al. (2016) China QE NSCLBP 35 35 NR (45- NR NR NR NR NR NR IL-6, IL-10 (48) 75) (45- 75) Degenhardt USA EX NSLBP 33 7 37.7 32.0 72.7 71.0 25.4 (4.3) 24.1 NR IL-1b, IL-6, et al. (2017) (11.7) (9) (4.3) TNF-a, CRP (49) Klyne et al. Australia CC NSALBP 109 55 29.0 (8) 27.0 46.8 69.1 24.1 (3.7) 22.9 NR CRP, IL-6, (2018) (40) (6) (4.1) IL-1b, TNF-a Tarebeith et al. Israel OB NSLBP 556 522 46.3 39.6 54.64 54.64 29 (NR) 27 NR GDF-15 (2019) (50) (NR) (NR) (NR) HPA axis markers Muhtz et al. Germany QE NSCLBP 20 33 44.9 33.3 60 63.6 NR NR 87.7 Cortisol (2013) (24) (14.6) (12.0) Sveinsdottir Norway OB NSCLBP 305 845 44.0 46.0 54 50 NR NR NR Cortisol et al. (2015) (9.34) (9.7) (26) BMI, body mass index (kg/m2); CC, case control; EX, experimental; OB, observational; QE, quasi-experimental; HC, healthy controls; NSLBP, non-specific low back pain; NSSLBP, non- specific subacute low back pain; NSCLBP, non-specific chronic low back pain; NSALBP, non-specific acute low back pain; NR, not reported; CRP, C-reactive protein; GDF, growth differentiation factor; hsCRP, high-sensitivity CRP; IL, interleukin; TNF, tumor necrosis factor. these three studies was low or medium and the MQ was rated as reporting significant differences presented a high RoB and a low medium (45, 50) or high (46). Results regarding GDF-15 and MQ (48), whereas the study reporting non-significant sTNF-R1 were related to NSLBP without specifying pain differences showed a moderate RoB and a medium MQ (45). duration, whereas TGF-b and IL-23 results were regarding Although we found a trend of IL-6 high levels and TNF-a in chronic NSLBP. NSLBP compared to HC, only one of the seven studies (14%) Two studies examined the association between IL-10 levels reported significant differences in IL-6 (48) in a sample of 70 and chronic NSLBP (45, 48). One study (50%) reported participants (chronic NSLBP = 35, HC = 35) and one of the four significantly lower IL-10 levels in chronic NSLBP (48) in a studies (25%) reported significant differences in TNF-a (19) in a sample of 70 participants (chronic NSLBP = 35, HC = 35), sample of 87 participants (chronic NSLBP = 58, HC = 29), both whereas the other study indicated a trend toward higher IL-10 referring to chronic NSLBP. The RoB from these two studies was levels in chronic NSLPB as compared with HC in a sample of 62 high (48) and moderate (19), and the MQ was rated as low (48) participants (chronic NSLBP = 37, HC = 25) (45). The study and high (19), respectively. Although results of TNF-a and IL-6 Frontiers in Immunology 06 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 TABLE 2 Risk of bias assessment using an adapted version of the National Heart, Lung, and Blood Institute tool for included studies (n = 14). Study NIHLBI Score RoB Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Immune-inflammatory biomarkers Gebhardt et al. (2006) (42) 1 1 1 1 −1 1 1 1 1 −1 6 Moderate Wang et al. (2008) (43) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Wang et al. (2010) (19) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Roy et al. (2010) (27) 1 1 −1 1 0 1 1 1 1 −1 5 Moderate Heffner et al. (2011) (44) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Luchting et al. (2014) (45) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Queiroz et al. (2015) (46) 1 1 1 1 1 1 1 1 1 −1 8 Low Luchting et al. (2016) (47) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Li et al. (2016) (48) 1 0 −1 −1 −1 1 1 1 1 −1 1 High Degenhardt et al. (2017) (49) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Klyne et al. (2018) (40) 1 1 1 1 −1 1 1 1 1 1 8 Low Tarabeih et al. (2019) (50) 1 1 1 1 −1 1 1 1 1 −1 6 Moderate HPA axis markers Muhtz et al. (2013) (24) 1 1 −1 1 −1 1 1 1 1 −1 4 Moderate Sveinsdottir et al. (2015) (26) 1 1 1 1 −1 1 1 1 1 1 8 Low Q1: Was the research question or objective in this paper clearly stated? Q2: Was the study population clearly specified and defined? Q3: Was the participation rate of eligible persons at least 50%? Q4: Were all the subjects selected or recruited from the same or similar populations (including the same time period)? Were inclusion and exclusion criteria for being in the study prespecified and applied uniformly to all participants? Q5: Were a sample size justification, power description, or variance and effect estimates provided? Q6: For the analyses in this paper, were the exposure(s) of interest measured prior to the outcome(s) being measured? Q7: For exposures that can vary in amount or level, did the study examine different levels of the exposure as related to the outcome (e.g., categories of exposure or exposure measured as continuous variable)? Q8: Were the outcome measures (dependent variables) clearly defined, valid, reliable, and implemented consistently across all study participants? Q9: Were key potential confounding variables measured and adjusted statistically for their impact on the outcome(s)? Q10: Is the proportion of participants with missing data in the variable irrelevant (or is adequately justified to be irrelevant) or is it justified that statistical techniques to deal with missing data are appropriate (e.g., weighting adjustments or imputation methods)? NIHLBI, National Heart, Lung, and Blood Institute. Options: 1 = yes; 0 = unclear or cannot be determined; −1 = not reported. Total score range was 0–10: low risk of bias (8–10), medium (4–7), and high (≤3). cytokines have only one study in each case with significant HPA axis markers results, all studies assessing these biomarkers show a pattern of Concerning HPA axis function, two studies assessed cortisol increased levels in NSLBP compared to HC. levels using salivary samples (24, 26). Although both studies In addition, two studies out of two found higher levels of exhibit significant results, one study found higher levels (26) and CRP in NSLBP than in HC with no statistical consensus. Thus, the other one found lower levels (24) of cortisol in NSLBP only one of them (50%) that assessed participants with acute compared to HC. Both studies focused their results on chronic NSLBP found significant differences in a sample of 164 NSLBP, but their sample sizes varied from 1,150 (chronic participants (acute NSLBP = 109, HC = 55) with a low effect NSLBP = 305, HC = 845) (24) to 53 participants (chronic size (g = 0.33) (40), whereas the non-significant results were NSLBP = 20, HC = 33) (26), respectively. The RoB was low captured in NSLBP without specifying the pain duration in a (26) and moderate (24) and the MQ was low (24) and medium sample of 40 participants (NSLBP = 33, HC = 7) (49). In the first (26). To sum up, we found evidence of altered cortisol levels in case, the RoB was low and the MQ was high (40), whereas the NSLBP compared to HC. RoB was moderate and the MQ was medium in the study with Table 4 and Figure 2 summarize the findings on immune- non-significant differences (49). inflammatory and HPA axis biomarkers in NSLBP. We found no statistically significant differences for serum hsCRP in chronic NSLBP (27, 42), with one out of two studies not controlling for any confounder variables (27). Furthermore, Discussion we also found non-significant differences for IL-1b [chronic NSLBP (47, 49); acute NSLBP (40)], IL-17, and interferon To the best of our knowledge, this is a pioneering systematic gamma (IFN-g) (chronic NSLBP) (45) in comparison to HC review focusing on alterations in immune-inflammatory and with only one study specifying pain duration (46). The RoB in HPA axis biomarker levels in individuals with NSLBP compared these aforementioned studies was rated as low (40) or moderate to HC. Although previous reports showed that alterations in (27, 42, 45, 47, 49) and the MQ was rated as high (40), medium levels of IL-6 (28, 30), TNF-a (28–30), and CRP (28, 30) were (42, 45, 47, 49), or low (27). associated with NSLBP and acute NSLBP (29), the results of our Frontiers in Immunology 07 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 TABLE 3 Synthesis of all identified evidence. Biomarker Immune Pheno- Studies NSLBP HC Directionality NSLBP vs. HC Significant* differences NSLBP vs. HC type (n) (n) (n) (n, %) (n, %) IL-6 M1 7 321 228 ↑ Higher levels (7/7, 100%) (1/7, 14%) IL-10 Treg/TH2 2 72 60 ↑ Higher levels (1/2, 50%) (1/2, 50%) ↓ Lower levels (1/2, 50%) IL-17 TH17 1 37 25 ↑ Higher levels (1/1, 100%) (0/1, 0%) IL-23 TH17 1 37 25 ↑ Higher levels (1/1, 100%) (1/1, 100%) IL-1b M1 3 161 81 ↑ Higher levels (1/3, 33%) (0/3, 0%) = Equal levels (1/3, 33%) ↓ Lower levels (1/3, 33%) TNF-a M1 4 320 211 ↑ Higher levels (4/4, 100%) (1/4, 25%) sTNF-R1 M1 (CIRS at high 1 71 71 ↑ Higher levels (1/1, 100%) (1/1, 100%) levels) TGF-b Treg 1 37 25 ↑ Higher levels (1/1, 100%) (1/1, 100%) IFN-g TH 1 1 37 25 ↑ Higher levels (1/1, 100%) (0/1, 0%) GDF-15 Treg 1 556 522 ↑ Higher levels (1/1, 100%) (1/1, 100%) CRP M1/APP 2 142 62 ↑ Higher levels (2/2, 100%) (1/2, 50%) hsCRP M1/APP 2 52 1582 ↑ Higher levels (1/2, 50%) (0/2, 0%) ↓ Lower levels (1/2, 50%) Cortisol – 2 20 30 ↑ Higher levels (1/2, 50%) (2/2, 100%) ↓ Lower levels (1/2, 50%) *p ≤ 0.5. Inflammatory macrophage M1, T helper (TH)1, and TH17 cytokines, coupled with acute phase proteins (APP) reactants including C-reactive protein (CRP) and high-sensitivity (hs)CRP, are pro-inflammatory and may be conceptualized as the immune-inflammatory response system (IRS). T regulatory (Treg) and TH2 cytokines are anti-inflammatory and part of the compensatory immune-regulatory system (CIRS). Further explanation of the function of the immune phenotypes can be found in the works of André s-Rodrı́guez et al. (32) and Maes and Carvalho (13). systematic review are not completely in line with previous remains uncertain whether GDF-15 increased levels observed findings when using exclusively HC as a comparison group. in chronic NSLBP is a causative mediator or a risk biomarker. For the first time, preliminary and limited evidence of increased The pro-inflammatory cytokine IL-23 is mainly produced by levels of sTNF-R1 (46), GDF-15 (50), and IL-23 (45) was found dendritic cells and macrophages. IL-23 favors the synthesis of in NSLBP compared to HC, although this evidence was derived pro-inflammatory cytokines IL-17 and IL-22 by promoting the from only one study for each biomarker. expansion of TH17 cells (54). High circulating levels of IL-23 The soluble receptor sTNF-R1 binds pro-inflammatory have been found in chronic NSLBP compared to HC, but with a cytokine TNF-a. This biomarker plays the main role in diminished expression in T cells (45). In individuals with regulating TNF-a in vivo activity and its circulating levels are chronic pain, a decrease in TH17/Treg cells ratio and IL-23 increased in inflammatory diseases. At low or intermediate expression, as well as an increase in TGF-b in T cells, has been concentrations, the sTNF-R1 preserves and enhances the TNF- found. It is significant to note that, in these individuals, the TNF- a activity, whereas, at high levels, it inhibits the TNF-a activity a, IFN-g, and IL-4 circulating levels were undetectable (45). protecting from an excessive inflammatory response (51). It has These results point to an association between chronic NSLBP been described that TNF-a causes muscular catabolism (52) and and immune suppression more than an activation as it has been loss of bone by osteoclasts activation and inhibition that could be observed in other chronic diseases (55). The question is whether related to chronic pain. The high levels of sTNF-R1 found in this disturbed balance is (1) a consequence of the long-time pain chronic NSLBP participants (46) might play a compensatory inducing the dysregulation of the immune system to an anti- response role to previous high TNF-a levels (13). inflammatory phenotype or (2) the dysregulation existing The peptide hormone GDF-15 belongs to the TGF (TGF-b) previously and predisposing to suffer chronification of pain superfamily and is expressed in multiple tissues mainly after symptoms because treated individuals who improved in their tissue damage, apoptosis, inflammation, and oxidative stress. pain condition continue to show the same imbalance (45). Under normal physiological conditions, TGF-b expression is The results for sTNF-R1, GDF-15, TGF-b, and IL-23 should low, with a role in recovering and reorganizing muscle structure be interpreted with caution because only one study was available (53), whereas, under many inflammatory pathological for each biomarker. Nonetheless, findings from GDF-15 comes conditions, it is overexpressed due to its protective role. It from one study with 1,078 participants (50), whereas sTNF-R1, Frontiers in Immunology 08 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 TABLE 4 Summary of studies’ results on immune-inflammatory and HPA axis biomarkers (n = 14). Author (year) Diagnosis Mean pain n Biomarker, Mean (SD) NSLBP HC NSLBP HC p g IL-6 (n = 7) Roy et al. (2010) (27) NSCLBP NR 11 10 3.97 pg/ml (0.44) 3.12 pg/ml (0) .06 0.87 Heffner et al. (2011) (44) NSCLBP NR 25 25 1.2 pg/ml (1.0) 1.1 pg/ml (0.6) .67 0.12 Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 2.1 pg/ml (NR) 1.4 pg/ml (NR) >.05 NR Queiroz et al. (2015) (46) NSLBP 5.31 71 71 2.25 pg/ml (1.80) 1.63 pg/ml (3.67) .37 0.21 Li et al. (2016) (48) NSCLBP NR 35 35 170% level 100% level <.05 NA Degenhardt et al. (2017) (49) NSLBP 4a 33 7 0 pg/ml (11.30) 0 pg/ml (0.07) 09 0 Klyne et al. (2018) (40) NSALBP NR 109 55 0.8 pg/ml (1.58) 0.7 pg/ml (0.54) .24 0.07 IL-10 (n = 2) Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 4.1 pg/ml (NR) 3.6 pg/ml (NR) >.05 NR Li et al. (2016) (48) NSCLBP NR 35 35 75% level 100% level <.01 NR IL-17 (n = 1) Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 4.3 pg/ml (NR) 4.0 pg/ml (NR) >.05 NR IL-23 (n = 1) Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 1.21 pg/ml (0.43) 0.94 pg/ml (0.29) <.01 0.70 IL-1b (n = 3) Luchting et al. (2016) (47) NSCLBP 3.5 19 19 1. 9 NR/NR (NR) 1.8 NR/NR (NR) >.05 NR Degenhardt et al. (2017) (49) NSLBP 4a 33 7 0 pg/ml (0.03) 0 pg/ml (0.03) .28 0 Klyne et al. (2018) (40) NSALBP NR 109 55 0.2 pg/ml (0.24) 0.3 pg/ml (0.31) .20 0.37 TNF-a (n = 4) Wang et al. (2008) (43) NSCLBP 5.8 120 120 57.6% positive 12.3% positive >.05 NR Wang et al. (2010) (19) NSCLBP 5.19b 58 29 2.58 pg/ml (NR) 0.1 pg/ml (NR) <.01 NR Degenhardt et al. (2017) (49) NSLBP 4a 33 7 3.68 pg/ml (4.28) 0.60 pg/ml (2.92) .23 0.74 Klyne et al. (2018) (40) NSALBP NR 109 55 1.1 pg/ml (1.53) 1.0 pg/ml (0.76) .24 0.07 sTNF-R1 (n = 1) Queiroz et al. (2015) (46) NSLBP 5.31 71 71 1275 pg/ml (404.18) 1087 pg/ml (448.87) .02 0.26 TGF-b (n = 1) Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 0.21 (0.07) 0.14 (0.05) .01 1.10 IFN-g (n = 1) Luchting et al. (2014) (45) NSCLBP 3.37a 37 25 4.19 (3.54) 3.60 (2.20) >.05 0.20 GDF-15 (n = 1) Tarebeith et al. (2019) (50) NSLBP NR 556 522 2.65 NR/NR (0.08) 2.58 NR/NR (0.08) <.01 0.87 CRP (n = 2) Degenhardt et al. (2017) (49) NSLBP 4a 33 7 0.82 mg/ml (3.45) 0.65 mg/ml (6.96) .71 0.04 Klyne et al. (2018) (40) NSALBP NR 109 55 1.8 mg/ml (2.60) 1.0 mg/ml (1.83) <.01 0.33 hsCRP (n = 2) Gebhardt et al. (2006) (42) NSCLBP 4.9a 41 1572 1.3 ug/ml (1.39) 1.49 ug/ml (NR) >.05 NR Roy et al. (2010) (27) NSCLBP NR 11 10 2.50 ug/ml (0.79) 1.05 ug/ml (0.34) .11 0.75 Cortisol (n = 2) Muhtz et al. (2013) (24) NSCLBP 6.03a 20 33 2.25 mg/l (NR) 3.35 mg/l (NR) <.01 NR Sveinsdottir et al. (2015) (26) NSCLBP NR 305 845 7.63 nmol/l (7,26) 4.23 nmol/l (24,26) .02 NR HC, healthy controls; NSCLBP, non-specific chronic low back pain; NSALBP, non-specific acute low back pain; NR, not reported; NA, not applicable; CRP, C-reactive protein; hsCRP, high- sensitivity CRP; IL, interleukin; TNF, tumor necrosis factor; GDF, growth differentiation factor. a, Visual Analogue Scale (VAS), score 0-10. Bold, statistically significant differences compared to HC. Bold, statistically significant differences compared to HC. Frontiers in Immunology 09 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 with moderate RoB and medium (42) to low (27) MQ. Potential adaptive mechanisms underlying CRP/hsCRP differences can be explained by acute NSLBP and chronic NSLBP targets. In the same way, the current systematic review found inconsistent evidence for IL-10, IL-6, and TNF-a alterations in NSLBP. Specifically, IL-10 was significantly lower in one study (48) and non-significantly increased in another study (45) compared to HC. IL-10 is essential for the homeostasis of the immune system by playing an anti-inflammatory role effect inhibiting pro-inflammatory cytokines such as TNF-a, IL-6, and IL-1 (56, 57). However, as literature shows, low levels of IL-10 are presented in chronic widespread pain (57). This fact FIGURE 2 may point out a pro-inflammatory state that could promote IL, interleukin; IFN, interferon; TNF, tumor necrosis factor; TGF, transforming growth factor; GDF, growth differentiation factor; tissue inflammation. The scientific evidence indicated that IL-10 CRP, C-reactive protein; hsCRP, high-sensitivity CRP; NSCLBP, data should be interpreted with caution because studies assessing non-specific chronic low back pain; NSALBP, non-specific acute low back pain; NSLBP, non-specific low back pain; NR, not this biomarker showed high (48) to moderate (45) RoB and low reported; BMI, body mass index (kg/m2); HC, healthy controls. (48) to medium (45) MQ, although both studies were aiming for chronic NSLBP. Moreover, these studies recruited small samples, from 62 (45) to 70 (48) participants, which limits the interpretation of their findings and makes it difficult to make TGF-b, and IL-23 results have sample sizes ranging from 62 solid conclusions. [Luchting et al. (45) for TGF-b and IL-23] to 142 participants IL-6 and TNF-a pro-inflammatory cytokines are important [Queiroz et al. (46) for sTNF-R1]. The study that reported higher neuromodulators in central sensitization by the activation of sTNF-R1 levels in NSLBP (46) included only elderly women glial cells (astrocytes and microglia) within the spinal cord (from 65 to 82 years old), and therefore, these results could be inducing hyperalgesia and allodynia (58). Even though higher related to the aging process or to gender issues. Furthermore, the levels of IL-6 and TNF-a appeared to increase in NSLBP IL-23 and TGF-b findings appeared to be driven exclusively by compared to HC, solid conclusions cannot be established. chronic NSLBP (45), whereas findings related to GDF-15 and Only one out of seven studies reported statistically significant sTNF-R1 were pointing out NSLBP. The RoB of the three latter differences in IL-6 (48) and only one out of four reports observed studies was moderate (45, 50) or low (46) and the MQ was statistically significant differences in TNF-a (19) regarding medium (45, 50) and high (46). chronic NSLBP in both findings. The RoB from these two CRP alterations (28, 30) in NSLBP and in acute NSLBP (29) studies was high (48) and moderate (19) and the MQ low (48) were partially replicated in the present work. Although two and high (19), respectively. In contrast to previous reviews (28– studies found an absolute elevation of CRP levels in NSLBP 30), we conclude that there is inconclusive evidence for compared to HC, only one study reported statistically significant significantly higher levels of IL-6 and TNF-a in NSLBP differences with a low effect size in acute NSLBP (40). This latter compared to HC. The low presence of significant results in study yielded low RoB and high MQ (40) with a total of 70 our review compared to previous reviews may be explained by participants, whereas the non-significant study (49) showed our use of HC as a comparator group as well as by target moderate RoB and medium MQ. The small sample size of the condition heterogeneity. For example, four IL-6 studies were last study (n = 40) and its target population (chronic NSLBP) conducted including chronic NSLBP in which one reported may explain these results discrepancies through unpowerful significant evidence, one included NSLBP participants, and statistical results without significant conclusions. Previous another used acute NSLBP. In TNF-a biomarker studies, two systematic research clarifies how high levels of pain intensity studies focused on chronic NSLBP with one reporting significant are correlated with CRP increased levels (28, 30) in NSLBP and results, whereas one focused on NSLBP and another on acute NSLBP (29), whereas the present study emphasized how acute NSLBP. the target population (i.e., acute NSLBP) is the key detonator to In agreement with previous reviews, we found no evidence exhibit significant results in a worsening dysregulation of CRP for alterations in IL-1b, IL-17, and IFN-g. Of these, IL-1b level expression. In congruence with significant results in NSLBP showed most clearly non-significant results, with three out of (30), we did not find evidence for alterations in hsCRP, with two three studies reporting non-significance for serum IL-1b (40, 47, out of two studies showing non-significant results for serum 49). The three studies assessing IL-1b targeted chronic NSLBP hsCRP (27, 42). Studies reporting hsCRP results were congruent without specifying pain duration (47, 49) and acute NSLBP (40). with the condition (i.e., chronic NSLBP) and they were Regarding HPA functioning, the two identified studies found composed of sample sizes of 1,613 and 21 participants (27, 42) significantly altered levels of salivary cortisol in chronic NSLBP Frontiers in Immunology 10 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 compared to HC, with low (26) or moderate RoB (24) and on the biomarkers analyzed, including analgesics, NSAIDs, medium (26) or low MQ (24). The sample sizes of these studies corticosteroids, and opioids, but also of antidepressants and varied from 53 to 1,150 participants. As there are age-dependent mood stabilizers needs to be taken into account, given the high changes in HPA axis function, both studies controlled age as a comorbidity of pain disorders with mood disorders such as potentially confounding variable in the analyses. However, both depression and anxiety (59). The strategies usually employed to studies found significant differences: the larger study (26) address this problem involve a medication washout or statistical supports the fact that cortisol levels are increased in NSLBP, control. In our review, we have assessed these aspects of the MQ whereas Muhtz et al. (24) suggest decreased levels in chronic of the included studies using the ICS scale (Appendices B and C). NSLBP. Agreeing with Juruena et al. (22), being exposed to a Finally, only published studies in English or Spanish were pain condition (constant or not) is a potential stressor that included in this study, so other evidence could have been activates HPA functioning through cortisol levels, but being omitted. Nevertheless, we included additional search sources exposed to a chronic pain condition yields to dysfunction in such as the review of references from other studies, review of cortisol response that can endanger the HPA system. Further references from other systematic reviews, and gray literature. research is needed to target the direction of these The strengths of this study are the number of databases neuroendocrine mechanisms and their relationship with NSLBP. explored, compliance with PRISMA guidelines, prospective The identification of signature immune-inflammatory and registration in PROSPERO, preliminary review and validation HPA axis biomarkers in NSLBP could provide new targets for of the Boolean searches according to PRESS guidelines, the pharmacological therapy. Furthermore, to the extent that inclusion of Rayyan as a tool to minimize possible loss of variations in the levels of these biomarkers could be related to evidence, and consensual review between two reviewers in the the response of patients to the different treatments different phases of screening and extraction of the data. In (pharmacological and/or psychological) available, they could addition, the inclusion of RoB and MQ tools, consideration of provide clinicians with objective measures to guide therapeutic an exclusive healthy comparison group, and selection of baseline choices along the lines of personalized medicine for this highly data not previously reported in other systematic reviews treatment-resistant condition. contribute to generate an additional value to the limited scope of our findings. Longitudinal studies are required to establish cause and effects to evaluate the predictive value of Limitations and strengths these biomarkers. Our findings must be interpreted taking the following limitations and strengths into account. First, the small amount Conclusions of evidence found for most explored biomarkers as well as the limited sample sizes of some studies made it difficult to report In contrast to previous systematic reviews, our results strong conclusions about the immune-inflammatory pathways indicate that there is not enough evidence to conclude that underpinning NSLBP. Second, the moderate RoB (about 71%) CRP, IL-6, and TNF-a levels are altered in NSLBP in and the medium (50%) to low MQ (about 21%) of the studies comparison to HC. This study showed preliminary and limited included in this systematic review could lead to imprecise evidence (only one study for each biomarker) that GDF-15, IL- conclusions. Third, results of our heterogeneity analysis 23, and sTNF-R1 are significantly increased in NSLBP as well as indicated that a considerable number of studies (about 45%) TGF-b in chronic NSLBP compared to HC, altogether did not control at least 10 of the 20 established confounding suggesting a pro-inflammatory state in NSLBP. Regarding the variables. All of these 20 potential confounder variables are anti-inflammatory cytokine IL-10, lower levels, although with depicted in the ICS (Appendix B). Fourth, sample source variety inconsistent evidence, were found in chronic NSLBP. can yield to provide misleading results: nine studies used serum, Preliminary evidence for IL-17 and IFN-g in chronic NSLBP two used plasma, one used serum and plasma, and two studies was not found. IL-1b and hsCRP showed non-significant used salivary cortisol. Fifth, given the characteristics of the associations with NSLBP. Pro-inflammatory biomarkers are studies, it was impossible to perform a complementary altered due to the pain exposure, whereas an anti- analysis between biomarkers and the NSLBP type (acute, inflammatory response is presented as a consequence of the subacute, and chronic), as well as taking into account clinical inflammation process in chronic pain situations. Regarding variables that have been associated with pain in previous studies cortisol, one study showed an increase and the other a (e.g., pain intensity, depression, sleep quality, smoking, and decrease in salivary levels in NSLBP compared to HC. medications). Medication, in particular, is a confounding However, with the incongruent evidence regarding the variable that is difficult to disentangle from the pain disorder directionality of cortisol response, both studies indicate a itself in this type of research. Possible intake not only of pain dysregulation of HPA functioning due to constant pain medication, in particular medications that have a clear impact exposure. To sum up, 10 out of 14 studies reported their Frontiers in Immunology 11 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 findings on chronic NSLBP, so the current results could be (FI_B/00216). AF-S acknowledges the funding from the Serra partially relaying under chronic condition. Overall, the number Hú nter program (Generalitat de Catalunya; reference number of available studies, RoB, and MQ limited the interpretation of UAB-LE-8015). The funding sources and sponsors had no these results. Consequently, more scientific evidence is needed to influence on the design of the study, data collection and confirm and quantify the magnitude of the findings reported in analysis, or the writing of the manuscript. the current systematic review, as well as to better understand the role of these specific biomarkers under NSLBP condition. Acknowledgments Data availability statement We thank Dr. Alessandro Chiarotto for his valuable comments on a previous draft of this manuscript. The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors. Conflict of interest The authors declare that the research was conducted in the Author contributions absence of any commercial or financial relationships that could be construed as a potential conflict of interest. JL, AS, SE, AF-S, and XB designed the study. JS-M, and MC- C performed the eligibility criteria, data extraction, and study coding. JS-M, and AC-C performed the data analysis, synthesized all extracted data, and drafted the manuscript. MG Publisher’s note and ME realized a value contribution specifically in the introduction, discussion, and conclusion section regarding All claims expressed in this article are solely those of the biological point of view as well as revised and approved the authors and do not necessarily represent those of their affiliated final version of the manuscript. MM critically revised and organizations, or those of the publisher, the editors and the supervised the final draft. All authors commented on, revised, reviewers. Any product that may be evaluated in this article, or and approved the draft and the final manuscript. claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Funding Supplementary material This work was supported by the Institute of Health Carlos III (ISCIII; PI19/00112) cofinanced with European Union ERDF The Supplementary Material for this article can be found funds. JS-M has a PFIS predoctoral contract from the ISCIII online at: https://www.frontiersin.org/articles/10.3389/ (FI20/00034). AC-C has a FI predoctoral contract from AGAUR fimmu.2022.945513/full#supplementary-material References 1. Savigny P, Watson P, Underwood M. Early management of persistent non- 6. Golob AL, Wipf JE. Low back pain. Med Clin North Am (2014) 98:405–28. specific low back pain: Summary of NICE guidance. BMJ (2009) 338:b1805. doi: 10.1016/j.mcna.2014.01.003 doi: 10.1136/bmj.b1805 7. Maher C, Underwood M, Buchbinder R. Non-specific low back pain. Lancet 2. Vlaeyen JWS, Maher CG, Wiech K, Van Zudert J, Beraldo Meloto C, (2016) 389:736–47. doi: 10.1016/S0140-6736(16)30970-9 Diatchenko L, et al. Low back pain (primer). Nat Rev Dis Primers (2018) 4:52. doi: 10.1038/s41572-018-0052-1 8. Wynne-Jones G, Cowen J, Jordan JL, Uthman O, Main CJ, Glozier N, et al. 3. van Tulder M, Becker A, Bekkering T, Breen A, Gil del Real MT, Hutchinson Absence from work and return to work in people with back pain: A systematic A, et al. Chapter 3 European guidelines for the management of acute nonspecific review and meta-analysis. Occup Environ Med (2014) 71:448–56. doi: 10.1136/ low back pain in primary care. Eur Spine J (2006) 15:169–91. doi: 10.1007/s00586- oemed-2013-101571 006-1071-2 9. Taylor JB, Goode AP, George SZ, Cook CE. Incidence and risk factors for 4. World Health Organization (WHO). Musculoskeletal health (2022). Available first-time incident low back pain: a systematic review and meta-analysis. Spine J at: https://www.who.int/news-room/fact-sheets/detail/musculoskeletal-conditions. (2014) 14:2299–319. doi: 10.1016/j.spinee.2014.01.026 5. Hoy D, March L, Brooks P, Blyth F, Woolf A, Bain C, et al. The global burden 10. Sanzarello I, Merlini L, Rosa MA, Perrone M, Frugiuele J, Borghi R, et al. of low back pain: Estimates from the global burden of disease 2010 study. Ann Central sensitization in chronic low back pain: A narrative review. J Back Rheum Dis (2014) 73:968–74. doi: 10.1136/annrheumdis-2013-204428 Musculoskelet Rehabil (2016) 29:625–33. doi: 10.3233/BMR-160685 Frontiers in Immunology 12 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 11. Hannibal KE, Bishop MD. Chronic stress, cortisol dysfunction, and pain: A systematic review, meta-analysis and meta-regression. Brain Behav Immun (2019) psychoneuroendocrine rationale for stress management in pain rehabilitation. Phys 87:881–9. doi: 10.1016/j.bbi.2019.12.020 Ther (2014) 94(12):1816–25. doi: 10.2522/ptj.20130597 33. Murri MB, Pariante C, Mondelli V, Masotti M, Atti AR, Mellacqua Z, et al. 12. Zhong J, Shi G. Editorial: Regulation of inflammation in chronic disease. HPA axis and aging in depression: Systematic review and meta-analysis. Front Immun (2019) 10:737. doi: 10.3389/fimmu.2019.00737 Psychoneuroendocrinology (2014) 41:46–62. doi: 10.1016/j.psyneuen.2013.12.004 13. Maes M, Carvalho AF. The compensatory immune-regulatory reflex system 34. Villafañe JH, Pedersini P, Bertozzi L, Drago L, Fernandez-Carnero J, Bishop (CIRS) in depression and bipolar disorder. Mol Neurobiol (2018) 55(12):8885–903. MD. Exploring the relationship between chronic pain and cortisol levels in subjects doi: 10.1007/s12035-018-1016-x with osteoarthritis: Results from a systematic review of the literature. Osteoarthr Cartil (2020) 28:572–80. doi: 10.1016/j.joca.2020.02.836 14. Kawasaki Y, Zhang L, Cheng JK, Ji RR. Cytokine mechanisms of central sensitization: Distinct and overlapping role of interleukin-1b, interleukin-6, and 35. McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. tumor necrosis factor-alpha in regulating synaptic and neuronal activity in the PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin superficial spinal cord. J Neurosci (2008) 28:5189–94. doi: 10.1523/ Epidemiol (2016) 75:40–6. doi: 10.1016/j.jclinepi.2016.01.021 JNEUROSCI.3338-07.2008 36. Perestelo-Pé rez L. Standards on how to develop and report systematic 15. Sommer C, Kress M. Recent findings on how proinflammatory cytokines reviews in psychology and health. Int J Clin Health Psychol (2013) 13:49–57. cause pain: peripheral mechanisms in inflammatory and neuropathic hyperalgesia. doi: 10.1016/S1697-2600(13)70007-3 Neurosci Lett (2004) 361:184–7. doi: 10.1016/j.neulet.2003.12.007 37. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan–a web and 16. Liu YZ, Wang YX, Jiang CL. Inflammation: The common pathway of stress- mobile app for systematic reviews. Syst Rev (2016) 5:210. doi: 10.1186/s13643-016- related diseases. Front Hum Neurosci (2017) 11:316. doi: 10.3389/ 0384-4 fnhum.2017.00316 38. National Heart Lung and Blood Institute. Study quality assessment tools (2018). Available at: http://www.nhlbi.nih.gov/health-topics/study-quality- 17. Kelly GA, Blake C, Power CK, O’Keeffe D, Fullen BM. The association assessment-tools. between chronic low back pain and sleep: A systematic review. Clin J Pain (2011) 27:169–81. doi: 10.1097/AJP.0b013e3181f3bdd5 39. Klyne DM, Barbe MF, Hodges PW. Systemic inflammatory profiles and their relationships with demographic, behavioural and clinical features in acute low 18. Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The back pain. Brain Behav Immun (2017) 60:84–92. doi: 10.1016/j.bbi.2016.10.003 association between obesity and low back pain: A meta-analysis. AJE (2010) 171:135–54. doi: 10.1093/aje/kwp356 40. Klyne DM, Barbe MF, van den Hoorn W, Hodges PW. ISSLS PRIZE IN CLINICAL SCIENCE 2018: longitudinal analysis of inflammatory, psychological, 19. Wang H, Ahrens C, Rief W, Gantz S, Schiltenwolf M, Richter W. Influence and sleep-related factors following an acute low back pain episode–the good, the of depression symptoms on serum tumor necrosis factor-alpha of patients with bad, and the ugly. Eur Spine J (2018) 27:763–77. doi: 10.1007/s00586-018-5490-7 chronic low back pain. Arthritis Res Ther (2010) 12:1–7. doi: 10.1186/ar3156 41. Klyne DM, Hodges PW. Circulating adipokines in predicting the transition 20. Wang H, Ahrens C, Rief W, Schiltenwolf M. Influence of comorbidity with from acute to persistent low back pain. Pain Med (2020) 21(11):2975–85. depression on interdisciplinary therapy: Outcomes in patients with chronic low doi: 10.1093/pm/pnaa052 back pain. Arthritis Res Ther (2010) 12:R185. doi: 10.1186/ar3155 42. Gebhardt K, Brenner H, Stürmer T, Raum E, Richter W, Schiltenwolf M. 21. Griep EN, Boersma JW, Lentjes EG, Prins AP, van der Korst JK, de Kloet The course of high-sensitive c-reactive protein in correlation with pain and clinical ER. Function of the hypothalamic-pituitary-adrenal axis in patients with function in patients with acute lumbosciatic pain and chronic low back pain–a 6 fibromyalgia and low back pain. J Rheumatol (1998) 25(7):1374–81. months prospective longitudinal study. Eur J Pain (2006) 10:711–9. doi: 10.1016/ 22. Juruena MF, Bocharova M, Agustini B, Young AH. Atypical depression and j.ejpain.2005.11.005 non-atypical depression: Is HPA axis function a biomarker? a systematic review. J 43. Wang H, Schiltenwolf M, Buchner M. The role of TNF-a in patients with Affect Disord (2018) 233:45–67. doi: 10.1016/j.jad.2017.09.052 chronic low back pain–a prospective comparative longitudinal study. Clin J Pain 23. Cohen S, Janicki-Deverts D, Doyle WJ, Turner RB. Chronic stress, (2008) 24:273–8. doi: 10.1097/AJP.0b013e31816111d3 glucocorticoid receptor resistance, inflammation, and disease risk. Proc Natl 44. Heffner KL, France CR, Trost Z, Ng HM, Pigeon WR. Chronic low back Acad Sci (2012) 109(16):5995–9. doi: 10.1073/pnas.1118355109 pain, sleep disturbance, and interleukin-6. Clin J Pain (2011) 27:35. doi: 10.1097/ 24. Muhtz C, Rodriguez-Raeckeç R, Hinkelmann K, Moeller-Bertram T, Kiefer ajp.0b013e3181eef761 F, Wiedemann K, et al. Cortisol response to experimental pain in patients with 45. Luchting B, Rachinger-Adam B, Zeitler J, Egenberger L, Möhnle P, Kreth S, chronic low back pain and patients with major depression. Pain Med (2013) et al. Disrupted TH17/Treg balance in patients with chronic low back pain. PloS 14:498–503. doi: 10.1111/j.1526-4637.2012.01514.x One (2014) 9:e104883. doi: 10.1371/journal.pone.0104883 25. Sudhaus S, Fricke B, Schneider S, Stachon A, Klein H, von Düring M, et al. 46. Queiroz BZ, Pereira DS, De Britto NM, Lopes RA, Felı́cio DC, Pereira DG, The cortisol awakening response in patients with acute and chronic low back pain. et al. Functional performance and plasma cytokine levels in elderly women with relations with psychological risk factors of pain chronicity. Schmerz (2007) 21:202– and without low back pain. J Back Musculoskelet Rehabil (2015) 28:343–9. 4. doi: 10.1007/s00482-006-0521-4 doi: 10.3233/BMR-140526 26. Sveinsdottir V, Eriksen HR, Ursin H, Hansen AM, Harris A. Cortisol, 47. Luchting B, Heyn J, Woehrle T, Rachinger-Adam B, Kreth S, Hinske LC, health, and coping in patients with nonspecific low back pain. Appl Psychophysiol et al. Differential expression of P2X7 receptor and IL-1b in nociceptive and Biofeedback (2016) 41:9–16. doi: 10.1007/s10484-015-9300-2 neuropathic pain. J Neuroinflamm (2016) 13:100. doi: 10.1186/s12974-016-0565-z 27. Roy RA, Boucher JP, Comtois AS. Inflammatory response following a short- 48. Li Y, Liu J, Liu Z, Duan DP. Inflammation in low back pain may be detected term course of chiropractic treatment in subjects with and without chronic low from the peripheral blood: Suggestions for biomarker. Biosci Rep (2016) 36:e00361. back pain. J Chiropr Med (2010) 9:107–14. doi: 10.1016/j.jcm.2010.06.002 doi: 10.1042/BSR20160187 28. van den Berg R, Jongbloed EM, de Schepper EIT, Bierma-Zeinstra SMA, 49. Degenhardt BF, Johnson JC, Fossum C, Andicochea C, Stuart MK. Changes Koes BW, Luijsterburg PAJ. The association between pro-inflammatory in cytokines, sensory tests, and self-reported pain levels after manual treatment of biomarkers and nonspecific low back pain: A systematic review. Spine J (2018) low back pain. Clin Spine Surg (2017) 30:E690–701. doi: 10.1097/ 18:2140–51. doi: 10.1016/j.spinee.2018.06.349 BSD.0000000000000231 29. Morris P, Ali K, Merritt M, Pelletier J, Macedo LG. A systematic review of 50. Tarabeih N, Shalata A, Trofimov S, Kalinkovich A, Livshits G. Growth and the role of inflammatory biomarkers in acute, subacute and chronic non-specific differentiation factor 15 is a biomarker for low back pain-associated disability. low back pain. BMC Musculoskelet Disord (2020) 21:1–12. doi: 10.1186/s12891- Cytokine (2019) 117:8–14. doi: 10.1016/j.cyto.2019.01.011 020-3154-3 51. Grzanka R, Damasiewicz-Bodzek A, Kasperska-Zajac A. Tumor necrosis 30. Lim YZ, Wang Y, Cicuttini FM, Hughes HJ, Chou L, Urquhart DM, et al. factor-alpha and Fas/Fas ligand signaling pathways in chronic spontaneous Association between inflammatory biomarkers and non-specific low back pain: A urticaria. Allergy Asthma Clin Immunol (2019) 15(1):1–9. doi: 10.1186/s13223- s y s t e m a t ic r e v ie w . C l i n J P a i n ( 20 2 0 ) 3 6 :3 7 8– 8 9. d o i: 1 0 .1 0 9 7/ 019-0332-7 AJP.0000000000000810 52. Cesari M, Penninx BW, Pahor M, Lauretani F, Corsi AM, Williams GR, 31. Shamseer L, Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. et al. Inflammatory markers and physical performance in older persons: the Preferred reporting items for systematic review and meta-analysis protocols InCHIANTI study. J Gerontol A Biol Sci Med Sci (2004) 59(3):M242–8. (PRISMA-p) 2015: Elaboration and explanation. Br Med J (2015) 349:g7647. doi: 10.1093/gerona/59.3.m242 doi: 10.1136/bmj.g7647 53. Assadi A, Zahabi A, Hart RA. GDF15, an update of the physiological and 32. André s-Rodrı́guez L, Borràs X, Feliu-Soler A, Pé rez-Aranda A, Angarita- pathological roles it plays: A review. Pflügers Archiv (2020) 472:1535–46. Osorio N, Moreno-Peral P, et al. Peripheral immune aberrations in fibromyalgia: A doi: 10.1007/s00424-020-02459-1 Frontiers in Immunology 13 frontiersin.org Sanabria-Mazo et al. 10.3389/fimmu.2022.945513 54. Li H, Tsokos GC. IL-23/IL-17 axis in inflammatory rheumatic diseases. Clin 59. Walker AK, Kavelaars A, Heijnen CJ, Dantzer R. Neuroinflammation and Rev Allergy Immunol (2020) 60:31–45. doi: 10.1007/s12016-020-08823-4 comorbidity of pain and depression. Pharmacol Rev (2014) 66(1):80–101. 55. Hong M, Zheng J, Ding Z, Chen J, Yu L, Niu Y, et al. Imbalance between doi: 10.1124/pr.113.008144 Th17 and treg cells may play an important role in the development of chronic unpredictable mild stress-induced depression in mice. Neuroimmunomodulation COPYRIGHT (2013) 20(1):39–50. doi: 10.1159/000343100 © 2022 Sanabria-Mazo, Colomer-Carbonell, Carmona-Cervelló , Feliu- 56. Sanjabi S, Zenewicz LA, Kamanaka M, Flavell RA. Anti-inflammatory and Soler, Borràs, Grasa, Esteve, Maes, Edo, Sanz and Luciano. This is an open- pro-inflammatory roles of TGF-b, IL-10, and IL-22 in immunity and access article distributed under the terms of the Creative Commons autoimmunity. Curr Opin Pharmacol (2009) 9(4):447–53. doi: 10.1016/ Attribution License (CC BY). The use, distribution or reproduction in j.coph.2009.04.008 other forums is permitted, provided the original author(s) and the 57. Zhang JM, An J. Cytokines, inflammation, and pain. Int Anesthesiol Clin copyright owner(s) are credited and that the original publication in this (2007) 45:27–37. doi: 10.1097/AIA.0b013e318034194e journal is cited, in accordance with accepted academic practice. No 58. Watkins LR, Maier SF. When good pain turns bad. Curr Dir Psychol Sci use, distribution or reproduction is permitted which does not comply (2003) 12:232–6. doi: 10.1046/j.0963-7214.2003.01268.x with these terms. Frontiers in Immunology 14 frontiersin.org