The association between inflammatory biomarkers and low back disorder: a systematic review and meta-analysis

Figures & data

Figure 1. Flowchart showing the procedure and results of the literature search in accordance with the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines.

Table 1. Demographic information, list of inflammatory markers and clinical outcomes of the cohort studies included for meta-analysis.

Author	Country	Year	Sex (M/F)	Age	Diagnosis	Inflammatory markers	Low back clinical outcomes		Leg pain, (type)
							Pain score, (type)	Disability score, (type)
Al-Rawaf et al.	Saudi Arabia	2021	20/15	51.8	Moderate LBP	hsCRP, IL-6, TNFa	5.4 (NRS)	34.3 (ODI)	–
			19/16	49.7	Severe LBP		10 (NRS)	61.5 (ODI)	–
Caposella S. et al.	Switzerland	2018	9/14	52.5	Chronic LBP	IL-2, IL-4, IL-6, IL-10, GM-CSF, MCP-1, TNFa, CXCL6, IL-1 beta, CXCL12	7.7 (NRS)	–	–
Cheng L. et al.	PRC	2013	10/5	45	Non-ruptured LDH	IL-17, Th17	6.33 (VAS 10cm)	–	–
			8/7	43	Ruptured LDH		8.32 (VAS 10cm)	–	–
Cuellar J. M., et al.	USA	2010	13/9	44	Chronic LBP	IL-2, GM-CSF	7.4 (VAS 10cm)	–	–
De Queiroz B. Z., et al.	Brazil	2017	0/155	70.7	Acute LBP	IL-6, TNFa, IL-1	6.78 (NRS)	13.81 (RMDQ)	–
Gjefsen E. et al.	Norway	2021	14/32	42.1	Modic change 1 with LBP	IL-1b, IL-2, IL-4, Il-6, IL-8, IL-10, TNFa, GM-CSF, CXCL (6, 12)	6.4 (NRS)	31.3 (ODI)	–
			18/19	45.8	Modic change 2 with LBP		6.2 (NRS)	33.9 (ODI)	–
Hider S.L. et al.	UK	2019	37/56	52.6	Sciatica with LBP	TNFa, IL-1b, IL-6, MCP-1	6.7 (NRS)	12.1 (RMDQ)	6.2 (NRS)
Kraychete et al.	Brazil	2010	12/11	42.8	Herniated disc	IL-1b, IL-6, TNFa	9.0 (NRS)	–	–
Ohtori S. et al.	Japan	2011	30	63	Lumbar spinal stenosis	TNFa, IL-6	3.2 (VAS 10cm)	–	0.8 (VAS 10cm)
Rannou et al.	France	2007	24/12	52	Chronic LBP	hsCRP	57 (VAS 100mm)	48 (Quebec disability score)	–
Surbakti K.P., Nasution I.	Indonesia	2020	8/42	46.72	Non-specific LBP	hsCRP, IL-1, IL-6	49.3 (VAS 100mm)	–	–
Teodorczyk-Injean J.A. et al.	Canada	2019	13/9	32.8	Acute LBP	TNFa, IL-1b, IL-2, IL-10, IL-6	6.1 (VAS 10cm)	37 (ODI)	–
			14/11	36.5	Chronic LBP		5.2 (VAS 10cm)	28.1 (ODI)	–

Note: M: male, F: female, LBP: low back pain, hsCRP: high sensitivity C-reactive protein, IL: interleukin, TNFa: tumour nuclear factor alpha, GM-CSF: granulocyte macrophage colony stimulating factor, MCP: monocyte chemoattractant protein, CXCL: chemokine ligand, NRS: numeric pain rating scale, VAS: visual analogue scale, ODI: oswestry disability index, RMDQ: roland morris disability questionnaire, PRC: People’s Republic of China, USA: United States of America, UK: United Kingdom.

Table 2. Demographic information, list of inflammatory markers and clinical outcomes of the treatment studies included for meta-analysis.

Author	Country	Year	Sex (M/F)	Age	Diagnosis	Inflammatory markers	Pre-treatment low back clinical outcomes		Leg pain, (type)	Treatment	Post-treatment low back clinical outcomes		Post-op leg pain, (type)
							Pain score, (type)	Disability score, (type)			Pain score, (type)	Disability score, (type)
Degenhardt B.F., et al.	USA	2017	2/11	39	Chronic LBP	IL-1b, IL-6, TNFa, hsCRP	4.83 (NRS)	–	–	Manual treatment	3.33 (NRS)	–	–
			4/9	35			3.67 (NRS)	–	–	Sham ultrasound	2.5 (NRS)	–	–
			3/4	39			4 (NRS)	–	–	No treatment	3.5 (NRS)	–	–
Elkan P., et al.	Sweden	2016	87/90	40.1	LDH	hsCRP	55 (VAS 100mm)	52 (ODI)	65 (VAS 100mm)	Surgery	15 (VAS 100mm)	27 (ODI)	10 (VAS 100mm)
France R., Urban B,	USA	1990	10/18	45.68	Chronic LBP	Beta-endorphin	71 (VAS 100mm)	32.7 (mcgill)	–	3 weeks inpatient pain management	38.7 (VAS 100mm)	26.5 (mcgill)	–
Gebhardt et al.	Germany	2006	14/27	42.2	Chronic LBP	hsCRP	4.9 (VAS 10cm)	–	–	3 weeks conservative pain treatment	2.2 (VAS 10cm)	–	–
			15/16	44.8	Acute lumbo sciatic pain		–	–	5.5 (VAS 10cm)		–	–	1.8 (VAS 10cm)
Koivisto K. et al.	Finland	2019	15/5	49.2	Non-specific LBP	CTX-1, MCP-1, IL-6, IL-8, TNFa, hsCRP	6.6 (VAS 10cm)	30.1 (ODI)	3.0 (VAS 10cm)	IV zoledronic acid	–	–	–
			11/9	51.5			6.8 (VAS 10cm)	34.9 (ODI)	2.9 (VAS 10cm)	placebo	–	–	–
Park CH., Lee CH.	South Korea	2011	55/0	62.1	Foraminal stenosis	hsCRP	8.0 (VAS 10cm)	–	–	Transforaminal epidural steroid injection	2.5 (VAS 10cm)	–	–
Wei-Chun L. et al.	USA	2015	10/20	61	Chronic LBP	Beta-endorphin, IL-1b, IL-2, IL-6, TNFa, IL-4, IL-10	6.31 (BPI-sf)	7.43 (RMDQ)	–	Auricular point acupressure	2.66 (BPI-sf)	4.77 (RMDQ)	–
			10/21	66	Chronic LBP		6.07 (BPI-sf)	9.43 (RMDQ)	–	Sham APA	5.28 (BPI-sf)	9 (RMDQ)	–
Yeun J. et al.	Hong Kong	2017	4/8	67.7	Chronic LBP	TNFa	53.9 (VAS 100mm)	5.6 (RMDQ)	–	Concurrent Gua Sha and hot pack	40.3 (VAS 100mm)	3.5 (RMDQ)	–

Note: M: male, F: female, LBP: low back pain, hsCRP: high sensitivity C-reactive protein, IL: interleukin, TNFa: tumour nuclear factor alpha, GM-CSF: granulocyte macrophage colony stimulating factor, MCP: monocyte chemoattractant protein, CXCL: chemokine ligand, CTX1: Type I collagen cross linked c-telopeptide, NRS: numeric pain rating scale, VAS: visual analogue scale, ODI: oswestry disability index, RMDQ: roland morris disability questionnaire. BPI-sf: brief pain inventory – short form, USA: United States of America.

Figure 2. Assessment of the methodological quality of randomized controlled trials according to the Cochrane collaboration’s tool for assessing risk of bias.

Table 3. Assessment Of the methodological quality of observational studies according to the Newcastle-Ottawa scale (NOS).

Risk bias observational (Newcastle Ottawa)
Author	Year	Selection	Comparability	Outcome	Total
Al-Rawaf H. A. et al.	2021	4	2	3	9
Caposella S. et al.	2018	4	2	2	8
Cheng L et al.	2013	4	1	2	7
Cuellar J.M. et al.	2010	4	2	2	8
de Queiroz B.Z. et al.	2017	4	2	3	9
Elkan P et al.	2016	3	2	3	8
France R, Urban B	1990	3	2	3	8
Gebhardt et al	2006	4	2	3	9
Gjefsen E. et al.	2021	4	2	3	9
Hider S.L. et al.	2019	4	2	3	9
Kraychete D.C. et al.	2010	4	1	2	7
Ohtori S. et al.	2011	3	1	3	6
Park CH, Lee SH	2011	3	2	2	7
Rannou et al.	2007	4	2	2	8
Surbakti K.P. & Nasution I.	2020	3	1	2	6
Teodorczyk-Injeyan J et al.	2019	3	2	2	7

Note: A study awarded with 3 or 4 stars in the selection domain and 1 or 2 stars in the comparability domain and 2 or 3 stars in the outcome domain was considered high qualit.

Figure 3. (A–E) Pre- and post-treatment back pain scores in visual analogue scale 10 cm (VAS 10 cm), visual analogue scale 100 mm (VAS 100 mm), numerical rating scale (NRS) and brief pain inventory short form were reported as the mean score and standard deviations (SD), and represented in Figure 3(A–D), respectively. Forest plot of back pain score changes in different scales following treatments were reported as standardized mean difference (SMD) and 95% confidence intervals (CI). Sensitivity analysis based on the type of pain scale used was performed. Four methods were used to measure pain, VAS (10 cm), VAS (100 mm), NRS and the brief pain inventory short form.

Table 4. Inflammatory biomarker changes following different treatments were reported as standardized mean difference and 95% confidence intervals. Grading of recommendations assessment, development and evaluation (GRADE) level of quality assessment.

Biomarkers	Author	Year	Total	Pre-op mean	SD	Post-op mean	SD	SMD (95%CI)	GRADE quality of evidence
Beta-endorphin
	Wei-Chun L et al.	2015	30	138.21	47.95	129.54	42.57	0.19 (–0.32, 0.70)
	Wei-Chun L et al.	2015	31	121.81	35.29	113.92	23.88	0.26 (–0.24, 0.76)
	France R, Urban B	1990	28	67.5	25.6	66.70	23.9	0.03 (–0.49, 0.56)
Subtotal ($I^2$ = 0.0%, p = 0.818)								0.17 (–0.13, 0.46)
CTX-1
	Koivisto K. et al.	2019	20	0.3	0.2	1.90	4.23	–0.53 (–1.17, 0.10)
	Koivisto K. et al.	2019	20	0.2	0.2	0.30	0.16	–0.55 (–1.18, 0.06)
Subtotal ($I^2$ = 0.0%, p = 0.969)								–0.54 (–0.99, -0.10)*	M^2
IL-10
	Wei-Chun L et al.	2015	30	16.17	11.38	27.55	9.82	–1.07 (–1.61, –0.53)
	Wei-Chun L et al.	2015	31	17.7	21.14	31.19	13.44	–0.76 (–1.28, –0.25)
Subtotal ($I^2$ = 0.0%, p = 0.418)								–0.91 (–1.28, –0.53)*	M^2
IL-1 beta
	Wei-Chun L et al.	2015	30	29.83	12.73	14.83	10.00	1.31 (0.75, 1.87)
	Wei-Chun L et al.	2015	31	32.56	14.77	17.30	22.25	0.81 (0.29, 1.33)
Subtotal ($I^2$ = 40.0%, p = 0.197)								1.05 (0.56, 1.54)*	M^2
IL-6
	Koivisto K. et al.	2019	20	0.7	0.4	20	0.70	0.00 (–0.62, 0.62)
	Koivisto K. et al.	2019	20	0.8	0.5	20	0.73	0.20 (–0.43, 0.82)
	Degenhardt B.F. et al	2017	13	0.047	0.116	13	0.02	0.30 (–0.47, 1.08)
	Degenhardt B.F. et al	2017	13	0.213	0.382	13	0.18	0.12 (–0.65, 0.89)
	Degenhardt B.F. et al	2017	7	3.87	9.05	7	0.33	0.55 (–0.52, 1.62)
	Wei-Chun L. et al.	2015	30	33.28	18.47	30	31.10	0.13 (–0.39, 0.63)
	Wei-Chun L. et al.	2015	31	36.48	25.82	31	35.21	0.04 (–0.45, 0.54)
Subtotal ($I^2$ = 0.0%, p = 0.981)								0.13 (–0.11, 0.37)
TNFa
	Koivisto K. et al.	2019	20	1.6	0.798	20	1.80	–0.22 (–0.84, 0.41)
	Koivisto K. et al.	2019	20	2	0.558	20	1.80	0.33 (–0.29, 0.96)
	Degenhardt B.F. et al	2017	13	2.97	3.43	13	3.73	–0.26 (–1.03, 0.51)
	Degenhardt B.F. et al	2017	13	4.013	4.23	13	3.19	0.24 (–0.53, 1.01)
	Degenhardt B.F. et al	2017	7	3.29	4.54	7	3.05	0.05 (–1.00, 1.10)
	Wei-Chun L. et al.	2015	30	111.68	44.13	30	114.23	–0.06 (–0.56, 0.45)
	Wei-Chun L. et al.	2015	31	127.41	51.81	31	121.80	0.11 (–0.38, 0.61)
	Yuen J et al.	2017	12	6.4	6.9	12	2.90	0.63 (–0.19, 1.45)
	Yuen J et al.	2017	12	6.2	5.2	12	3.60	0.57 (–0.25, 1.38)
Subtotal ($I^2$ = 0.0%, p = 0.680)								0.11 (-0.11, 0.34)
hsCRP
	Park CH, Lee SH	2011	55	3.2	3.4	55	0.70	1.02 (0.62, 1.42)
	Koivisto K. et al.	2019	20	0.9	0.558	20	0.83	0.11 (–0.51, 0.73)
	Koivisto K. et al.	2019	20	1.2	1.755	20	1.17	0.02 (–0.60, 0.64)
	Degenhardt B.F. et al	2017	13	0.503	0.515	13	0.55	–0.09 (–0.86, 0.68)
	Degenhardt B.F. et al	2017	13	0.987	1.0797	13	0.80	0.17 (–0.60, 0.94)
	Degenhardt B.F. et al	2017	7	3.31	3.95	7	4.10	–0.16 (–1.21, 0.89)
Subtotal ($I^2$ = 65.6%, p = 0.013)								(0.24 (–0.22, 0.71)

Note: Quality of evidence: H = high, M = moderate, L = low, VL = very low. * bolded SMD values signifies a significant result (p < 0.05).

GRADE rating - ¹-rated down for imprecision,²-rated down for risk of bias, ³-rated down for inconsistency, ⁴-rated down for publication bias, ⁵-rated up for large magnitude of effect (strong evidence of association (+1), very strong evidence of association (+2)).

Table 5. The table presents a detailed summary of the evidence, including statistical model (effect size and associated confidence intervals, regression data and the certainty of evidence).

Outcome	Test	Statistical Model
		ES/r	95%CI	P-value
Back Pain	Meta-regression
	Beta-endorphin	2.76	–3.70, 9.22	0.207
	CXCL12	2.75	–3.24, 8.75	0.187
	CXCL6	–0.55	–10.14, 9.04	0.827
	GM-CSF	0.32	–4.10, 4.73	0.851
	hsCRP	–3.44	–5.16, –1.69	0.003*
	IL-1 beta	2.31	–6.72, 11.33	0.564
	IL-10	2.50	–3.91, 8.91	0.377
	IL-6	0.52	–3.10, 4.14	0.333
	IL-2	1.04	–3.97, 6.05	0.638
	IL-4	–1.74	–5.25, 2.05	0.271
	IL-8	–1.63	–6.67, 3.41	0.379
	MCP-1	4.46	2.72, 6.20	0.004*
	TNFa	1.45	–1.30, 4.21	0.283
Leg Pain	Meta-regression
	hsCRP	–1.83	–4.02, 0.37	0.077
	IL-6	–1.20	–1.20, –0.41	0.023*
	MCP-1	4.34	1.30, 7.38	0.025*
	TNFa	–0.21	–1.65, 1.22	0.586
Disability Score	Meta-regression
	hsCRP	–4.04	–4.54, –3.55	<0.001*
	IL-10	1.19	–4.52, 6.90	0.637
	Il-16	–0.53	–3.71, 2.64	0.684
	IL-1 beta	2.76	–5.71, 11.22	0.467
	IL-2	0.75	–4.11, 5.61	0.726
	IL-4	–1.58	–3.70, 0.54	0.114
	IL-6	1.43	–4.11, 6.98	0.580
	IL-8	3.36	2.71, 4.01	<0.001*
	TNFa	0.52	–2.74, 3.78	0.739

Note: * bolded ES/r, 95%CI and P-value signifies a significant result (p < 0.05).

Figure 4. Relationship between hsCRP, LBP and confounding (A) factors (modic changes, herniated disk confirmed on MRI and age) and collider (B) factors (chronic duration of LBP) that are included in the studies analysed.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.