Distinct patterns of cytokines, chemokines, and growth factors in synovial fluid after ACL injury in comparison to osteoarthritis

Abstract

This study analyzed knee synovial fluid after anterior cruciate ligament (ACL) tear and in osteoarthritis (OA) to test the hypotheses that concentrations of cytokines, chemokines, and growth factors differ (a) by diagnosis and, (b) after ACL tear by time from injury and presence/absence of concomitant meniscus tear. Synovial fluid samples were collected from two groups, ACL tears (with or without meniscus tear) (N=13) and Kellgren-Lawrence grade 3 and 4 OA (N=16), undergoing clinically indicated aspiration of the knee joint. Multiple cytokines, chemokines, and growth factors were assessed using a multiplexed 45-protein panel. Comparisons were made for the concentrations of all molecules between ACL tear and OA patients, isolated vs. combined ACL and meniscus tears, and categorized by time from injury: acute and early subacute (<15 days, N=8) vs. late subacute (>15 days and < 3 months, N=5). ACL tear patients have higher levels of six molecules (IL-4, IL-5, IL-13, PlGF-1, bNGF, TNF-α) in knee synovial fluid compared to OA patients. Isolated ACL tears express higher levels of IL-4, IL-13 and IFN-γ and lower levels of IL-7 than ACL tears with a concomitant meniscus tear. SDF-1α, PlGF-1, IL-1RA, HGF, bNGF, and BDNF levels are elevated immediately after injury and drop off significantly in the late subacute phase (after 15 days). Synovial fluid from knees with ACL tears have elevated metabolic activity compared to knees with OA. The cytokine profiles after ACL tears are influenced by the time from injury and the presence of meniscus tears. These findings offer valuable insights into the levels of cytokines, chemokines, and growth factors in the knee after ACL injury, information which may have important implications for the diagnosis, prognosis and treatment of this common pathology.

INTRODUCTION

Patients with anterior cruciate ligament (ACL) tears typically present with effusions, which often contain hemarthrosis^{1; 2}. The current clinical approach to managing these effusions varies, with aspiration considered if they cause pain or limit knee range of motion. These synovial fluid effusions contain inflammatory mediators, believed to correlate with the extent of cartilage damage³ and potentially reflecting alterations in the molecular processes involved in joint homeostasis. Post-injury, synovial fluid may initiate an inflammatory response, potentially contributing to repair potential as well as the early cascade of joint degeneration⁴.

Previous studies have highlighted that acute knee injuries can trigger the release of inflammatory mediators leading to long-term cartilage damage^{5; 6}. ACL tears, in particularly, pose a high risk for knee osteoarthritis (OA) development within 10–20 years⁷. Surprisingly, no studies yet have compared the levels of pro- and anti-inflammatory cytokines, chemokines, and growth factors in knee synovial fluid between ACL tear patients and late-stage OA patients. Some studies have examined these molecules in subjects undergoing ACL-reconstruction procedure^{8; 9} or those with meniscus and/or ligament injuries¹⁰. These studies have basically reported a “pro-inflammatory” response in injured joints, with all anti-inflammatory, anti-catabolic, or pro-anabolic proteins either absent or present at low concentrations. These changes may serve as initiating factors for cartilage degeneration, setting the stage for OA development¹¹. It is plausible that the injured joint exhibits both a robust inflammatory reaction in conjunction with a coordinated anabolic response, which may be crucial for tissue healing and restoration of joint homeostasis and function^{8; 12}. An improved understanding of the sequence of expression of such factors, and their differences between early injury and late-stage OA, may provide new targets for OA therapy.

However, conflicting data exist concerning the cytokine burden in the synovial fluid of ACL-deficient knees. While numerous studies have reported elevated pro-inflammatory cytokines persisting for up to three months post-ACL tear^{11; 13}, some studies have indicated an initial surge in these cytokines within the first 24 hours after injury, followed by a decline in concentration^{11; 12}. Information on cytokine, chemokine, and growth factor molecules in the synovial fluid from knees with isolated ACL tears compared to combined ACL and meniscus tears is limited and primarily at the time of corrective surgery^{8; 10; 14; 15}. This study seeks to test the hypotheses that the pattern of synovial fluid cytokines, chemokines, and growth factors in knees with an ACL tear differs (a) from knees affected by OA, and (b) by time from injury and presence/absence of concomitant meniscus tear.

METHODS

Patient recruitment and synovial fluid collection

The Institutional Review Board of the study site approved the research protocol. We recruited and obtained consent from eligible patients through the practice of an academic sports medicine fellowship trained orthopedic surgeon. Patients diagnosed with OA were included if they exhibited radiographically confirmed Kellgren-Lawrence (K-L) grade ≥2 degenerative changes and were undergoing aspiration and corticosteroid injection as part of their conservative management. In contrast, patients with ACL tear were diagnosed based on their clinical history, physical examination, and magnetic resonance imaging, with a clinical indication for aspiration aimed at reducing pain and improving the range of knee motion. ACL tear patients had little or no radiographic evidence for OA (K-L 0 or 1). The diagnostic groups consisted of two categories (i) ACL tear (N = 13), and (ii) grade 3 and 4 OA (N = 16) (Table 1). The presence of meniscus tears was confirmed via MRI for all patients, as not all patients underwent surgical intervention. Nevertheless, among those who did undergo surgical treatment, all had verified meniscus tears that were subsequently addressed through partial meniscectomy or repair.

Table 1:

Detail of study patients that provided synovial fluid samples

Patient ID	Group/Injury pattern	Age (years)	Sex	BMI (kg/m2)	Diagnosis	TFI (days)	K-L grade	Smoking	Diabetes	Effusion size (mL)	Previous injection	Time since last injection (months)	NSAID used
P14-SF025	ACLT	16	Female	23.4	ACLT	4	0	No	No	17	NO	-	No
P14-SF036	ACLT	15	Male	25.7	ACLT	9	0	No	No	30	NO	-	Intermittent OTC
P14-SF010	ACLT	57	Female	25.7	ACLT, MCL sprain	12	0	No	No	20	NO	-	Intermittent OTC
P14-SF011	ACLT+MT	49	Female	26.6	ACLT, MMT, LMT	6	0	No	No	25	NO	-	Intermittent OTC
P14-SF012	ACLT+MT	41	Female	26.6	ACLT, MMT	16	0	No	No	45	NO	-	Intermittent OTC
P14-SF017	ACLT+MT	41	Female	27	ACLT, MMT	2	0	No	No	60	NO	-	Intermittent OTC
P14-SF020	ACLT+MT	52	Female	17.9	ACLT, MMT	73	0	Yes	No	6	NO	-	No
P14-SF023	ACLT+MT	23	Male	22	ACLT, MMT, LMT	31	0	No	No	30	NO	-	No
P14-SF026	ACLT+MT	32	Male	25.4	ACLT, LMT	1	0	No	No	30	NO	-	Prescription (meloxicam)
P14-SF028	ACLT+MT	66	Male	32.4	ACLT, MMT, LMT	31	1	No	No	20	NO	-	Intermittent OTC
P14-SF013	ACLT+MT	26	Male	25.1	ACLT, LMT	4	0	No	No	70	NO	-	Intermittent OTC
P14-SF029	ACLT+MT	19	Female	26.6	ACLT, LMT	6	0	No	No	20	NO	-	Intermittent OTC
P14-SF032	ACLT+MT	20	Male	24.4	ACLT, LMT	36	0	No	No	40	NO	-	Intermittent OTC
P14-SF002	OA	59	Male	35.2	Knee OA	-	2	No	No	30	YES	4	Intermittent OTC
P14-SF003	OA	55	Male	31	Knee OA	-	2	No	No	50	NO	-	No
P14-SF004	OA	65	Female	25	Knee OA	-	2	No	No	40	YES	11	No
P14-SF005	OA	57	Female	21	Knee OA	-	4	No	No	35	NO	-	No
P14-SF014	OA	55	Female	35.5	Knee OA	-	2	No	No	50	NO	-	Intermittent OTC
P14-SF015	OA	45	Male	32.1	Knee OA	-	2	No	No	35	NO	-	No
P14-SF018	OA	55	Male	28.2	Knee OA	-	4	No	No	60	YES	5	No
P14-SF019	OA	50	Male	40.1	Knee OA	-	4	Yes	No	115	YES	3	Intermittent OTC
P14-SF022	OA	58	Female	27.3	Knee OA	-	4	No	No	17	NO	-	Intermittent OTC
P14-SF024	OA	48	Male	27.6	Knee OA	-	4	No	No	50	NO	-	Intermittent OTC
P14-SF027	OA	54	Male	39.6	Knee OA	-	3	Yes	No	20	YES	7	No
P14-SF030	OA	69	Female	25.1	Knee OA	-	4	No	No	20	YES	4	No
P14-SF031	OA	61	Male	28	Knee OA	-	4	No	No	20	NO	-	No
P14-SF033	OA	52	Male	31.8	Knee OA	-	4	No	No	10	YES	3	Intermittent Celebrex
P14-SF037	OA	58	Female	32.2	Knee OA	-	3	No	No	20	YES	3	Intermittent OTC
P14-SF038	OA	53	Female	28	Knee OA	-	4	No	No	5	YES	36	Intermittent OTC

ACLT = anterior cruciate ligament tear; MT = meniscus tear; OA = osteoarthritis; BMI = body mass index; MCL = medial collateral ligament; MMT = medial meniscus tear;

LMT = lateral meniscus tear; TFI = time from injury; K-L grade = Kellgren-Lawrence grade; NSAID = non-steroidal anti-inflammatory drug; OTC = over the counter

The synovial fluid was aseptically aspirated using an 18-gauge needle, entering the joint through a superolateral approach, with the knee fully extended. There was no observable evidence of iatrogenic contamination in the synovial fluid samples at the time of aspiration. All OA and some ACL tear samples contained yellow serous fluid whereas some ACL samples contained dark sanguineous fluid. Subsequently, the collecting syringe was capped and promptly transferred to the laboratory within 30–60 minutes of collection. The synovial fluid was then subjected to centrifugation at 3000g for 20 minutes at 4°C to eliminate cells and particulate material, with the resulting supernatant stored at −80°C until further assay.

Analysis of cytokines, chemokines, and growth factor molecules

We conducted a comprehensive analysis of biochemical molecules using a multiplex ELISA-based immunoassay (Luminex), specifically, the Cytokine/Chemokine/Growth Factor 45-Plex Human ProcartaPlex^™ Panel 1 (Cat. # EPX450-12171-901). After being centrifuged at 15,000g for 10 min at 4°C, the synovial fluid samples were added to the beads. Following this, we rigorously adhered to the manufacturer’s protocol (ProcartaPlex^™ Multiplex Immunoassay [MAN0017081], Thermo Fisher Scientific). The results were quantified and expressed as pg/mL. The selection of this panel was based on the established roles of these molecules in injury and OA as outlined in Table 2.

Table 2:

Molecular markers used in this study

Analyte	Name	Alias	Type	Function	PubMed ID
MIP-1α	Macrophage inflammatory protein 1 alpha	CCL3	C-C family chemokine	Pro-inflammatory	8558069
MIP-1β	Macrophage inflammatory protein 1 beta	CCL2	C-C family chemokine	Pro-inflammatory	12401480
MCP-1	Monocyte chemoattractant protein 1	CCL2	C-C family chemokine	Pro-inflammatory	19441883
RANTES	Regulated on activation, normal T cell expressed and secreted	CCL5	C-C family chemokine	Pro-inflammatory	11286708
SDF-1α	Stromal cell-derived factor 1	CXCL12	C-C family chemokine	Pro-inflammatory	15339043
Eotaxin	Eotaxin	CCL11	C-C family chemokine	Pro-inflammatory	8609214
GRO-α	Growth related oncogene alpha	CXCL1	C-X-C family chemokine	Pro-inflammatory	17534244
IP-10	Interferon γ-induced protein 10 kDa	CXCL10	C-X-C family chemokine	Pro-inflammatory	21802343
IL-8	Interleukin 8	CXCL8	C-X-C family chemokine	Pro-inflammatory	8315568
TNF-α	Tumor necrosis factor alpha		Cytokine	Pro-inflammatory	10936147
TNF-β	Tumor necrosis factor beta		Cytokine	Pro-inflammatory	18508030
EGF	Epidermal growth factor		Growth factor	Anti-inflammatory	24527320
FGF-2	Fibroblast growth factor		Growth factor	Anti-inflammatory	19565481
GM-CSF	Granulocyte-macrophage colony-stimulating factor		Growth factor	Pro-inflammatory	16474424
HGF	Hepatocyte growth factor		Growth factor	Pro-inflammatory	25667190
bNGF	Beta nerve growth factor		Growth factor	Pro-inflammatory	24438745
PDGF-BB	Platelet-derived growth factor BB		Growth factor	Anti-inflammatory	16413799
PlGF-1	Placenta growth factor 1		Growth factor	Pro-inflammatory	10606981
VEGF-A	Vascular endothelial growth factor A		Growth factor	Pro-inflammatory	17945514
VEGF-D	Vascular endothelial growth factor A		Growth factor	Pro-inflammatory	27163679
BDNF	Brain-derived neurotrophic factor		Growth factor	Pro-inflammatory	24399456
SCF	Stem cell factor		Growth factor	Pro-inflammatory	9269751
LIF	Leukemia inhibitory factor		IL-6 family cytokine	Anti-inflammatory	26807429
IFN-α	Interferon alpha		Interferon (cytokine)	Pro-inflammatory	12365360
IFN-g	Interferon gamma		Interferon (cytokine)	Pro-inflammatory	8624619
IL-1RA	Interleukin 1 receptor antagonist		Interleukin (cytokine)	Anti-inflammatory	9597123
IL-1α	Interleukin 1 alpha		Interleukin (cytokine)	Pro-inflammatory	9497936
IL-1β	Interleukin 1 beta		Interleukin (cytokine)	Pro-inflammatory	18925684
IL-2	Interleukin 2		Interleukin (cytokine)	Pro-inflammatory	15473953
IL-4	Interleukin 4		Interleukin (cytokine)	Anti-inflammatory	11145704
IL-5	Interleukin 5		Interleukin (cytokine)	Anti-inflammatory	21986312
IL-6	Interleukin 6		Interleukin (cytokine)	Pro- and anti-inflammatory	16899107
IL-7	Interleukin 7		Interleukin (cytokine)	Pro-inflammatory	12010786
IL-9	Interleukin 9		Interleukin (cytokine)	Anti-inflammatory	21368237
IL-10	Interleukin 10		Interleukin (cytokine)	Anti-inflammatory	11244051
IL-12p70	Interleukin 12 p70		Interleukin (cytokine)	Anti-inflammatory	19275692
IL-13	Interleukin 13		Interleukin (cytokine)	Anti-inflammatory	8096327
IL-15	Interleukin 15		Interleukin (cytokine)	Pro-inflammatory	24167367
IL-17A	Interleukin 17 A		Interleukin (cytokine)	Pro-inflammatory	21852080
IL-18	Interleukin 18		Interleukin (cytokine)	Pro-inflammatory	24115947
IL-21	Interleukin 21		Interleukin (cytokine)	Pro-inflammatory	12429707
IL-22	Interleukin 22		Interleukin (cytokine)	Pro-inflammatory	23405899
IL-23	Interleukin 23		Interleukin (cytokine)	Pro-inflammatory	21585245
IL-27	Interleukin 27		Interleukin (cytokine)	Pro- and anti-inflammatory	23244718
IL-31	Interleukin 31		Interleukin (cytokine)	Pro-inflammatory	18926762
NO	Nitric oxide		NO/free radical	Pro-inflammatory	7513156

Statistical analysis

We conducted a priori estimation of the sample size using the online G*Power tool (Heinrich Heine Universität Düsseldorf, Germany). Assuming a 2-tailed independent t-test with an α = 0.05 and a 1:1 ratio of patients with ACL tear and those with OA, we determined that a sample size of 12 patients per group would be sufficient to detect an effect size of 1.2 between groups with 80% power. However, the sample size for the acute and subacute ACL tear groups did not meet the power calculation criteria mentioned above and is therefore considered a convenience sample.

We employed IBM SPSS Statistics for the comparative analysis of cytokine, chemokine, and growth factor molecules across different patient groups, including patients with OA versus all ACL injured patients, patients with isolated ACL tears versus those with combined ACL and meniscus tears, and acute and early subacute (< 15 days since injury) versus late subacute (> 15 days and less than 3 months since injury) ACL injured patients¹¹. In each analysis, we applied an ANCOVA model, considering patient type and sex as factors, while age and body mass index served as covariates. Comparative assessments between patient groups were based on estimated marginal means (± standard error of the mean), with statistical significance set at the p value less than 0.05. In Tables, 95% confidence intervals (with upper and lower limit) have also been displayed.

RESULTS

Patients’ characteristics

A total of 29 patients were recruited, comprising 16 individuals with K-L grade 3–4 OA and 13 with knee injury. As expected, there were significant differences between the OA and knee injury patient groups. OA patients were significantly older (55.88±1.51 vs. 35.15±4.72 years p = 0.0001), had a higher body mass index (30.48±1.32 vs. 25.29±0.91 kg/m² p = 0.005), and exhibited more severe degenerative changes (K-L score 3.25±0.23 vs. 0.077±0.077, p < 0.0001) in comparison to injured patients. Among the injured patients, nine were intermittently using over-the-counter non-steroidal anti-inflammatory drugs (NSAIDs), three had not taken any NSAIDs, and one was prescribed a NSAID. None of the injured patients had a history of a prior corticosteroid injection. Within the OA patient cohort, seven were using over-the-counter NSAIDs intermittently, eight were not using any NSAIDs, and one was taking a prescription NSAID intermittently. Furthermore, nine of the OA patients had a history of a previous corticosteroid injection in the knee, with a median interval of 4 months before the study (range: 3–36 months).

Injured vs. OA

Using multivariate analysis, we identified six molecules (IL-4, IL-5, PlGF-1, IL-13, TNF-α, bNGF) that displayed significantly higher levels in patients with ACL tears (Table 3) in comparison to those with OA (Fig. 1). Three additional molecules approached statistically significant elevation in ACL tear patients when compared to OA patients, namely IL-6 (p = 0.052), IL-1RA (p = 0.063), and IFN-γ (p = 0.074).

Table 3:

Comparison of molecular markers between ACLT and OA patients

Marker	ACLT (N = 13)				OA (N = 16)				P value*
	Mean	S.E.M.	95% CI		Mean	S.E.M	95% CI
			Lower	Upper			Lower	Upper
MIP-1α	1.45	0.66	0.08	2.82	2.35	0.58	1.15	3.54	.391
SDF-1α	101.43	20.28	59.57	143.30	51.60	17.69	15.10	88.10	.125
IL-27	6.78	1.10	4.52	9.05	4.93	0.96	2.95	6.90	.285
LIF	29.99	8.24	12.98	46.99	12.33	7.18	−2.49	27.15	.179
IL-1β	1.25	0.15	0.93	1.56	1.09	0.13	0.81	1.36	.512
IL-2	6.70	0.91	4.82	8.58	4.78	0.80	3.14	6.42	.187
IL-4	3.11	0.20	2.69	3.53	2.42	0.18	2.05	2.79	.039
IL-5	2.87	0.36	2.13	3.61	1.60	0.31	0.95	2.24	.031
IP-10	−8.16	2.57	7.23	17.86	6.76	2.24	2.13	11.39	.159
IL-6	285.42	64.36	152.58	418.25	80.43	56.11	−35.38	196.25	.051
IL-7	0.54	0.13	0.26	0.81	0.69	0.12	0.45	0.93	.453
IL-8	1.17	1.49	−1.90	4.25	0.83	1.30	−1.85	3.51	.881
IL-10	0.84	0.20	0.43	1.25	0.70	0.17	0.34	1.06	.653
PlGF-1	3.45	0.85	1.69	5.22	0.09	0.75	−1.45	1.63	.018
Eotaxin	11.39	1.96	7.34	15.43	8.81	1.71	5.29	12.33	.403
IL-12p70	0.78	0.05	0.67	0.89	0.73	0.05	0.63	0.83	.586
IL-13	1.74	0.07	1.60	1.88	1.37	0.06	1.25	1.49	.001
IL-17A	1.12	0.10	0.91	1.32	0.91	0.09	0.72	1.09	.189
IL-31	0.09	0.02	0.04	0.13	0.06	0.02	0.02	0.10	.394
IL-1RA	664.11	226.47	196.70	1131.53	−18.66	197.45	−426.19	388.86	.063
SCF	11.77	2.17	7.29	16.25	12.89	1.89	8.99	16.80	.742
RANTES	0.48	0.15	0.16	0.79	0.08	0.13	−0.19	0.35	.104
IFN-γ	4.11	0.42	3.24	4.97	2.88	0.37	2.12	3.63	.072
GM-CSF	9.08	1.28	6.44	11.72	7.30	1.11	5.00	9.60	.377
TNF-α	2.82	0.23	2.33	3.30	1.70	0.20	1.28	2.12	.005
HGF	203.54	40.44	120.08	287.00	138.32	35.26	65.55	211.08	.308
MIP-1β	1.82	1.47	−1.22	4.86	2.08	1.28	−0.57	4.73	.908
IFN-α	0.03	0.00	0.02	0.03	0.02	0.00	0.02	0.03	.474
MCP-1	184.79	39.64	102.97	266.60	119.31	34.56	47.98	190.64	.296
IL-9	ND	ND	ND	ND	ND	ND	ND	ND	-
VEGF-D	0.78	0.96	−1.21	2.76	1.03	0.84	−0.71	2.76	.867
TNF-β	ND	ND	ND	ND	ND	ND	ND	ND	-
bNGF	3.92	0.23	3.45	4.39	2.88	0.20	2.47	3.29	.007
EGF	0.23	0.02	0.20	0.26	0.22	0.01	0.19	0.24	.473
BDNF	0.42	0.09	0.24	0.60	0.23	0.08	0.08	0.39	.193
GRO-α	0.60	0.51	−0.45	1.64	0.68	0.44	−0.23	1.59	.917
IL-1α	0.05	0.01	0.02	0.08	0.03	0.01	0.01	0.05	.383
IL-23	3.50	0.19	3.10	3.90	3.26	0.17	2.91	3.61	.427
IL-15	0.26	0.07	0.12	0.41	0.27	0.06	0.14	0.39	.975
IL-18	7.25	0.75	5.69	8.81	5.37	0.66	4.02	6.73	.121
IL-21	0.56	0.28	−0.02	1.15	0.42	0.25	−0.09	0.93	.741
FGF-2	0.18	0.06	0.06	0.29	0.11	0.05	0.01	0.22	.473
IL-22	64.70	24.15	14.86	114.53	44.74	21.05	1.28	88.19	.598
PDGF-BB	33.42	2.00	29.30	37.55	28.81	1.74	25.22	32.41	.149
VEGF-A	1212.89	350.03	490.45	1935.32	474.12	305.18	−155.74	1103.99	.185
NO	2.87	0.87	1.08	4.66	1.17	0.76	−0.39	2.73	.218

ACLT = anterior cruciate ligament tear; OA = osteoarthritis; ND = not detected; S.E.M. = standard error of the mean

^{* =}bold values indicate statistical significance

Figure 1. Assessment of cytokines/chemokines/growth factors in synovial fluid of ACL tear patients and OA patients.

An assessment of concentration of a panel of 46 cytokines/chemokines/growth factors in the synovial fluid of ACL tear (N = 13) and OA (N = 16) patients revealed that six molecules were significantly different between the two groups. The estimated marginal mean ± SEM. were plotted for IL-4, IL-5, IL-13, PlGF1, bNGF and TNF-α. All values are pg/mL and all of the molecules shown here had higher concentration in ACL tear group. Similar letters indicate statistical significance at a P value of < 0.05. ACLT = anterior cruciate ligament tear; OA = osteoarthritis

Isolated ACL tears vs. combined ACL and meniscus tears

Isolated ACL tears (N = 3) expressed higher IL-4 (p = 0.016), IL-13 (p = 0.022), IFN-γ (p = 0.002) (Fig. 2) and lower IL-7 (p = 0.027) (Table 4) than combined ACL and meniscus tears (N = 10).

Figure 2. Assessment of cytokines/chemokines/growth factors by pattern of injury.

An assessment of concentration of a panel of 46 cytokines/chemokines/growth factors in the synovial fluid of patients with isolated ACL tear (N = 3) and combined ACL tear and meniscus tear (N = 10) showed that IL-4 and IL-7 were significantly different between two groups. Concentration of IL-4 was significantly higher in ACL tear group, while that IL-7 was higher in ACL tear + meniscus tear group. Values are expressed as pg/mL and plotted as mean ± SEM. Similar letters indicate statistical significance at a P value of < 0.05. ACLT = anterior cruciate ligament tear; ACLT+MT = anterior cruciate ligament tear + meniscus tear

Table 4:

Comparison of molecular markers between ACLT and ACLT+MT patients

Marker	ACLT (N = 3)				ACLT+MT (N = 10)				P value*
	Mean	S.E.M.	95% CI		Mean	S.E.M	95% CI
			Lower	Upper			Lower	Upper
MIP-1α	2.97	1.20	0.21	5.74	2.03	0.63	0.57	3.49	.511
SDF-1α	119.17	43.13	19.70	218.63	95.91	22.79	43.36	148.46	.649
IL-27	8.62	1.26	5.72	11.52	5.76	0.66	4.22	7.29	.081
LIF	14.44	18.76	−28.81	57.69	32.34	9.91	9.48	55.19	.428
IL-1β	1.54	0.16	1.16	1.92	1.15	0.09	0.95	1.35	.074
IL-2	6.95	1.47	3.57	10.33	7.04	0.77	5.25	8.83	.958
IL-4	3.90	0.27	3.28	4.52	2.97	0.14	2.64	3.29	.016
IL-5	2.65	0.76	0.89	4.41	2.84	0.40	1.91	3.77	.828
IP-10	11.69	5.32	−0.58	23.95	13.70	2.81	7.22	20.18	.749
IL-6	302.40	174.02	−98.90	703.69	329.24	91.94	117.21	541.26	.896
IL-7	0.25	0.09	0.05	0.45	0.52	0.05	0.42	0.63	.027
IL-8	0.71	0.24	0.15	1.27	0.14	0.13	−0.15	0.44	.076
IL-10	0.74	0.41	−0.21	1.68	1.01	0.22	0.51	1.51	.572
PlGF-1	4.69	2.40	−0.85	10.23	2.61	1.27	−0.32	5.53	.469
Eotaxin	7.56	2.97	0.72	14.40	11.29	1.57	7.68	14.91	.303
IL-12p70	0.94	0.09	0.72	1.16	0.80	0.05	0.68	0.91	.214
IL-13	1.89	0.08	1.71	2.08	1.63	0.04	1.53	1.73	.022
IL-17A	1.38	0.16	1.00	1.76	1.09	0.09	0.89	1.29	.157
IL-31	0.06	0.04	−0.03	0.15	0.11	0.02	0.06	0.16	.269
IL-1RA	1654.05	524.07	445.54	2862.55	344.55	276.89	−293.96	983.07	.060
SCF	12.35	3.86	3.44	21.26	12.32	2.04	7.61	17.03	.995
RANTES	0.22	0.42	−0.75	1.19	0.44	0.22	−0.08	0.95	.665
IFN-γ	5.69	0.47	4.60	6.77	3.35	0.25	2.77	3.92	.002
GM-CSF	10.44	1.33	7.39	13.50	8.78	0.70	7.16	10.39	.303
TNF-α	3.02	0.52	1.82	4.23	3.00	0.28	2.36	3.64	.970
HGF	300.04	80.92	113.44	486.64	203.03	42.75	104.44	301.62	.324
MIP-1β	−0.64	3.35	−8.37	7.10	3.21	1.77	−0.88	7.29	.345
IFN-α	0.02	0.00	0.02	0.03	0.03	0.00	0.02	0.03	.609
MCP-1	120.02	58.58	−15.08	255.11	160.80	30.95	89.42	232.17	.559
IL-9	ND	ND	ND	ND	ND	ND	ND	ND	-
VEGF-D	0.02	0.98	−2.25	2.28	0.72	0.52	−0.48	1.92	.547
TNF-β	ND	ND	ND	ND	ND	ND	ND	ND	-
bNGF	3.66	0.57	2.35	4.97	3.94	0.30	3.25	4.63	.675
EGF	0.25	0.02	0.20	0.31	0.24	0.01	0.21	0.27	.658
BDNF	0.44	0.22	−0.07	0.96	0.45	0.12	0.17	0.72	.984
GRO-α	2.13	0.73	0.45	3.82	0.23	0.39	−0.66	1.12	.052
IL-1α	0.05	0.02	0.00	0.11	0.04	0.01	0.01	0.07	.762
IL-23	3.79	0.36	2.95	4.63	3.34	0.19	2.90	3.78	.311
IL-15	0.41	0.09	0.20	0.62	0.26	0.05	0.14	0.37	.184
IL-18	9.35	1.45	6.01	12.68	6.34	0.76	4.58	8.10	.106
IL-21	1.02	0.62	−0.41	2.44	0.60	0.33	−0.15	1.36	.574
FGF-2	0.14	0.12	−0.14	0.42	0.20	0.06	0.06	0.35	.640
IL-22	22.75	49.56	−91.53	137.04	93.32	26.19	32.94	153.71	.248
PDGF-BB	35.00	4.42	24.81	45.19	32.76	2.34	27.37	38.14	.668
VEGF-A	907.48	998.37	−1394.77	3209.72	1343.60	527.49	127.20	2559.99	.712
NO	2.43	2.45	−3.21	8.06	3.64	1.29	0.66	6.61	.676

ACLT = anterior cruciate ligament tear; ACLT+MT = anterior cruciate ligament tear + meniscus tear;

ND = not detected; S.E.M. = standard error of the mean;

^{* =}bold values indicate statistical significance

Time from injury

Six molecules were more highly expressed during the acute and early subacute phase of injury: SDF-1α (p = 0.028), PlGF-1 (p = 0.003), IL-1RA (p = 0.029), HGF (p = 0.025), bNGF (p = 0.015), and BDNF (p = 0.029) (Fig. 3 and Table 5).

Figure 3. Time-dependent changes in the synovial fluid concentrations of cytokines/chemokines/growth factors.

An assessment of concentration of a panel of 46 cytokines/chemokines/growth factors in the synovial fluid of patients with ACL tear (including ACL tear + meniscus tear) showed that the concentrations of six molecules (SDF-1α, PlGF-1, IL-1RA, HGF, bNGF, and BDNF) significantly changed between acute (N = 8) and chronic (N = 5) injury groups. The means ± S.E.M. were plotted with all values in pg/mL.

Table 5:

Comparison of molecular markers between acute (<15 days) and chronic (>15 days) groups

Marker	Acute (N = 8)				Chronic (N = 5)				P value*
	Mean	S.E.M.	95% CI		Mean	S.E.M.	95% CI
			Lower	Upper			Lower	Upper
MIP-1α	2.06	0.78	0.25	3.87	2.52	0.99	0.23	4.80	0.744
SDF-1α	138.84	20.31	92.00	185.68	44.49	25.68	−14.74	103.71	0.028
IL-27	5.99	0.96	3.77	8.21	7.00	1.22	4.19	9.81	0.559
LIF	32.21	12.37	3.69	60.74	22.52	15.64	−13.55	58.59	0.662
IL-1β	1.30	0.13	1.00	1.59	1.16	0.16	0.79	1.52	0.538
IL-2	7.39	0.92	5.27	9.51	6.46	1.16	3.78	9.14	0.576
IL-4	3.38	0.23	2.84	3.91	2.87	0.29	2.20	3.54	0.243
IL-5	3.32	0.41	2.36	4.27	2.02	0.52	0.82	3.22	0.107
IP-10	15.91	3.15	8.65	23.18	9.26	3.98	0.07	18.44	0.257
IL-6	380.48	107.69	132.14	628.82	237.35	136.17	−76.65	551.35	0.464
IL-7	0.40	0.07	0.24	0.56	0.55	0.09	0.35	0.76	0.253
IL-8	0.47	0.16	0.10	0.84	−0.04	0.20	−0.51	0.43	0.104
IL-10	1.18	0.24	0.62	1.74	0.61	0.31	−0.10	1.32	0.216
PlGF-1	5.67	0.89	3.62	7.71	−0.82	1.12	−3.40	1.77	0.003
Eotaxin	11.36	1.99	6.78	15.94	9.10	2.51	3.31	14.89	0.529
IL-12p70	0.89	0.06	0.75	1.02	0.74	0.08	0.57	0.92	0.209
IL-13	1.71	0.07	1.55	1.88	1.66	0.09	1.45	1.87	0.677
IL-17A	1.16	0.12	0.89	1.44	1.14	0.15	0.79	1.49	0.896
IL-31	0.10	0.03	0.04	0.16	0.10	0.03	0.02	0.17	0.961
IL-1RA	1205.98	309.23	492.90	1919.06	−216.27	390.99	−1117.88	685.34	0.029
SCF	9.94	2.15	4.97	14.91	15.91	2.72	9.63	22.19	0.147
RANTES	0.64	0.24	0.08	1.19	0.02	0.30	−0.69	0.72	0.174
IFN-γ	4.44	0.47	3.36	5.53	3.01	0.59	1.64	4.38	0.116
GM-CSF	9.70	0.87	7.70	11.69	8.33	1.10	5.80	10.86	0.389
TNF-α	3.31	0.30	2.63	4.00	2.55	0.37	1.68	3.41	0.172
HGF	300.18	39.60	208.86	391.51	111.46	50.07	−4.02	226.93	0.025
MIP-1β	2.40	2.28	−2.86	7.65	2.27	2.88	−4.37	8.92	0.975
IFN-α	0.02	0.00	0.02	0.03	0.03	0.00	0.02	0.03	0.873
MCP-1	179.07	35.74	96.64	261.49	110.58	45.19	6.37	214.80	0.300
IL-9	ND	ND	ND	ND	ND	ND	ND	ND	-
VEGF-D	0.99	0.61	−0.42	2.39	−0.07	0.77	−1.84	1.71	0.346
TNF-β	ND	ND	ND	ND	ND	ND	ND	ND	-
bNGF	4.41	0.25	3.84	4.98	3.09	0.31	2.36	3.81	0.015
EGF	0.25	0.01	0.22	0.29	0.23	0.02	0.18	0.27	0.297
BDNF	0.64	0.10	0.40	0.88	0.16	0.13	−0.15	0.46	0.029
GRO-α	1.00	0.57	−0.32	2.33	0.14	0.73	−1.54	1.81	0.410
IL-1α	0.06	0.01	0.03	0.09	0.03	0.02	−0.02	0.07	0.208
IL-23	3.32	0.24	2.76	3.87	3.63	0.30	2.93	4.33	0.468
IL-15	0.25	0.06	0.10	0.40	0.35	0.08	0.16	0.53	0.410
IL-18	7.97	0.99	5.69	10.24	5.58	1.25	2.70	8.46	0.200
IL-21	1.02	0.37	0.17	1.87	0.20	0.47	−0.88	1.27	0.233
FGF-2	0.27	0.07	0.12	0.42	0.06	0.08	−0.13	0.26	0.105
IL-22	63.11	33.90	−15.06	141.29	99.22	42.86	0.37	198.06	0.555
PDGF-BB	32.68	2.85	26.10	39.25	34.13	3.60	25.82	42.44	0.776
VEGF-A	2043.57	500.27	889.94	3197.20	49.52	632.54	−1409.12	1508.16	0.050
NO	5.12	1.31	2.11	8.13	0.73	1.65	−3.08	4.54	0.088

S.E.M. = standard error of the mean; CI = confidence interval; ND = not detected;

^{* =}bold values indicate statistical significance

DISCUSSION

Synovial fluid concentrations of various cytokines, chemokines, and growth factors differ between patients with knee injury and knee OA. Our investigation also examined how the levels of these molecules in knees with ACL tears relate to the time elapsed since injury and the presence or absence of concomitant meniscus tears. The findings provide valuable insights into the biological response of the knee to ACL tear, while identifying potential avenues for diagnostic, prognostic and therapeutic approaches.

We identified six molecules that were elevated in ACL tears compared to OA, similar to previously published data¹². Among these molecules, four (IL-4, IL-5, PlGF-1, IL-13) are potent anti-inflammatory molecules, while the other two (TNF-α, bNGF) play pro-inflammatory roles. IL-4, with its pleiotropic anti-inflammatory and chondroprotective effects has been proposed as a therapeutic candidate in OA and rheumatoid arthritis^{16; 17}. IL-13, another pleiotropic immuno-regulatory anti-inflammatory cytokine, promotes chondrocyte proliferation and chondrogenic differentiation¹⁸. Our study revealed that IL-12 levels were higher in ACL tear patients compared to OA patients, suggesting a potential role in the response of the knee to ACL rupture. Interestingly, IL-4 and IL-13 share a receptor system, mediating their biological activities¹⁹.

Recent studies have shown that patients with K-L grade 3–4 have significantly higher levels of IL-5 in synovial fluid as compared with K-L grade 1–2, suggesting that higher IL-5 levels are associated with advanced OA²⁰. IL-4, IL-5, and IL-13 are Th2 cytokines, indicating the induction of a Th2 response following ACL injury. Additionally, our data align with previous findings reporting higher serum levels of IL-5, IL-6, and IL-13 in early OA compared to advanced OA²¹. IL-1RA, which we found significantly elevated in the acute phase of ACL injury compared to OA, aligns with other studies³. However, IL-1α and IL-1β levels were not different between ACL tear and OA groups in our study, highlighting the complex and varying cytokine responses following joint trauma^{3; 22 9; 10}. Thus it is likely that there are different phases of biological reaction (cytokine-activity) following joint trauma²³. Overall, our results suggest a pattern consistent with a modest anabolic response in the knee following ACL tears.

IL-6, with its multifaceted roles, including immune system involvement, cartilage synthesis nociception in OA joints, and bone homeostasis^{24; 25 26}, has been shown to be elevated in OA synovial fluid^26–28 and is positively correlated with patient-reported pain after ACL injury and reconstruction²⁹. Our data support this finding and suggest that prompt neutralization of IL-6 accumulation in synovial fluid may help prevent the development of OA in ACL-deficient knees^{10; 11}. Elevated levels of PlGF-1, a member of VEGF family with a role in chemotactic migration of human mesenchymal progenitor cells³⁰, indicate its potential role in inflammation-induced hypoxia and tissue infiltration, mirroring higher levels of IL-6 and TNF-α. It has also been detected at higher levels in the synovial fluid of rheumatoid patients compared to OA patients³¹, where it performs a pro-inflammatory role by stimulating the expression of TNF-α and IL-6 from synovial fluid mononuclear cells³². Again, higher concentrations of PlGF-1 in ACL tear patients corroborate well with our observation of higher concentrations of IL-6 and TNF-α. TNF-α is a known master pro-inflammatory cytokine and has long been associated with OA pathogenesis³³. A cross-sectional study reported that synovial fluid concentration of TNF-α is elevated after acute injury compared with a reference group⁶. Our observation that TNF-α is higher after ACL tear is consistent with another study, which reported a large increase in TNF-α levels after ACL rupture¹³. However, OA joints exhibit significantly more TNF-α¹⁰. Furthermore, TNF-α can induce the expression of IFN-γ, a pleiotropic cytokine produced by activated T-cells responsible for macrophage activation/differentiation and inducing transcription of several pro-inflammatory genes³⁴. We observed higher levels of IFN-γ in ACL tear patients, similar to TNF-α. Considering the crosstalk between these two molecules³⁵, it is likely that they are working synergistically in the injured joint.

Another growth factor that was significantly elevated in ACL tear synovial fluid was bNGF, a member of the neurotrophins family of proteins³⁶. bNGF levels are elevated in a wide-array of acute and chronic pain states such as rheumatoid arthritis and inflammatory arthritis^{37; 38} and may contribute directly to joint inflammation via activation of inflammatory cells. The higher levels of bNGF we observed in ACL tears patients are consistent with a stimulated inflammatory process

The presence of a concomitant meniscus tear was associated with a lower level of anti-inflammatory IL-4, IL-13 and INF-γ, and a higher concentration of pro-inflammatory IL-7. This may be early evidence of a biological basis for a higher risk of chondral degeneration and OA in patients with combined ACL and meniscus tears compared to isolated ACL tears. IL-4 exhibits potent anti-inflammatory activities by inhibiting the synthesis of master pro-inflammatory cytokines IL-1β and TNF-α³⁹. Conversely, IL-7 is a pro-inflammatory cytokine produced by OA chondrocytes which exacerbates MMP-13 expression and is associated with proteoglycan loss and OA⁴⁰. The negative impact of IL-7 on cartilage is mediated by inflammation-driven cartilage degeneration, T-cell associated bone loss, and direct catabolism on cartilage⁴¹. As noted above, IL-4 is a ligand whose biological activity is mediated through a receptor system dedicated to both IL-4 and IL-13¹⁹. The combination of less anti-inflammatory IL-4, IL-13 and IFN-γ and more pro-inflammatory IL-7 suggests that the presence of a concomitant meniscus tear generates a stronger inflammatory signal in the synovial fluid after ACL injury. This may be an opportunity for early intervention to decrease the future risk of cartilage damage and joint degeneration in patients with combined ACL and meniscus tears.

No previous studies have measured all of these molecules, except for IL-1RA. We and others have found that IL-1RA is higher in the early phase of post-ligament injury¹¹. bNGF is suspected to exhibit pro-inflammatory effects^{37; 38}, inhibits proteoglycan and collagen synthesis⁴² and perpetuates numerous other inflammatory mediators in the cycle of OA pathogenesis⁴³. These findings suggest that there is a robust pro-inflammatory reaction in the joint immediately after ACL tears. By 2 weeks, this inflammatory response begins to decrease. Despite a lack of consensus on the classification of time from injury, other studies support the conclusion that levels of various molecules in ACL tears synovial fluid are higher in samples from injured knees closer to injury compared with knees further out from injury^{12; 44}. The association of SDF-1α, PLGF-1, HGF, and BDNF levels with time from injury is a novel finding. There is some evidence that levels of SDF-1α are associated with OA. It has been implicated in inflammation-induced hypoxia and murine collagen-induced OA by attracting leukocytes to inflamed joint⁴⁵. Although SDF-1α plays a multifaceted role in in joint disease, our data suggest that it may be an important molecule in the early response to injury. We observed significantly higher levels of PlGF-1, a member of the VEGF family. This molecule is increased in the synovial fluid of rheumatoid arthritis patients and its blockade has been shown to attenuate the progression of arthritis in mice³². Its higher levels immediately after injury likely stimulate angiogenesis as well as chemotaxis of inflammatory cells into the inflamed tissues. HGF, which enhances angiogenesis and induces synovial cell proliferation⁴⁶, was also increased closer to injury. Lastly, elevated BDNF within 15 days of injury may be associated with the mechanism of joint pain, as others have suggested a similar role in the acute stage of knee OA inflammation⁴⁷. These findings, in conjunction with previous studies on anabolic and catabolic molecules^{8; 12} and gene expression^{48; 49}, suggest that the biologic window to even consider ligament repair after ACL tear may be limited to a few days to weeks from injury. More research is needed to understand if and how the biology of the injured joint influences the potential of healing for the acute stage of inflammation few ACL tears that are candidates for repair.

In this study, we did not include measurements of the classical (bio)markers for OA in our synovial fluid samples⁵⁰. A major reason for this is that most, if not all, of these markers have been validated in serum and/or urine but their utility in synovial fluid is just beginning to emerge. When sufficient validation has been established, these molecules will likely warrant investigation. A limitation to the current study is lack of chronic ACL tear patients more than a year out from injury. However, our focus was on the early post-injury time points and it is reasonable to suspect that levels may approach a steady state further out. Nevertheless, that area likely deserves further study. Since we did not aspirate all patients presenting to the clinic with ACL tears or OA, there may be selection bias. The variable use of NSAIDs could also impact our findings. With the current sample size, it is not feasible to precisely assess the impact of this variable because it likely depends on the specific medication and dose, patient clearance, and time from last dose to aspiration, as well as the frequency of use. In this translational study, the variable use of NSAIDs reflects a clinical reality which certainly deserves further investigation in the future with larger cohorts. Similarly, the history of a previous intra-articular corticosteroid injection could impact the findings of the OA patients and could not be assessed in the current study for similar reasons. The findings in this study are limited by the fact that they come from a subset of each patient population. For that reason, the findings do not represent a comprehensive comparison of ACL tear and OA patients but establish an initial comparison between a subset of each cohort which identifies differences that deserve further study in larger, more comprehensive patient populations. Furthermore, we did not record or control for previous knee surgery although all ACL patients were primary ACL injuries, not recurrent ACL graft tears. Finally, we did not include data from healthy joints, which is challenging for technical and ethical reasons. Future research should aim to address these limitations and further investigate the downstream effects of these molecules on the potential for ligament healing and long-term joint health.

Conclusion

Our study highlights changes in cytokine, chemokine, and growth factor levels in the synovial fluid following ACL injury based on time from injury and presence/absence of concomitant meniscus tear, as well as how these levels compare to knees with end stage OA. The presence of concomitant meniscus tear amplifies the inflammatory response, potentially contributing to an increased risk of cartilage degeneration and the development of OA. These findings lay the foundation for future studies to explore whether modulating these molecules is possible and has therapeutic potential to delay or prevent joint degeneration or possibly potentiate ligament repair if indicated.