Introduction
Total hip arthroplasty (THA) is one the most successfully orthopedic interventions for severe osteoarthritis1. The revolutionary treatment for hip arthritis shows an annual increase over 5.4% and an estimated increase of 200% in demand for THA by 20302. Recently, three surgical approaches were commonly used in THA procedure, including the direct anterior, direct lateral and posterior approaches, with cost-effective treatment for pain relief, and function improvement in hip3. Although numerous studies confirm the effects of THA on joint diseases4,5, the clinical outcomes and prognosis are still unsatisfied. Thus, deeper understanding on the risk factors and seeking potential predicted biomarkers for prognosis of THA patients are of great significance.
High mobility group box protein-1 (HMGB1) is a kind of nuclear proteins binding to DNA6. Recent evidence shows the pathogenic function of HMGB1 in inflammatory diseases. As reported, cellular activation, stress, damage, or necrosis induced the release of HMGB1, which, in turn, triggered inflammatory response in cells7. HMGB1 bound to Toll-like receptors (TLRs) and the receptor for advanced glycation end products (RAGE) to activate inflammation in chronic and acute liver diseases, such as hepatocellular carcinoma, hepatic ischemia/reperfusion injury, and nonalcoholic fatty liver disease8. Recent studies suggested that increased HMGB1 was involved in the pathogenesis of arthritis. Monoclonal anti-HMGB1 partially alleviated joint destruction and obviously improved the clinical outcomes of mice with collagen type II-induced arthritis9. Elevated HMGB1 was found in inflammatory synovial tissue of RA patients and collagen-induced arthritis rats; besides, HMGB1 expression was positively correlated with TNF-α and interleukin-1b (IL-1β) at the peak10. Another study also illustrated that the progression of arthritis was accelerated by overexpressed HMGB1 but inhibited by reduced HMGB1, suggesting HMGB1 was a mediator for inflammation and a novel target for arthritis11. In addition, the close association between inflammatory factors and postoperative recovery and complications was also reported. Significantly positive relationships was observed between the peak levels of C-reactive protein (CRP) and IL-6 and complication rate after THA, suggesting inflammatory factors might be the predictors for the post-operative complications in patients underwent THA12. However, limited researches demonstrated the correlation between HMGB1 and inflammatory factors and the prognosis of THA patients.
In the present study, we demonstrated that increased HMGB1 and inflammatory factors were correlated with poor prognosis of THA patients, indicating the potential therapeutic role of HMGB1 and inflammatory factors for the clinic outcomes and prognosis of THA patients.
Methods
Patients
A total of 208 THA patients who were admitted in our hospital during January 2020 to January 2022 were included in this prospective observational study. The inclusion criteria were as follows: (1) All adult patients aged over 18 years diagnosed with secondary osteoarthritis including post-traumatic, osteonecrosis, or secondary to inflammatory arthropathy; (2) patients underwent primary THA; (3) no abnormality around the hip joint was found in the anatomy; and (4) patients without deep venous thrombosis when admission. The exclusion criteria were as follows: (1) serious pre-operative complications, such as malignant tumors, severe renal, liver, or cardiovascular diseases; (2) mental diseases; and (3) pathological hip fractures caused by tumor. This study was approved by the ethic committee of our hospital. Written informed consent was obtained from all patients.
THA surgery
All the patients received standardized operation conducted by operating surgeons as described previously13. Three approaches were chosen for THA in all patients according to special conditions, including posterior approach, direct lateral approach, and anterior approach. A standardized intraoperative protocol and routine post-operative management protocol were obeyed, including the use of tranexamic acid and local infiltration anesthesia. A X-ray examination was performed at 6 weeks after surgery.
Detection of serum HMGB1 and inflammatory factors
Venous blood samples of 5 ML were collected from all patients at admission, 1 day, 3days d, 7 days, 30 days, and 90 days after surgery. The serum HMGB1 and inflammatory factors were determined using the following ELISA kits: HMGB1 ELISA kit (#MBS451177, 62.5-4000 pg/mL), IL-1β ELISA kit (#MBS175901, 3.9 -250 pg/mL), and IL-6 ELISA kit (#MBS2701078, 7.8-500 pg/mL). The serum level of CRP was detected using a Hitachi 7600 Automatic Biochemical Analyzer (Hitachi Corporation).
Clinical outcomes and follow-up
Demographic data including sex, BMI, pre-operative, and post-operative complications were collected from the database. The levels of Harris and Fugl-Meyer were detected on admission and 90 days after surgery in all patients, and the levels of 36-item short-form health survey (SF-36) and Pittsburgh sleep quality index (PSQI) were detected on 90 days after surgery in all patients. For patients with Harris Hip Score ≥ 80 were defined as good prognosis, the others with Harris Hip Score < 80 were defined as poor prognosis. All patients were followed up for 3 months.
Data analysis
The normally distributed data were expressed as mean ± SD. Comparison for continuous data were analyzed by Student t-test. The rates were analyzed by Chi-square test. Receiver operating characteristic curve (ROC) was used for assessing the diagnostic value of HMGB1 in patients with poor prognosis. In addition, the risk factors for poor prognosis of THA patients were analyzed by logistic regression model. All the calculation and graph were conducted using SPSS 18.0 and Graphpad Prism Software 6.0.
Results
Dynamic changes and the correlation of serum HMGB1 and inflammatory factors in all THA patients
Serum HMGB1 and inflammatory factors in all patients were detected at the admission, 1 day, 3 days, 7 days, 30 days, and 90 days after surgery (Fig. 1). Compared with the admission, serum levels of HMGB1, IL-1β, and IL-6 were obviously increased on 1 day and 3 days after surgery, and then gradually declined (p < 0.05). Serum CRP showed a peak value at 1 day after surgery and then gradually declined (p < 0.05). No significant difference was observed after 30 days after the surgery. In addition, the results showed significantly positive correlation between HMGB1 and CRP on 1 day after surgery (p < 0.05, Table 1a). Positive correlations were found among HMGB1, IL-1β, and IL-6 on 3 day after surgery (p < 0.05, Table 1b). Analysis showed no obvious correlations among HMGB1, IL-1β, IL-6, and CRP on 7 day after surgery (p > 0.05, Table 1c).
HMGB1 | CRP | IL-1β | IL-6 | |
---|---|---|---|---|
HMGB1 | ||||
Person's correlation | 1 | 0.784 | −0.053 | 0.001 |
p | - | < 0.001 | 0.444 | 0.994 |
CRP | ||||
Person's correlation | 0.784 | 1 | −0.081 | −0.031 |
p | < 0.001 | - | 0.243 | 0.661 |
IL-1β | ||||
Person's correlation | −0.053 | −0.081 | 1 | 0.020 |
p | 0.444 | 0.243 | - | 0.779 |
IL-6 | ||||
Person's correlation | 0.001 | −0.031 | 0.020 | 1 |
p | 0.994 | 0.661 | 0.779 | - |
HMGB1: high level of high mobility group box 1; CRP: C-reactive protein; IL-1β: interleukin-1β; IL-6: interleukin-6.
HMGB1 | CRP | IL-1β | IL-6 | |
---|---|---|---|---|
HMGB1 | ||||
Person's correlation | 1 | 0.084 | 0.764 | 0.788 |
p | - | 0.226 | < 0.001 | < 0.001 |
CRP | ||||
Person's correlation | 0.084 | 1 | 0.059 | 0.100 |
p | 0.226 | - | 0.399 | 0.152 |
IL-1β | ||||
Person's correlation | 0.764 | 0.059 | 1 | 0.609 |
p | < 0.001 | 0.399 | - | < 0.001 |
IL-6 | ||||
Person's correlation | 0.788 | 0.100 | 0.609 | 1 |
p | < 0.001 | 0.152 | < 0.001 | - |
HMGB1: high level of high mobility group box 1; CRP: C-reactive protein; IL-1β: interleukin-1β; IL-6: interleukin-6..
HMGB1 | CRP | IL-1β | IL-6 | |
---|---|---|---|---|
HMGB1 | ||||
Person's correlation | 1 | 0.051 | 0.024 | 0.030 |
p | - | 0.446 | 0.731 | 0.668 |
CRP | ||||
Person's correlation | 0.051 | 1 | 0.027 | −0.029 |
p | 0.446 | - | 0.695 | 0.673 |
IL-1β | 1 | 0.007 | ||
Person's correlation | 0.024 | 0.027 | - | 0.923 |
p | 0.731 | 0.695 | ||
IL-6 | ||||
Person's correlation | 0.030 | −0.029 | 0.007 | 1 |
p | 0.668 | 0.673 | 0.923 | - |
HMGB1: high level of high mobility group box 1; CRP: C-reactive protein; IL-1β: interleukin-1β; IL-6: interleukin-6.
HMGB1 was correlated with post-operative complications and prognosis of THA patients
To further study the association between HMGB1 and prognosis of all THA patients, all the participants were divided into high HMGB1 group and low HMGB1 group according to the mean value of serum HMGB1 704.04 pg/mL on 3 day after surgery. Table 2 showed basic clinical characteristics of all patients in two groups. Serum levels of CRP, IL-1β, and IL-6 were significantly higher in high HMGB1 group compared with low HMGB1 group on 3 days after surgery (p < 0.05). No significant difference was found for the other basic characteristics between two groups.
Characteristics | High HMGB1 group (n = 108) | Low HMGB1 group (n = 100) | p value |
---|---|---|---|
Age, year | 62.35 ± 11.18 | 63.66 ± 10.53 | 0.383 |
Female, n (%) | 78 (72.22) | 69 (69.00) | 0.610 |
BMI, kg/m2 | 25.93 ± 3.96 | 26.58 ± 5.23 | 0.314 |
Pre-operative complications, n (%) | |||
Diabetes | 15 (13.89) | 11 (11.00) | 0.529 |
Hypertension | 16 (14.81) | 16 (16.00) | 0.813 |
Pre-operative Harris score | 34.89 ± 8.79 | 36.07 ± 10.72 | 0.388 |
Pre-operative Fugl-Meyer score | 14.92 ± 2.99 | 14.13 ± 3.80 | 0.101 |
CRP on 3 days after surgery, mg/L | 73.22 ± 19.76 | 66.29 ± 20.57 | < 0001 |
IL-1β on 3 days after surgery, pg/mL | 100.72 ± 8.76 | 85.59 ± 8.43 | < 0001 |
IL-6 on 3 days after surgery, pg/mL | 400.04 ± 37.02 | 334.22 ± 33.12 | 0.014 |
Approaches for THA, n (%) | |||
Posterolateral approach | 60 (55.56) | 59 (59.00) | 0.616 |
Direct anterior approach | 48 (44.44) | 41 (41.00) |
HMGB1: high level of high mobility group box 1; CRP: C-reactive protein; IL-1β: interleukin-1β; IL-6: interleukin-6; THA: Total hip arthroplasty; BMI: body mass index.
The association between serum HMGB1 and the prognosis of THA patients was also analyzed. The levels of Harris, Fugl-Meyer, SF-36, and PSQI were compared between high HMGB1 group and low HMGB1 group on 90 d after surgery. As shown in table 3, the levels of Harris, Fugl-Meyer, and SF-36 were in low HMGB1 group, which were significantly higher than those in high HMGB1 group (p < 0.05). The incidence of infection and deep venous thrombosis in high HMGB1 group was much higher than low HMGB1 group (p < 0.05). However, no significant difference was observed for PSQI and the incidence of nausea and vomiting in two groups. All these results indicated that low HMGB1 reduced the incidence of post-operative complications and improved prognosis of THA patients.
Variables | High HMGB1 group (n = 108) | Low HMGB1 group (n = 100) | p value |
---|---|---|---|
Harris on 90 days | 86.37 ± 6.65 | 89.72 ± 6.02 | < 0.001 |
Fugl-Meyer on 90 days | 23.85 ± 2.64 | 27.64 ± 4.23 | < 0.001 |
SF-36 on 90 days | 74.56 ± 9.42 | 83.62 ± 7.22 | < 0.001 |
PSQI on 90 days | 8.26 ± 4.45 | 7.69 ± 4.23 | 0.346 |
Post-operative complications, n (%) | |||
Nausea and vomiting | 12 (11.11) | 12 (12.00) | 0.841 |
Infection | 11 (10.19) | 2 (2.00) | 0.020 |
Deep venous thrombosis | 27 (25.00) | 4 (4.00) | < 0.001 |
Dislocation rates | 2 (1.85) | 1 (1.00) | - |
Length of stay | 9.22 ± 2.01 | 8.84 ± 1.75 | 0.146 |
HMGB1: high level of high mobility group box 1; PSQI: Pittsburgh sleep quality index.
High level of HMGB1 predicted poor prognosis for THA patients
Finally, we analyzed the diagnostic value of HMGB1 for THA patients. For all THA patients, Harris Hip score < 80 was considered as poor prognosis and Harris score ≥ 80 was considered as good prognosis. ROC curve was used to investigate the diagnostic value of HMGB1 on 3 days after surgery. The data showed that the cutoff value of HMGB1 for poor prognosis was 806.45 pg/mL, with AUC 0.926 (95% CI 0.876~0.975), sensitivity 87.50%, and specificity 82.8% (Fig. 2). Logistic regression model suggested that HMGB1 was the risk factor for poor prognosis of THA patients (odd ratio > 1, p < 0.05, Table 4). These findings illustrated that serum HMGB1 exhibited valuable diagnostic potential for poor prognosis of THA patients; besides, HMGB1 was also the risk factors for poor prognosis in THA patients.
Variables | B | SE | Wald | Sig. | OR (95%CI) |
---|---|---|---|---|---|
HMGB1 on 3 days | 0.020 | 0.010 | 4.130 | 0.042 | 1.020 (1.001~1.040) |
IL-1β on 3 days | 0.004 | 0.046 | 0.006 | 0.936 | 1.004 (0.918~1.097) |
IL-6 on 3 days | 0.021 | 0.016 | 1.590 | 0.207 | 1.021 (0.989~1.054) |
CRP on 3 days | −0.023 | 0.020 | 1.320 | 0.251 | 0.978 (0.941~1.016) |
HMGB1: high level of high mobility group box 1; CRP: C-reactive protein; IL-1β: interleukin-1β; IL-6: interleukin-6; SE: standard error; OR: odds ratio; CI: confidence interval.
Discussion
THA has been one of the major orthopedic operations widely performed in clinic practice14. THA is an ideal method for treating hip diseases, such as femoral neck fracture, femoral head necrosis, hip dysplasia, and so on15. Although THA significant improves the life quality of patients with hip diseases, the prognosis is not satisfied. Thus, new diagnostic biomarkers to predict the prognosis of THA patients are very essential. In the present research, we demonstrated that high level of HMGB1 and inflammatory factors was associated with poor prognosis in THA patients.
HMGB1 has been proven to be associated with the pathological mechanism and prognosis of patients with joint diseases. As reported, HMGB1 gene expression was increased in the articular cartilage in the mouse model of ischemic osteonecrosis and elevated HMGB1 activated IL-6 expression in the supernatants of chondrocytes, suggesting HMGB1 expression was significantly involved in the pathogenesis of Legg-Calvé-Perthes disease16. Upregulatd serum HMGB1 was found in ankylosing spondylitis patients; besides, serum HMGB1 could be identified as laboratory indicator to reflect the activity of ankylosing spondylitis17. HMGB1 inhibitors exhibited protective effect on osteoarthritis through suppressing IL-1β-induced MMPs expression through TLR4/NF-kB pathway in human articular chondrocytes-knee18. Suppression of BRD4 alleviated inflammatory response and catabolism of chondrocytes and inhibited the activation NF-kB pathway through recruiting HMGB1 in in articular cartilage of osteoarthritis19. Increasing number of evidence demonstrate the pathogenic features of HMGB1 in various diseases, suggesting the potential of HMGB1 as a new therapeutic option for rheumatoid arthritis20 and osteoarthritis21. Nevertheless, the association between HMGB1 and the prognosis of patients received THA surgery is still uncertain. Our study for the 1st time investigated the role of HMGB1 in disease progression and its correlation with the clinic outcomes of THA patients. Our data showed that lower HMGB1 reduced the incidence of post-operative complications and improved prognosis of THA patients. In addition, serum HMGB1 was a diagnostic biomarker and risk factor for poor prognosis of THA patients.
Recent data showed the relationship between inflammation and HMGB1 reported in joint diseases. A prospective study revealed obvious elevation of serum HMGB1 and white blood cells (WBC) in patients with prosthetic joint infection, suggesting a underlying correlation between HMGB1 and WBC in joint infection22. As a mediator of inflammation in rheumatoid arthritis, HMGB1 level increased with proinflammatory S100 protein and IL-8 in synovial fluid collected from patients diagnosed with the early onset juvenile idiopathic arthritis23. A previous study illustrated that cytoplasmic and extracellular HMGB1 expression elevated along with the development of collagen-induced arthritis; additionally, cytoplasmic HMGB1 expression was consistent with the level of TNF-α and IL-1β in advanced arthritis24. HMGB1 was reported to sustain chronic synovitis through several mechanisms including the upregulation of IL-1, IL-6, and TNF-α; however, extensive production of proinflammatory cytokines mediated tissue damage. Besides, a significantly correlation was found between serum HMGB1 and IL-6 in postmenopausal women with rheumatoid arthritis25. Consistent with previous studies, our findings revealed positive correlation between HMGB1 and IL-1β on 3 day after surgery, and positive correlation between HMGB1 and CRP was found on 1 day after surgery. The involvement of inflammatory factors in joint diseases was also reported in various studies. IL-1β aggravated the production of osteoclastogenic Tregs, resulting in the bone erosion in arthritis26. The increase of IL-6 and sIL-6R might increase the risk of joint destruction through upregulating VEGF expression in rheumatoid arthritis patients27. Our study also identified HMGB1, IL-1β, and IL-6 as the risk factors for poor prognosis in THA patients.
Conclusion
In summary, the present study found that the contents of HMGB1 and inflammatory factors were decreased after THA surgery and positive correlations were found between HMGB1 and inflammatory factors in patients. Besides, high level of HMGB1 increased the complication rates and predicted poor prognosis for patients with THA surgery. Our results might provide new insights for the role of HMGB1 and inflammatory factors in THA patients.