The crossroads of the hypercoagulability and patient outcomes in osteoarthritis: interactions and connections

doi:10.21203/rs.3.rs-4718192/v1

Download PDF

Research Article

The crossroads of the hypercoagulability and patient outcomes in osteoarthritis: interactions and connections

https://doi.org/10.21203/rs.3.rs-4718192/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Our study aimed to probe whether coagulation indices are linked to patient-reported outcomes (PROs) in OA.

Methods

A thorough review of the literature on OA and coagulation indices was conducted using bibliometric approaches. Clinical data were retrospectively analyzed in OA patients (7,068) and healthy controls (HC, 795). Coagulation indices—prothrombin time (PT), fibrinogen (FIB), activated partial thromboplastin time (APTT), thrombin time (TT), D-dimer, and platelet count (PLT)—as well as immune-inflammatory indices, PROs (visual analogue scale and Short Form 36), were analyzed for correlations.

Results

Co-cited literature revealed that research related to OA and coagulation indices focused on inflammation, pain, and clinical utility. The levels of PLT, TT, FIB, and D-dimer were elevated in the OA group compared to the HC group. Hypercoagulable states are present in the OA. The results of the ROC demonstrate that they can differentiate between OA and healthy individuals. Coagulation indices were strongly linked to immune-inflammatory indicators and PROs. Logistic regression analysis indicated that PLT, D-dimer, and C-reactive protein (CRP) were all predictive of disease activity. However, PLT and D-dimer combined with CRP had a superior predictive effect than CRP alone.

Conclusion

PLT and D-dimer may serve as appropriate biomarkers to correlate with OA disease activity.

osteoarthritis

coagulation index

inflammation

patient-reported outcomes

data mining

Osteoarthritis (OA) is a bone disease characterised by disturbances to cartilage, meniscus, synovium, ligaments, and subchondral bone ¹. Arthralgia, joint malformations, and soft tissue haemorrhage are the symptoms of osteoarthritis. Low-grade systemic inflammation impairs joint tissue homeostasis and exacerbates the deterioration of OA ². Clot hypercoagulability is another characteristic of systemic inflammation, and OA also exhibits this characteristic together with a diminished capacity to lyse clots, potentially as a result of actual amyloid production ³. Localised circulatory disorders in the joints affect the physiochemical microenvironment of the subchondral bone. Frequent localised joint injuries cause osteomalacia, or even erosion of the underlying bone, resulting in the joints becoming more painful, swollen, and stiff ^4–5. Therefore, it is crucial to study the progression of OA from the perspective of inflammatory and hypercoagulable states.

The significance of innate immunity is related to the emergence of synovitis, downstream inflammatory activation, and articular cartilage catabolic processes, which contribute to the evolution of OA ⁶. Long-term chronic inflammation, cytokine imbalance, and immune disorders in the body contribute to damage to the vascular endothelium, which directly or indirectly activates the coagulation-fibrinolytic system and interferes with anticoagulation, thereby causing microcirculatory disturbances, leading to systemic circulatory abnormalities in patients with OA with hypercoagulability and prothrombotic state ⁷. The immune system and the coagulation system work together to positively modify the immunological response. Coagulation factors such as thrombin stimulate immune cells and increase the synthesis of cytokines that promote inflammation ⁸. The primary element of blood clotting, fibrin, serves as a scaffold for immune cells and encourages their recruitment and activation at the site of infection or injury. Additionally crucial to OA are platelets, which are involved in vascular integrity, hemostasis, and inflammatory processes ⁹. Pro-inflammatory and anti-inflammatory factors—interleukin (IL)-6, IL-1β, tumour necrosis factor (TNF)-α, IL-10, and IL-4—are associated with the development of angiogenesis and chemotaxis in OA. The majority are created, activated, adhered to, or expelled by platelets ¹⁰.

Patient-reported outcomes (PROs) are information derived directly from patients' reports of their own health status, functional status, and perceptions of treatment ¹¹. The Visual Analog Scale (VAS) is widely employed to assess a variety of health-related issues, including pain, mood, and quality of life, and has been recognized as providing robust and valid data ¹². An additional indicator of medical-related quality of life is the MOS item short from the health survey (SF-36) ¹³. Bibliometrics is an instrument for applying mathematical and statistical methods to rapidly grasp the hottest research topics and trends in a particular field ¹⁴.

This research was conducted to understand the current trends in research related to coagulation and hypercoagulation in patients with OA through bibliometrics. We explore outbreaks of keywords related to coagulation in OA and visualise, evaluate, and predict research status, hotspots, and future trends. Furthermore, we investigated the involvement of coagulation indices—prothrombin time (PT), fibrinogen (FIB), activated partial thromboplastin time (APTT), thrombin time (TT), D-dimer, and platelet count (PLT) in the pathophysiology of OA. The purpose of our research was to verify whether coagulation indices are related to immune-inflammation and whether they correlate with patient-reported outcomes and disease activity status (as defined by VAS).

2.1 Data

2.1.1 Literature sources

The Web of Science (WOS) Core Collection database is an excellent source of science statistics and evaluation ¹⁵. The search terms were as follows: TS = (Osteoarthritides OR Osteoarthrosis OR Osteoarthroses OR (Osteoarthrosis Deformans) OR Osteoarthritis) AND TS = (hypercoagulable state OR hypercoagulability OR platelet OR blood platelet OR soterocyte OR fibrinogen OR D-dimer). Only English-language articles were chosen among numerous forms of publication. Each investigator independently reviewed all the articles to ensure their relevance. All articles' entire records were searched up to December 20, 2023. In total, 1368 articles were analyzed. Table 1 illustrates the results of the complete screening.

Table 1

TS search quires and refinement procedure.
Set	Results	Refinement
1	2,184	TS= (Osteoarthritides OR Osteoarthrosis OR Osteoarthroses OR (Osteoarthrosis Deformans) OR osteoarthritis) AND TS= (hypercoagulable state OR hypercoagulability OR platelet OR blood platelet OR soterocyte OR Fibrinogen OR D-dimer) Indexes = SCI-EXPANDED
2	2,152	Refined by LANGUAGES: (English)
3	1,368	Refined by DOCUMENT TYPES: (Articles)

2.1.2 Sources of clinical data

This research is an observational, retrospective, single-center study. We enrolled OA patients who were hospitalized between March 2009 and February 2023 at the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine. General information about the patients was retrieved from the hospital case information management system, including basic information such as body mass index (BMI), gender, comorbidities, and length of hospital stay (LOS). Exclusion criteria include: incomplete data; comorbid infections; severe circulatory, respiratory, and hematopoietic diseases; and other autoimmune diseases. The Charlson Comorbidity Index (CCI) ¹⁶ is a quantification of patient comorbidities for the purpose of predicting the risk of death from disease. A total of 7,068 osteoarthritis patients took part in the study. On the other side, 795 healthy control group volunteers (HC) were chosen from those who had regular medical checkups at the hospital's medical checkup center. It was not necessary to get informed consent for this retrospective observational investigation. Subjects' personal information was hidden before any analysis. This study was approved by the Ethics Research Committee of the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine (2023AH-52).

2.2 Methodology

2.2.1 Bibliometric Visualization Analysis

CiteSpace version 6.2.6 was implemented for the bibliometric analysis study, which examined essential pathways and knowledge inflection points in the advancement of topic areas ¹⁷. In our study, CiteSpace software was adopted to perform visualisation analysis, network analysis, cluster analysis, timeline analysis, and outbreak word detection for reference analysis and keywords. The specific parameters used are as follows: set the time range as January 2007–December 2023; set the time slice as 1, node type to select references and keywords, respectively; and choose the top 50 keywords with the highest frequency of occurrence in each time slice by setting Top N to 50. Paths under and under the pruned slice network in the pruned model are chosen by pruning, and the remaining parameters are default values.

2.2.2 Collection of coagulation indexes and immuno-inflammatory indexes

(1) Coagulation indexes: prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (FIB), thrombin time (TT), platelet count (PLT) and D-dimer. (2) Immunologic indexes: immunoglobulin (IG) A, G, M, complement 3 (C3), complement 4 (C4). (3) Inflammation indicators: C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR).

2.2.3 Patient-reported outcomes (PROs)

One of the most prevalent unidimensional methods for measuring pain intensity is the VAS, which is extensively utilised in rheumatic diseases ¹⁸. The scale is a 10-cm straight line, where "no pain at all" is indicated at one end and "pain to the extreme" at the other. Eight health-related quality of life questionnaires make up the SF-36 scale: physical function (PF), bodily pain (BP), role-physical (RP), vitality (VT), social function (SF), general health (GH), role-emotional (RE), and mental health (MH)¹⁹.

2.2.4 Association Rules

Apriori is the most impactful algorithm for mining the frequent item sets of Boolean association rules. An association rule is an implication of the form XY, where X and Y are called the prior (antecedent or left-hand-side, LHS) and the successor (consequent or right-hand-side, RHS) of the association rule, respectively. N denotes the total number of items in the sample, and X and Y represent an item set, respectively. The corresponding equations for support, confidence, and elevation are as follows:

2.2.5 Statistical analysis

For data processing, the SPSS 22.0 and R software were adopted. The Kolmogorov-Smirnov with Levene test was applied to test the data for normality and chi-square. Measures with a normal distribution were reported as mean and standard deviation, while measures with a non-normal distribution were displayed as median (P25, P75), and count data were displayed as n/%. To determine statistical significance, the Mann-Whitney test, Wilcoxon signed rank test, t-test, or chi-square test (as appropriate) were utilized to determine statistical significance. Correlations between coagulation indices and other indices were analyzed by Spearman or Pearson analysis (as appropriate).

To compare the diagnostic efficacy of coagulation indicators in OA, the receiver operator characteristic curve (ROC) of the subjects was displayed, and the area under the curve (AUC) was computed. Jorden's test was applied to determine the optimal critical value. Logistic regression models were constructed to assess the relationship between coagulation indices and immune-inflammatory indicators, as well as patient reports. In the multivariate logistic regression analysis, variables that were deemed significant in the univariate analysis were incorporated and modelled using stepwise regression. The absolute values of the t-statistic for each model parameter was determined to compare the relative weight of each predictor in the logistic regression model. McFadden's R₂ and Hosmer-Lemeshow goodness-of-fit tests were utilized to evaluate the logistic regression model. McFadden's R₂ values between 0.2 and 0.4 were deemed to indicate a good fit of the model ²⁰. The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were applied to evaluate the quality of fit and complexity of various models and to select the best-balanced model ²¹.

3.1 Trends in the issuance of papers

A total of 1,368 literary collections were included. The number of publications and citations in a particular period can objectively and quantitatively reflect general trends in a region. Annual number of publications issued and citations related to OA and coagulation metrics are depicted in Fig. 1A. The first peak was reached in 2020 with a total of 182 publications. Overall, there is a steady upward trend in annual publications in this area. In Fig. 1B, a fitted curve showing the overall annual growth trend is presented. Based on the time curve, the global cumulative number of publications in the field is expected to continue to rise over the next 10 years (R₂ = 0.922) (Fig. 1B). Furthermore, the citations of these publications showed an increasing trend year by year (Fig. 1A), with a total of 16,970 citations (15,873 after the self-citation), an average of 24.42 citations, and an overall h-index of 82. These data demonstrate the overall fluctuating upward trend of OA coagulation index-related studies during the last 16 years.

3.2 Analysis of the co-cited literature

The topics strongly related to the field of OA-coagulation index were highlighted by evaluating the co-citation network of 1,368 documents. A total of 244 documents (with at least 20 citations) were chosen to map the co-citation analysis network. Cluster analysis categorizes closely related topics to obtain hotspots in this research area. It forms clusters of varying strength based on the co-occurrence links of terms, with significant homophily among terms. Literature clusters were classified using different node colors (Fig. 2A). The top 11 hot topics were presented for clustering. The main research hot topics were osteoarthritis, platelet-rich plasma, intra-articular injection, clinical outcomes, growth factors, clinical efficacy, real-world evidence, pain, systematic reviews, platelet amplification, inflammation in arthritis, and efficacy. Most of the studies focused on OA inflammatory mediators, patient-reported outcome indicators, and blood derivatives.

The articles with a higher citation frequency can be regarded as more influential in a particular field. Figure 2B presents the details of the top 20 articles, with the blue line indicating the time periods and the red line indicating the strong citation frequency. The research by Kon, E. et al. ²² (26.9) gained the most citations over a 7-year period and had the strongest citation burst value. This research was published in 2,009 under the title "Platelet-rich plasma: intra-articular knee injections produced favorable results on degenerative cartilage plasma: intra-articular knee injections produced favorable results on degenerative cartilage lesions. Natural concentrates of autologous growth factors in the blood may stimulate cartilage anabolism and reduce catabolic processes. Platelet-rich plasma (PRP) influences whole joint homeostasis, decreases synovial hyperplasia, and ultimately boosts knee function, discomfort, and quality of life in OA patients. The citation outbreak published in “The American Journal of Sports Medicine” by van Buul GM et al. took the longest time ²³. This research demonstrates that platelet-rich plasma release reduces the inflammatory effects of IL-1β on human osteoarthritic chondrocytes, including suppression of NF-B activation.

This study demonstrates that platelet-rich plasma release attenuates multiple inflammatory effects mediated by IL-1β on human osteoarthritic chondrocytes, including inhibition of NF-κB activation. Notably, an explosion of articles in recent years, such as Dai X et al. ²⁴ and Cheng Y et al. ²⁵, emphasized blood hypercoagulation as a susceptibility factor for deep vein thrombosis and as a cause of osteoarthritis total knee arthroplasty. According to Liu N et al. ²⁶, blood rheology alterations and hypercoagulability contributed to the development of petechiae following total knee arthroplasty.

3.3 Analysis of co-cited keywords

Keywords are a high degree of summarization and condensation of the article's theme, as high-frequency keywords in different periods reflect the transformation of research hotspots in the field. A total of 458 keywords were identified, and those related to osteoarthritis-coagulation index were displayed in the network diagram (Fig. 3A). The top 10 high-frequency keywords included osteoarthritis, pain, risk factors, clinical trials, growth factors, platelet-rich plasma (PRP), prevalence, randomized controlled trials, growth factor beta, and chondrogenic differentiation. The term "emergent words" refers to keywords that have been widely cited over a period of time. Among them, the longest burst time is "inflammation" (2011–2018), and the strongest burst intensity is "pain" (16.55) (Fig. 3B). Meanwhile, we noticed that regenerative medicine, venous thromboembolism, and M₂ macrophage polarization were the most prevalent keywords in the last 3 years (Fig. 3B). On the basis of the keyword co-occurrence network, further keyword clustering analysis was generated (Fig. 3C). The results were co-clustered into 10 major categories, including inflammatory processes, osteoarthritis, injured articular cartilage, clinical utility, induced apoptosis, M₂ macrophage polarization, mesenchymal stem cell therapy, pain, controlled trial, and coagulation.

3.4 Results of coagulation index analysis

Age, gender, and BMI did not significantly differ between OA patients and healthy controls. Compared with the healthy group, PLT, TT, FIB, and D-D indicators were elevated in OA patients. Immuno-inflammatory markers (such as ferritin, IGA, IGG, CRP, and ESR) were significantly elevated in OA patients than in the HC group (Table 2).

Table 2

Baseline characteristics, coagulation indices, immunoinflammatory markers and PROs.
Variables		OA(n = 7,068)	HC(n = 795)	p-value
Baseline characteristics	Age(year)	61.48 ± 11.97	59.792 ± 8.74	0.132
	Male (%)	1,887(26.70)	224(28.18)	0.231
	BMI (kg/m²)	24.66 ± 2.07	24.52 ± 2.03	0.321
	LOS (years)	13.00(9.84,16.79)	NA	/
	CCI	1.15 ± 1.01	NA	/
Coagulation indicator	PLT (×10^9/L)	382(285,513)	108(89,118)	< 0.001
	PT (s)	9.81(9.00,11.9)	11.30(11.90,12.90)	0.096
	APTT (s)	22.60(20.30,30.41)	27.71(26.21,30.31)	0.064
	TT (s)	22.9(16.2,28.6)	16.1(15.1,17.7)	0.045
	FIB (g/L)	4.32(3.36,4.96)	2.32(1.09,2.79)	0.031
	D-dimer (mg/L)	0.9(0.42,2.61)	0.19(0.11,0.26)	< 0.001
Immune-inflammatory markers	CRP (mg/L)	11.64(7.78,16.59)	0.09(0.05,0.13)	< 0.001
	ESR (mm/h)	57(40,64)	5(1, 9)	< 0.001
	C3(g/L))	1.09(1.01,1.51)	0.86(0.78,0.90)	0.245
	C4(g/L)	0.26(0.23,0.41)	0.17(0.14,0.22)	0.321
	IGA(g/L)	1.22(1.21,2.57)	0.93(0.79,1.04)	0.032
	IGG(g/L)	10.43(9.13,15.28)	6.72(5.10,7.11)	0.012
	IGM(g/L)	0.56(1.79,1.07)	0.39(0.31,0.44)	0.056
	Ferritin (pg/mL)	179.20(141.08,231.81)	54.2(8.96,71.24)	< 0.001
PROs	VAS (cm)	6.0(5.5,7.5)	/	/
	PF (score)	30(10, 40)	/	/
	RP (score)	15(10, 35)	/	/
	BP (score)	28(21, 42)	/	/
	GH (score)	30(20, 40)
	VT (score)	35(25, 42)	/	/
	SF (score)	37.5(25,50)	/	/
	RE (score)	15(12,33.33)	/	/
	MH (score)	48(38,52)	/	/
Note: BMI: body mass index; LOS: length of hospital stay; CCI: charlson comorbidity index; PLT: platelet count; PT: prothrombin time; APTT: activated partial thromboplastin time; TT: thrombin time; FIB: fibrinogen; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate; C3: complement component 3; C4: complement component 4; IG: immunoglobulin; VAS: visual analog scale; PF: physical function; RP: role-physical; BP: bodily pain; GH: general health; VT: vitality; SF: social function; RE: role-emotional; MH: mental health.

3.5 ROC curve of coagulation indices

ROC curve analysis suggested (Fig. 4) that PLT, TT, FIB, and D-dimer had excellent specificity in differentiating between OA patients and healthy individuals. The best cut-off values of 350 and 0.55 for PLT and D-dimer had the highest sensitivity and specificity among these indices, with AUC (95% CI) of 0.968 (0.907–0.994) and 0.918 (0.841–0.965), respectively. The cut-off values were 20.8 for TT and 4.00 for FIB.

3.6 Correlation analysis between coagulation indices and immuno-inflammatory indexes, PROs

The consequences of the correlation analysis showed that PLT, TT, FIB and D-dimer were positively correlated with the immuno-inflammatory index and negatively correlated with the PROs; PT, and APTT are the reverse. The findings exhibited a strong correlation between the immune-inflammatory indices, PROs, and the coagulation indexes of OA patients (Table 3, Fig. 5).

Table 3

Correlation analysis.
Variables	PLT		PT		APTT		TT		FIB		D-dimer
Variables	r	p	r	p	r	p	r	p	r	p	r	p
Baseline characteristics
Age	0.011	0.564	-0.012	0.542	-0.013	0.519	0.032	0.102	0.012	0.552	0.011	0.568
LOS	0.014	0.988	-0.011	0.561	-0.011	0.608	0.026	0.176	0.002	0.990	-0.002	0.923
BMI	0.011	0.572	-0.163	0.001	-0.662	0.001	0.012	0.541	0.011	0.563	0.006	0.752
Immune-inflammatory markers
CRP	0.423	< 0.001	0.008	0.694	0.005	0.779	0.543	< 0.001	0.423	< 0.001	0.431	< 0.001
ESR	0.827	< 0.001	-0.795	< 0.001	-0.796	< 0.001	0.151	< 0.001	0.128	< 0.001	0.621	< 0.001
C3	0.334	< 0.001	-0.129	< 0.001	-0.129	< 0.001	0.326	< 0.001	0.267	< 0.001	0.343	< 0.001
C4	0.346	< 0.001	-0.134	< 0.001	-0.131	< 0.001	0.326	< 0.001	0.267	< 0.001	0.343	< 0.001
IGA	0.129	< 0.001	-0.939	< 0.001	-1.001	< 0.001	0.158	< 0.001	0.133	< 0.001	0.132	< 0.001
IGG	0.129	< 0.001	-0.913	< 0.001	-1.003	< 0.001	0.158	< 0.001	0.133	< 0.001	0.132	< 0.001
IGM	0.129	< 0.001	-0.923	< 0.001	-1.003	< 0.001	0.158	< 0.001	0.133	< 0.001	0.132	< 0.001
Ferritin	0.029	0.301	-0.156	< 0.001	-0.155	< 0.001	0.021	0.307	0.019	0.319	0.013	0.515
PROs
VAS	0.134	< 0.001	-0.931	< 0.001	-0.930	< 0.001	0.142	< 0.001	0.135	< 0.001	0.119	< 0.001
PF	-0.189	< 0.001	0.562	< 0.001	0.562	< 0.001	-0.180	< 0.001	-0.110	< 0.001	-0.183	< 0.001
RP	-0.163	< 0.001	0.824	< 0.001	0.827	< 0.001	-0.185	< 0.001	-0.111	< 0.001	-0.188	< 0.001
BP	-0.134	< 0.001	0.468	< 0.001	0.469	< 0.001	-0.163	0.001	-0.165	0.001	-0.161	0.002
GH	-0.133	< 0.001	0.686	< 0.001	0.685	< 0.001	-0.190	< 0.001	-0.134	< 0.001	-0.189	< 0.001
VT	-0.184	< 0.001	0.474	< 0.001	0.472	< 0.001	-0.174	< 0.001	-0.164	< 0.001	-0.173	< 0.001
SF	-0.175	< 0.001	0.374	< 0.001	0.375	< 0.001	-0.169	< 0.001	-0.175	< 0.001	-0.167	0.001
RE	-0.188	< 0.001	0.527	< 0.001	0.529	< 0.001	-0.154	0.005	-0.188	< 0.001	-0.176	< 0.001
MH	-0.014	0.618	-0.115	< 0.001	-0.116	< 0.001	-0.021	0.272	-0.01	0.622	-0.009	0.647
Note: LOS: length of hospital stay; BMI: body mass index; PLT: platelet count; PT: prothrombin time; APTT: activated partial thromboplastin time; TT: thrombin time; FIB: fibrinogen; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate; C3: complement component 3; C4: complement component 4; IG: immunoglobulin; VAS: visual analog scale; PF: physical function; RP: role-physical; BP: bodily pain; GH: general health; VT: vitality; SF: social function; RE: role-emotional; MH: mental health.

3.7 Association rule analysis

We evaluated the association between the coagulation index and the immuno-inflammation index using the optimal cut-off value established by the ROC curve, where the minimum support was set at 70% and the minimum confidence interval was used at 50%. The pre- and post-items with the highest confidence level were chosen for each item. The results revealed that PLT, FIB, D-dimer, and APTT had a higher association with CRP. The PT and TT had a higher correlation with ESR. The lift was greater than 1 (Table 4, Fig. 6).

Table 4

Correlation analysis between coagulation and immunoinflammatory markers.
Items (LHS ⇒ RHS)	Support (%)	Confidence (%)	Lift
PLT⇒ CRP	45.814	88.572	1.083
FIB⇒ CRP	44.539	86.875	1.005
D-dimer⇒ CRP	45.422	81.229	1.001
TT⇒ESR	29.686	82.684	1.003
APTT⇒CRP	23.341	70.748	1.024
PT⇒ ESR	25.814	70.274	1.019
Note: PLT: platelet count; PT: prothrombin time; APTT: activated partial thromboplastin time; TT: thrombin time; FIB: fibrinogen; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate.

3.8 Univariate and multivariate regression analysis

The optimal cutoff values for PLT, FBG, PT, APTT, D-dimer, and TT were determined from the ROC curves, and the critical values for the other variables were determined from the median values. Logistic regression models were performed to describe the associations of PLT, FIB, PT, APTT, D-dimer, and TT with immunoinflammatory indexes and PROs (using VAS ≥ 6 as a criterion for subgrouping). In univariate regression analysis, PLT (OR = 1.275, p < 0.001), FIB (OR = 1.667, p = 0.024), and D-dimer (OR = 2.346, p < 0.001) were independent risk factors for disease activity in OA. The ESR (OR = 2.326, p < 0.001) and CRP (OR = 2.312, p < 0.001) were markedly correlated with OA disease activity (Table 5).

Stronger correlations were observed in the multivariate logistic regression analysis, which controlled for sex, age, duration of illness, and BMI. The PLT (OR = 1.874, p = 0.015), D-dimer (OR = 1.534, p = 0.003) and CRP (OR = 1.456, p = 0.021) remained independent predictors of OA disease activity. In the multivariate model, TT (OR = 1.115, p = 0.078), FIB (OR = 1.431, p = 0.081), and ESR (OR = 2.105, p = 0.088) were not able to serve as independent predictors of disease activity (Table 6). The PLT, D-dimer and CRP were included as variables of interest in the multivariate simplified model. To determine whether PLT and D-dimer were linked to disease activity when CRP was being considered, we tested PLT and D-dimer while correcting for CRP. When CRP was considered, PLT (+ CRP) (OR = 2.423, p = 0.001) and D-dimer (+ CRP) (OR = 2.987, p = 0.003) remained significant predictors of disease activity, in addition to the afore-mentioned control variables. AIC and BIC indicate that models containing PLT, D-dimer, and CRP outperform models containing CRP only.

All independent risk factors screened by regression analysis were integrated to construct a nomogram (Fig. 7C). Each variable was assigned a score on the rating scale. The overall score was calculated by adding the scores of all variables (Fig. 7C). C-index values were applied to assess the discriminatory nature of the nomogram for survival prediction in the model (Fig. 7A). The internal validation cohort proved to have good discriminatory performance. The Hosmer-Lemeshow test findings proved that the calibration curves in the dataset once again performed well (p = 0.325) (Fig. 7A). In addition, DCA demonstrated that patients could obtain a satisfactory net benefit from the predictive model. A wide range of high-risk thresholds were observed in DCA (Fig. 7B). In conclusion, the evaluation of the nomogram based on column-line graphs revealed considerable discriminatory and calibrating power of our research model.

Table 5 Univariate logistic regression analysis

Note

BMI: body mass index; LOS: length of hospital stay; PLT: platelet count; PT: prothrombin time; APTT: activated partial thromboplastin time; TT: thrombin time; FIB: fibrinogen; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate; C3: complement component 3; C4: complement component 4; IG: immunoglobulin.

Table 6

Multi-factor logistic regression analysis (as defined by VAS ≥ 6)
Laboratory variable (critical value)	Odds ratio (95% CI)	t-stat	p-value	AIC	BIC
PLT (350)	1.874(1.056–4.134)	21.567	0.015	482.46	356.73
TT (20.8)	1.115(0.875–3.311)	9.785	0.078	Not performed	Not performed
FIB (4.00)	1.431(0.761–2.145)	6.643	0.081	Not performed	Not performed
D-dimer (0.55)	1.534(1.221–3.667)	22.135	0.003	456.75	342.76
ESR (57)	2.105(1.013–3.245)	8.621	0.088	Not performed	Not performed
CRP (11.64)	1.456(1.005–3.245)	20.754	0.021	355.78	267.75
PLT (+ CRP)	2.423(1.116–4.067)	30.234	0.001	211.345	135.643
D-dimer (+ CRP)	2.987(1.004–4.157)	26.567	0.003	222.451	115.689
Note: PLT: platelet count; TT: thrombin time; FIB: fibrinogen; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate.

Bibliometrics, through the adoption of mathematical and statistical techniques, aims to explore the distribution pattern, quantitative relationship and changing law of literature information, which can guide not only research design but also clinical practice ²⁷. We included a total of 1,368 articles of closely related literature in the field of OA -coagulation index. The number of publications has been on an upward trend in the last decade or so. These results suggest that there is a growing interest among researchers in the field of OA-coagulation metrics. Highly cited literature has focused on studies of inflammation, platelet derivatives, and patient-reported OA. With the development of regenerative medicine, platelet derivatives, such as platelet-rich plasma, have gained increasing attention for use in osteoarthritis ²⁸. Several prospective cohort studies have demonstrated that intra-articular administration of PRP will alleviate pain and enhance joint function in patients with OA ^29–30. In addition to facilitating coagulation, platelets constitute a rich source of cytokines and growth factors that are necessary for bone mineralization and soft tissue repair ³¹. The keywords associated with the OA -coagulation index were pain, risk factors, clinical trials, growth factors, and platelet-rich plasma. Cluster analysis of the keywords also focused on inflammation and coagulation. Consequently, it is reasonable to investigate the significance of markers associated with clotting in OA.

The hypercoagulable state involves immune system activation ⁸. The inflammatory response and alterations in the coagulation system correlate with the severity of osteoarthritis ³². Relevant mediators of the inflammatory process and inflammatory regulation include thrombin, fibrinogen, coagulation factor XIII, and factors of the fibrinolytic system, among many other hemostatic system components. During inflammation, cytokines may facilitate thrombosis by regulating the coagulation and fibrinolytic systems ⁷. The following cytokines are involved in the coagulation process: TNF-α (which increases platelet activation and aggregation), IL-8 (which activates neutrophils and releases coagulation-promoting factors), IL-6 (which produces fibrinogen to promote coagulation), and IL-1 (which upregulates the expression of endothelial cells and monocytes, leading to an increased procoagulant state) ^32–34. The immune cells' (such as neutrophils and macrophages) activation of platelets and coagulation factors is another essential component of immune cell-mediated coagulation. These immune cells secrete chemicals that promote inflammation, such as tissue factors, chemokines, and cytokines, which can start the coagulation cascade ^35–36. An increased platelet count is also a major cause of hypercoagulable states ³⁷. D-dimer is the product of fibrinolytic enzyme-mediated degradation of cross-linked fibrin clots ^38–39 and has been proven to be a sensitive marker for assessing disseminated intravascular coagulation (DIC). In conclusion, the interaction between coagulation and the immune system is of great importance for clinical practice and research ⁸. Coagulation factors have the power to stimulate immune cells and increase cytokine production, which promotes inflammation. At the site of damage, fibronectin serves as a scaffold for immune cells, facilitating their recruitment and activation.

Additionally, compared to healthy volunteers, OA patients had higher levels of coagulation (PLT, TT, FIB, and D-dimer) as well as immunoinflammatory indexes (ESR, CRP, IGA, IGG, and ferritin), indicating that both hypercoagulation and inflammation are involved in disease onset and progression. In addition, PLT, TT, FIB, and D-dimer have excellent discriminative capabilities and diagnostic value in OA. The PLT, PT, APTT, TT, FIB, and D-dimer correlated with several immuno-inflammatory indexes (CRP, ESR, C3, C4, IGG, IGA, IGM, and ferritin), indicating that coagulation indices correlated with immuno-inflammatory indexes. In addition to this, the coagulation index correlated with PROs (VAS score and SF-36 subscales), contributing to the differentiation of OA disease activity. The VAS and SF-36 have been proven to be valuable in measuring overall health status, disease activity, and self-perceived disease severity in patients with OA ⁴⁰. Association rules are designed to uncover interesting and frequently occurring patterns and associations between item sets of data. Similarly, association rule results again support the high correlation between coagulation index, CRP, and ESR. C-reactive protein levels rise dramatically during inflammatory processes in vivo ⁴¹. Separation of the red blood cells from the plasma is measured by the erythrocyte sedimentation rate. Red blood cells stick to each other during inflammation because the blood contains high levels of fibrinogen. "Rouleaux" are stacks of erythrocytes that settle more quickly ⁴². These results indicate that coagulation-related indices in OA are involved in the inflammatory response.

Following that, we identified additional parameters linked with VAS disease activity, which we analyzed by univariate regression analyses of PLT (OR = 1.275, p < 0.001), FIB (OR = 1.667, p = 0.024), D-dimer (OR = 2.346, p < 0.001), ESR (OR = 2.326, p < 0.001) and CRP (OR = 2.312, p < 0.001) were identified as independent risk factors for OA disease activity. In multivariate logistic regression, only PLT (OR = 1.874, p = 0.015), D-dimer (OR = 1.534, p = 0.003) and CRP (OR = 1.456, p = 0.021) remained independent predictors of OA disease activity. Based on the results of the AIC and BIC evaluations, we observed that the addition of PLT and D-dimer to the model controlling CRP increased the OR while decreasing the AIC and BIC, implying that the model is more accurate in identifying disease activity. Interestingly, our results imply that PLT and D-dimer may improve, rather than replicate, existing models for evaluating disease activity. The results further confirm the strong clinical utility and high benefits of the model by introducing nomogram plots, clinical decision curves, and correction curves.

The present study has a variety of strengths. First, it is a literature-based study with a large sample size and credible outcomes. Second, this is the first realistic study to confirm the diagnostic and prognostic utility of coagulation indices (PLT and D-dimer) in OA patients. However, our study does have a few drawbacks. First, since this was an observational study with single-center retrospective data and lacked longitudinal monitoring, there may have been selection bias. Second, because our study was limited to real-world patients, we were not able to examine or rule out the effect of treatment on the clotting index, which could have led to confounding biases. Therefore, further studies involving multiple centers will provide a larger sample to validate the findings.

Funding Statement

This work was supported by the Anhui Famous Traditional Chinese Medicine Liu Jian Studio Construction Project (Traditional Chinese Medicine Development Secret[2018]No.11),Anhui Province Traditional Chinese Medicine Leading Talent Project(Traditional Chinese Medicine Development Secret[2018] No. 23), High-level Chinese Medicine Key Discipline Construction Project-Traditional Chinese Medicine Bi Disease Study(zyyzdxk-2023100), Anhui Provincial Laboratory of Applied Foundation and Development of Internal Medicine of Modern Traditional Chinese Medicine (2016080503B041), Anhui University Natural Science Major Project (2023AH040112), Young Talents Training Project - "Xinglin Qingxiu Cultivation Plan" (0500-48-65) and the Key Projects of Scientific Research Projects of Higher Education Institutions in Anhui Province (Natural Sciences) (No. 2022AH050449).

Author Contributions

All authors reviewed the results and approved the final version of the manuscript.

Qiao Zhou: Data curation, Methodology, Software, Validation, Visualization, Writing-original draft, Writing- review & editing, Jian Liu: Conceptualization, Writing-review & editing; Yan Zhu: Conceptualization, Data curation, Methodology, Validation, Visualization, Writing-review & editing; Guizhen Wang: Conceptualization, Data curation, Methodology, Validation, Visualization, Writing-review & editing, Jinchen Guo: Data curation, Methodology, Validation, Visualization, Writing-review & editing.

Data availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics statement

The research protocol and procedures have been approved by the Ethics Committee of the First Affiliated Hospital of Anhui University of Chinese Medicine (Review No. 2023AH-52).

Acknowledgement

We thank all the authors for their contributions to this study and the Foundation for its financial support of this project.

Conflicts of Interest

The authors declare that they have no conflicts of interest to report regarding the present study.

Sanchez-Lopez E, Coras R, Torres A, Lane NE, Guma M. Synovial inflammation in osteoarthritis progression. Nat Rev Rheumatol. 2022;18(5):258–275. doi:10.1038/s41584-022-00749-9
Chow YY, Chin KY. The Role of Inflammation in the Pathogenesis of Osteoarthritis. Mediators Inflamm. 2020; 3;2020:8293921. doi: 10.1155/2020/8293921
Pretorius E, Mbotwe S, Bester J, Robinson CJ, Kell DB. Acute induction of anomalous and amyloidogenic blood clotting by molecular amplification of highly substoichiometric levels of bacterial lipopolysaccharide. J R Soc Interface. 2016;13(122):20160539. doi:10.1098/rsif.2016.0539
Ramasamy SK, Kusumbe AP, Schiller M, et al. Blood flow controls bone vascular function and osteogenesis. Nat Commun. (2016) 7:13601. doi:10.1038/ncomms13601
Aaron RK, Racine JR, Voisinet A, Evangelista P, Dyke JP. Subchondral bone circulation in osteoarthritis of the human knee. Osteoarthritis Cartilage. 2018;26(7):940–944. doi:10.1016/j.joca.2018.04.003
Liu-Bryan R. Synovium and the innate inflammatory network in osteoarthritis progression. Curr Rheumatol Rep. 2013;15(5):323. doi: 10.1007/s11926-013-0323-5.
Wang Y, Xie J, Pei F. Plasma D-dimer and FDP are promising biomarkers to predict perioperative fibrinolysis and bleeding following primary total joint arthroplasty: A STROBE compliant article. Medicine (Baltimore). 2021;21;100(20):e26058. doi: 10.1097/MD.0000000000026058
Wilhelm G, Mertowska P, Mertowski S, Przysucha A, Strużyna J, Grywalska E, Torres K. The Crossroads of the Coagulation System and the Immune System: Interactions and Connections. Int J Mol Sci. 2023;8;24(16):12563. doi: 10.3390/ijms241612563.
Kwon YJ, Koh IH, Chung K, Lee YJ, Kim HS. Association between platelet count and osteoarthritis in women older than 50 years. Ther Adv Musculoskelet Dis. 2020; 12:1759720X20912861. doi:10.1177/1759720X20912861
Tramś E, Malesa K, Pomianowski S, Kamiński R. Role of Platelets in Osteoarthritis-Updated Systematic Review and Meta-Analysis on the Role of Platelet-Rich Plasma in Osteoarthritis. Cells. 2022;23;11(7):1080. doi: 10.3390/cells11071080.
Aiyegbusi OL, Roydhouse J, Rivera SC, Kamudoni P, Schache P, Wilson R, Stephens R, Calvert M. Key considerations to reduce or address respondent burden in patient-reported outcome (PRO) data collection. Nat Commun. 2022;12;13(1):6026. doi: 10.1038/s41467-022-33826-4.
Turnbull A, Sculley D, Escalona-Marfil C, et al. Comparison of a Mobile Health Electronic Visual Analog Scale App With a Traditional Paper Visual Analog Scale for Pain Evaluation: Cross-Sectional Observational Study. J Med Internet Res. 2020;22(9):e18284. doi:10.2196/18284
Enomoto H, Fujikoshi S, Ogawa K, Tsuji T, Tanaka S. Relationship Between Pain Reduction and Improvement in Health-Related Quality of Life in Patients with Knee Pain Due to Osteoarthritis Receiving Duloxetine: Exploratory Post Hoc Analysis of a Japanese Phase 3 Randomized Study. J Pain Res. 2020;20;13:181–191. doi: 10.2147/JPR.S211072
Ninkov A, Frank JR, Maggio LA. Bibliometrics: Methods for studying academic publishing. Perspect Med Educ.2022;11(3):173–176. doi: 10.1007/s40037-021-00695-4
Wang L, Feng W, Duan J, Liang J. Pharmacovigilance Bibliometrics: Visualizing Thematic Development in the Category of Pharmacology and Pharmacy in Web of Science. Front Pharmacol. 2021;4;12:731757. doi: 10.3389/fphar.2021.731757
Charlson ME, Wells MT. Comorbidity: From a Confounder in Longitudinal Clinical Research to the Main Issue in Population Management. Psychother Psychosom. 2022;91(3):145–151. doi: 10.1159/000521952
Yuan X, Lai Y. Bibliometric and visualized analysis of elite controllers based on CiteSpace: landscapes, hotspots, and frontiers. Front Cell Infect Microbiol.2023;12;13:1147265. doi: 10.3389/fcimb.2023.1147265
Gholami Z, Faezi ST, Letafatkar A, Madreseh E. Pain neuroscience education, blended exercises and booster sessions as an effective therapy for pain, functional and psychological factors in patients with knee osteoarthritis: a study protocol for a single-blind randomised controlled trial with 22 factorial design during 6-month follow-up. BMJ Open. 2023; 12;13(5):e070336. doi: 10.1136/bmjopen-2022-070336
Ogura K, Bartelstein MK, Yakoub MA, Nikolic Z, Boland PJ, Healey JH. Minimal clinically important differences in SF-36 global score: Current value in orthopedic oncology. J Orthop Res. 2021; 39(10):2116–2123. doi: 10.1002/jor.24944
Weiss BA, Dardick W. An Entropy-Based Measure for Assessing Fuzziness in Logistic Regression. Educ Psychol Meas. 2016; 76(6):986–1004. doi: 10.1177/0013164415623820
Liu Q, Charleston MA, Richards SA, Holland BR. Performance of Akaike Information Criterion and Bayesian Information Criterion in Selecting Partition Models and Mixture Models. Syst Biol. 2023;19;72(1):92–105. doi: 10.1093/sysbio/syac081.
Kon E, Buda R, Filardo G, et al. Platelet-rich plasma: intra-articular knee injections produced favorable results on degenerative cartilage lesions. Knee Surg Sports Traumatol Arthrosc. 2010;18(4):472–479. doi:10.1007/s00167-009-0940-8
van Buul GM, Koevoet WL, Kops N, Bos PK, Verhaar JA, Weinans H, Bernsen MR, van Osch GJ. Platelet-rich plasma releasate inhibits inflammatory processes in osteoarthritic chondrocytes. Am J Sports Med. 2011; 39(11):2362–2370. doi:10.1177/0363546511419278
Dai X, Wang X, Huang Z, Wang K, Ding W. Exact Association Between Preoperative Blood Viscosity and Postoperative Deep Venous Thrombosis Risk in Knee Osteoarthritis Patients: A 10-Year Retrospective Study. Clin Appl Thromb Hemost. 2021;27:10760296211048896. doi:10.1177/10760296211048896
Cheng Y, Liu J, Su Y, Zhao H, Zhao Y, Wen M, Lu S, Zhang W, Wu J. Clinical Impact of Coagulation and Fibrinolysis Markers for Predicting Postoperative Venous Thromboembolism in Total Joint Arthroplasty Patients. Clin Appl Thromb Hemost. 2019; 25:1076029619877458. doi:10.1177/1076029619877458
Liu N, Luo S, Hang C, Zha Z, Li J, Wu W, Hou D. Changes in coagulation functions and hemorheological parameters may predict hematoma formation after total knee arthroplasty. J Orthop Surg Res. 2016; 11:35. doi:10.1186/s13018-016-0370-5
Zhang D, Fan B, Lv L, Li D, Yang H, Jiang P, Jin F. Research hotspots and trends of artificial intelligence in rheumatoid arthritis: A bibliometric and visualized study. Math Biosci Eng. 2023;10;20(12):20405–20421. doi: 10.3934/mbe.2023902
Mishra A, Tummala P, King A, Lee B, Kraus M, Tse V, Jacobs CR. Buffered platelet-rich plasma enhances mesenchymal stem cell proliferation and chondrogenic differentiation. Tissue Eng Part C Methods. 2009;15(3):431–5. doi: 10.1089/ten.tec.2008.0534.
Rodríguez-Merchán EC. Intra-Articular Platelet-Rich Plasma Injections in Knee Osteoarthritis: A Review of Their Current Molecular Mechanisms of Action and Their Degree of Efficacy. Int J Mol Sci. 2022;24;23(3):1301. doi: 10.3390/ijms23031301
Fernández-Fuertes J, Arias-Fernández T, Acebes-Huerta A, Álvarez-Rico M, Gutiérrez L. Clinical Response After Treatment of Knee Osteoarthritis With a Standardized, Closed-System, Low-Cost Platelet-Rich Plasma Product: 1-Year Outcomes. Orthop J Sports Med. 2022;31;10(3):23259671221076496. doi: 10.1177/23259671221076496
Anitua E, Sánchez M, Nurden AT, Nurden P, Orive G, Andía I. New insights into and novel applications for platelet-rich fibrin therapies. Trends Biotechnol. 2006; 24(5):227–34. doi: 10.1016/j.tibtech.2006.02.010.
Hügle T, Nasi S, Ehirchiou D, Omoumi P, So A, Busso N. Fibrin deposition associates with cartilage degeneration in arthritis. EBioMedicine. 2022;81:104081. doi: 10.1016/j.ebiom.2022.104081
Burzynski LC, Humphry M, Pyrillou K, Wiggins KA, Chan JNE, Figg N, Kitt LL, Summers C, Tatham KC, Martin PB, Bennett MR, Clarke MCH. The Coagulation and Immune Systems Are Directly Linked through the Activation of Interleukin-1α by Thrombin. Immunity. 2019;16;50(4):1033–1042.e6. doi: 10.1016/j.immuni.2019.03.003
Hirayama D, Iida T, Nakase H. The Phagocytic Function of Macrophage-Enforcing Innate Immunity and Tissue Homeostasis. Int J Mol Sci. 2017;29;19(1):92. doi: 10.3390/ijms19010092
Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol. 2011; 14;11(11):723–37. doi: 10.1038/nri3073
Ward MP, E Kane L, A Norris L, Mohamed BM, Kelly T, Bates M, Clarke A, Brady N, Martin CM, Brooks RD, Brooks DA, Selemidis S, Hanniffy S, Dixon EP, A O'Toole S, J O'Leary J. Platelets, immune cells and the coagulation cascade; friend or foe of the circulating tumour cell? Mol Cancer. 2021;31;20(1):59. doi: 10.1186/s12943-021-01347-1
Schmied L, Höglund P, Meinke S. Platelet-Mediated Protection of Cancer Cells From Immune Surveillance - Possible Implications for Cancer Immunotherapy. Front Immunol. 2021; 10;12:640578. doi: 10.3389/fimmu.2021.640578
So AK, Varisco PA, Kemkes-Matthes B, Herkenne-Morard C, Chobaz-Péclat V, Gerster JC, Busso N. Arthritis is linked to local and systemic activation of coagulation and fibrinolysis pathways. J Thromb Haemost. 2003; 1(12):2510–5. doi: 10.1111/j.1538-7836.2003.00462.x.
Weinberg JB, Pippen AM, Greenberg CS. Extravascular fibrin formation and dissolution in synovial tissue of patients with osteoarthritis and rheumatoid arthritis. Arthritis Rheum. 1991;34(8):996–1005. doi: 10.1002/art.1780340809
Wu CT, Chen B, Wang JW, Yen SH, Huang CC. Plasma D-dimer is not useful in the prediction of deep vein thrombosis after total knee arthroplasty in patients using rivaroxaban for thromboprophylaxis. J Orthop Surg Res. 2018; 11;13(1):173. doi: 10.1186/s13018-018-0883-1.
Goh GS, Liow MHL, Chen JY, Tay DK, Lo NN, Yeo SJ. The patient acceptable symptom state for the knee society score, oxford knee score and short form-36 following unicompartmental knee arthroplasty. Knee Surg Sports Traumatol Arthrosc. 2023;31(3):1113–1122. doi: 10.1007/s00167-021-06592-x.
Dainese P, Mahieu H, De Mits S, Wittoek R, Stautemas J, Calders P. Associations between markers of inflammation and altered pain perception mechanisms in people with knee osteoarthritis: a systematic review. RMD Open. 2023; 9(2):e002945. doi: 10.1136/rmdopen-2022-002945
Wixted CM, Charalambous LT, Kim BI, Case A, Hendershot EF, Seidelman JL, Seyler TM, Jiranek WA. D-Dimer, Erythrocyte Sedimentation Rate, and C-Reactive Protein Sensitivities for Periprosthetic Joint Infection Diagnosis. J Arthroplasty. 2023;38(5):914–917. doi: 10.1016/j.arth.2022.12.010

No competing interests reported.

Download PDF

Reviews received at journal
12 Sep, 2024
Reviews received at journal
09 Sep, 2024
Reviewers agreed at journal
09 Sep, 2024
Reviewers agreed at journal
07 Sep, 2024
Reviewers agreed at journal
07 Sep, 2024
Reviewers agreed at journal
06 Sep, 2024
Reviewers invited by journal
20 Aug, 2024
Editor assigned by journal
12 Jul, 2024
Submission checks completed at journal
11 Jul, 2024
First submitted to journal
10 Jul, 2024

You are reading this latest preprint version

The crossroads of the hypercoagulability and patient outcomes in osteoarthritis: interactions and connections

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1 Data

2.1.1 Literature sources

2.1.2 Sources of clinical data

2.2 Methodology

2.2.1 Bibliometric Visualization Analysis

2.2.2 Collection of coagulation indexes and immuno-inflammatory indexes

2.2.3 Patient-reported outcomes (PROs)

2.2.4 Association Rules

2.2.5 Statistical analysis

3. Results

3.1 Trends in the issuance of papers

3.2 Analysis of the co-cited literature

3.3 Analysis of co-cited keywords

3.4 Results of coagulation index analysis

3.5 ROC curve of coagulation indices

3.6 Correlation analysis between coagulation indices and immuno-inflammatory indexes, PROs

3.7 Association rule analysis

3.8 Univariate and multivariate regression analysis

4. Discussion

5. Advantages and disadvantages

Declarations

References

Additional Declarations

Status:

Version 1