2025 年 32 巻 4 号 p. 458-473
Aim: D-dimer, lipoprotein (a) (Lp(a)), and high-sensitivity C-reactive protein (hs-CRP) are known predictors of vascular events; however, their impact on the stroke prognosis is unclear. This study used data from the Third China National Stroke Registry (CNSR-III) to assess their combined effect on functional disability and mortality after acute ischemic stroke (AIS).
Methods: In total, 9,450 adult patients with AIS were enrolled between August 2015 and March 2018. Patients were categorized based on a cutoff value for D-dimer, Lp(a), and hs-CRP in the plasma. Adverse outcomes included poor functional outcomes (modified Rankin Scale (mRS score ≥ 3)) and one- year all-cause mortality. Logistic and multivariate Cox regression analyses were performed to investigate the relationship between individual and combined biomarkers and adverse outcomes.
Results: Patients with elevated levels of all three biomarkers had the highest odds of functional disability (OR adjusted: 2.01; 95% CI (1.47-2.74); P＜0.001) and mortality (HR adjusted: 2.93; 95% CI (1.55-5.33); P＜0.001). The combined biomarkers improved the predictive accuracy for disability (C-statistic 0.80 vs.0.79, P＜0.001) and mortality (C-statistic 0.79 vs.0.78, P=0.01).
Conclusion: Elevated D-dimer, Lp(a), and hs-CRP levels together increase the risk of functional disability and mortality one-year post-AIS more than any single biomarker.
Abbreviations: Acute Ischemic Stroke: AIS, Area Under Curve: AUC, High-Sensitivity C-Reactive Protein: hs-CRP, Integrated Discrimination Improvement: IDI, Lipoprotein (A): Lp (a), Modified Rankin Scale: mRS, National Institutes of Health Stroke Scale: NIHSS, Net Reclassification Improvement: NRI, Receiver Operating Characteristic Curve: ROC, Third China National Stroke Registry: CNSRIII, Transient Ischemic Attack: TIA, Trial Of ORG 10172 In Acute Stroke Treatment: TOAST
Stroke is a major global medical problem that causes both disability and mortality. According to the 2019 Global Burden of Disease study, China has the highest stroke burden worldwide, with 3.94 million new stroke cases and 2.19 million stroke-related deaths1, 2). Approximately 69.6% to 70.8% of all strokes are acute ischemic stroke (AIS) and transient ischemic attack (TIA)3). AIS has a significant impact on the quality of life of patients and places a burden on their families and their socioeconomic status. Therefore, to develop appropriate therapeutic approaches and reduce these serious outcomes, the identification of potential risk factors for functional disability and mortality is critical. Several predictive models for stroke disability and mortality have been developed, showing associations between such factors as age, sex, stroke severity, atrial fibrillation, heart failure, and diabetes4, 5). In addition to these cardiovascular risk factors, thrombotic, lipidic, and inflammatory risks play crucial roles in the onset and progression of vascular events through biomarkers such as D-dimer, lipoprotein (a) (Lp (a)), and high-sensitivity C-reactive protein (hs-CRP)6-8). Despite efforts to prevent these risks with antithrombotic, lipid-lowering, and anti-inflammatory therapies, the risks of functional disability and mortality persist, and the mechanism for reducing these risks remains unclear. Recent studies have suggested that the combination of thrombotic and inflammatory markers9, 10), or lipid and inflammatory markers11, 12) have a synergistic effect on adverse outcomes after vascular thrombosis events. However, the combined effects of these three markers on functional disability and mortality following AIS have not yet been investigated.
Our study aimed to investigate the individual and combined effects of the thrombotic biomarker D-dimer, lipid biomarker Lp (a), and inflammatory biomarker hs-CRP on stroke outcomes using data from the Third China National Stroke Registry (CNSR-III). In addition, we determined whether these biomarkers can improve the predictive value of the baseline model, including traditional risk factors.
CNSR-III is a national prospective registry that enrolled 15,166 patients with either AIS or TIA from 201 hospitals in China between 2015 and March 2018. According to the World Health Organization, the diagnosis of AIS requires confirmation using brain magnetic resonance imaging (MRI) or computed tomography (CT). The detailed design and methodology of the CNSR-III have been described previously13). A total of 169 of the 201 participating hospitals contributed to both the biomarker and imaging sub-studies. Patients with biomarker data for D-dimer, Lp (a), and hs-CRP levels were included in this study.
Baseline Data CollectionThe CNSR-III baseline clinical data were carefully collected using an electronic data collection system, followed by face-to-face interviews. Baseline data included demographic information such as age, sex, and body mass index, as well as the National Institutes of Health Stroke Scale (NIHSS) score, smoking, and alcohol consumption. The medical history of all patients was also recorded, including previous stroke (any type) or transient ischemic attack (TIA), hypertension, diabetes, dyslipidemia, coronary artery disease or myocardial infarction, and atrial fibrillation. The inpatient stroke treatment consisted of antiplatelet, antihypertensive, and statin medications. Etiological subtypes were classified according to the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria, which included large artery atherosclerosis, cardioembolic, small artery occlusion, other determined causes, or undetermined causes. The identification of indeterminant causes has been enhanced by computed tomography (CT) or magnetic resonance imaging (MRI), 12-lead electrocardiography, transthoracic echocardiography, 24-hour Holter electrocardiography, and imaging modalities (DSA, MRA, CTA, or ultrasound).
Measurement of BiomarkersFasting blood tests were collected within 24 h of hospital admission in EDTA anticoagulation blood collection tubes. All blood samples were frozen in cryotube at −80℃ freezer and transported to the central laboratory in Beijing Tiantan Hospital via a cold chain. D-dimer levels were determined using an OLYMPUS AU2700 analyzer (Beckman, Japan) and immunoturbidimetry (Kamiya Biomedical, Seattle, WA, USA). Lp (a) concentrations were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Mercodia AB, Sweden). hs-CRP was tested on a Cobas c501 analyzer using a cardiac CRP (latex) high-sensitivity assay (Roche, Basel, Switzerland). Measurements were performed in a College of American Pathologists-certified core laboratory with laboratory personnel blinded to the clinical data according to the manufacturer’s recommendations.
These biomarkers were stratified into low and high plasma levels according to their cut-off values. 0.5 µg/mL, 30 g/dL, and ＞3 mg/L for D-Dimer14), Lp (a)15), and hs-CRP16), respectively.
Functional Outcomes and Follow-UpThe clinical outcomes were closely monitored through face-to-face interviews at 3 months and telephonic interviews at 6 and 12 months. The interviews were conducted by trained research coordinators who were blinded to the baseline clinical status and followed a standardized interview protocol. All adverse clinical outcomes including all-cause mortality and poor functional outcomes were recorded. Functional outcomes were assessed during the follow-up period using the modified Rankin Scale (mRS) score, which ranged from 0 (no symptoms) to 6 (death). Poor functional outcomes were defined as an mRS score of 3–6. Death information was verified by obtaining a death certificate from the hospital or local citizen registry.
Statistical AnalysisStatistical analyses were performed using SAS software program, version 9.4 (SAS Institute, Inc.). Continuous variables were presented as median (interquartile range, IQR) and compared using the Kruskal-Wallis test. Categorical variables were expressed as frequency (%) and compared using either χ2 or Fisher’s exact test. Statistical significance was defined as a two-tailed p-value of ＜0.05. To evaluate the correlation between combined biomarkers and the risk of functional disability and mortality within one year, the participants were divided into eight groups based on their biomarker plasma levels. Group 1 (low D-dimer + low Lp(a) + low hs-CRP) was used as the reference. Group 2 had high D-dimer + Low Lp(a) + low hs-CRP; Group 3 had Low D-dimer + high Lp(a) + low hs-CRP; Group 4 had low D-dimer + low Lp(a) + high hs-CRP; Group 5 had high D-dimer + high Lp (a)+ low hs-CRP; Group 6 had high D-dimer + low Lp (a)+ high hs-CRP; Group 7 had low D-dimer + high Lp (a)+ high hs-CRP; and Group 8 had high D-dimer + high Lp (a)+ high hs-CRP. Group 1 was used as the reference. The study used a logistic regression model to evaluate adjusted odds ratios (OR) with their 95% confidence intervals (CI) for functional disability. Additionally, the study estimated hazard ratios (HR) with 95% CI for the risk of mortality using the Cox regression model. Model 1 was defined as the unadjusted OR and HR. Model 2 was adjusted for age, sex, BMI, history of stroke, history of coronary heart disease, history of arterial fibrillation, hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS score, and TOAST score. Model 3 was adjusted for the same variables as Model 2 plus inpatient antiplatelet, antihypertensive, and lipid-lowering therapy. A test of the interaction between the combined biomarkers and stroke prognosis was performed.
The association between both individual and combined biomarkers and stroke prognosis was estimated using the adjusted Model 3.
The predictive value of combining D-dimer, Lp(a), and hs-CRP for clinical outcomes was evaluated using several metrics, including the area under the receiver operating characteristic curve (AUC), C-statistics, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI). The basic model included age, sex, BMI, medical history of stroke, coronary heart disease, arterial fibrillation, hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS score, and TOAST.
Availability of Data and MaterialsAll data are available from the corresponding author upon reasonable request.
This study included 9,450 eligible patients out of 15,166 patients from CNSR III, as shown in Fig.1. Patients without biomarker measurements were excluded from this study. No baseline variables of interest were missing. Table 1 presents the clinical characteristics of the patients, with a median age of 63 (54-70) years, and 6633 (69%) were male. Among them, 1,266 (13%) patients had poor functional outcomes (mRS3-6) after 1 year, with an average age of 69 (61-77) years and a male rate of 63%. Patients with functional disability (mRS3-6) were found to be older, current smokers (24.5% vs. 32.6%; P＜0.001), and heavy alcohol drinkers (11% vs. 15%; P＜0.001). They also had a medical history of hypertension (66% vs. 62%; P＜0.001), diabetes (29%vs 23%; P＜0.001), stroke (34% vs. 21%; P＜0.001), coronary diseases (15%vs 10%; P＜0.001), and arterial fibrillation (14% vs. 6%; P＜0.001). Supplementary Tables 1, 2, and 3 show the main characteristics of high D-dimer versus low D-dimer, high Lp (a) versus low Lp (a), and high hs-CRP versus low hs-CRP levels, respectively.
Flowchart of the patient selection process
| All patients, 9450 (100) | Good outcome, 8184 (97) | Poor outcome, 1266 (13) | P | |
|---|---|---|---|---|
| Age, y, median (IQR) | 63 (54-70) | 62 (54-69) | 69 (61-77) | ＜0.001 |
| Gender, male | 6633 (69) | 5663 (69) | 801 (63) | ＜0.001 |
| BMI, kg/m2, median (IQR) | 24 (22-26) | 24 (23-27) | 24 (22-26) | ＜0.001 |
| Current Smoking | 3062 (31.5) | 2672 (32.6) | 311 (24.5) | ＜0.001 |
| Heavy Drinking | 1381 (14.2) | 1209 (15) | 137 (11) | ＜0.001 |
| Hypertension | 6073 (62.6) | 5077 (62) | 833 (66) | 0.010 |
| Diabetes | 2296 (23.7) | 1885 (23) | 363 (29) | ＜0.001 |
| Dyslipidemia | 824 (8.5) | 702 (8.6) | 108 (8.5) | 0.955 |
| History of stroke | 2208 (22.8) | 1716 (21) | 429 (34) | ＜0.001 |
| Coronary disease | 1043 (10.8) | 824 (10) | 197 (15) | ＜0.001 |
| Arterial fibrillation | 706 (7.3) | 501 (6) | 183 (14) | ＜0.001 |
| NIHSS, median (IQR) | 3 (1-6) | 3 (1-5) | 7 (4-11) | ＜0.001 |
| TOAST classification | ＜0.001 | |||
| Atherosclerosis | 2424 (25) | 1904 (23) | 437 (34) | |
| Cardioembolic | 637 (6.7) | 502 (6) | 116 (9) | |
| Small Vessel Disease | 2007 (20.7) | 1850 (22) | 118 (9) | |
| Determent | 106 (1) | 87 (1) | 17 (1) | |
| Indeterminant | 4521 (46.6) | 3841 (47) | 578 (46) | |
| Inpatient treatment | ||||
| Antiplatelet | 9329 (96.2) | 7932 (97) | 1177 (93) | ＜0.001 |
| Hypertension lowering | 4515 (46.6) | 3792 (46) | 612 (48) | 0.182 |
| Statin | 93311 (96) | 7892 (96) | 1193 (94) | ＜0.001 |
Abbreviation: BMI: Body Mass Index; n: patients’ number; NIHSS: National Institutes of Health Stroke Scale score; TOAST: Trial of ORG 10172 in Acute Stroke Treatment; P: significant value (0.05).
| Low D-dimer, 1849 (%) | High D-dimer, 7601 (%) | P | |
|---|---|---|---|
| Age | 59 (52-67) | 63 (55-71) | ＜0.001 |
| Gender, Male | 1360 (74) | 5140 (67) | ＜0.001 |
| BMI | 24 (23-27) | 24 (22-26) | 0.010 |
| Smoking | 672 (36) | 2311 (30) | ＜0.001 |
| Drinking | 330 (18) | 1016 (13) | ＜0.001 |
| Hypertension | 1154 (62) | 4756 (62) | 0.824 |
| Diabetes | 444 (24) | 1804 (24) | 0.727 |
| Dyslipidemia | 179 (10) | 631 (8) | 0.031 |
| History of stroke | 385 (21) | 1760 (23) | 0.038 |
| Coronary disease | 176 (9) | 842 (11) | 0.063 |
| Arterial fibrillation | 79 (4) | 605 (8) | ＜0.001 |
| NIHSS | 3 (1-5) | 3 (1-6) | ＜0.001 |
| TOAST | ＜0.001 | ||
| Atherosclerosis | 441 (24) | 1900 (25) | |
| Cardioembolic | 88 (5) | 530 (7) | |
| Small Vessel Disease | 422 (22) | 1546 (20) | |
| Determent | 15 (1) | 89 (1) | |
| Indeterminant | 883 (48) | 3536 (46) | |
| Inpatient treatment | |||
| Antiplatelet | 1800 (97) | 7309 (96) | 0.024 |
| Hypertension lowering | 899 (48) | 3505 (46) | 0.059 |
| Statin | 1792 (97) | 7293 (96) | 0.060 |
Abbreviation: BMI: Body Mass Index; n: patients’ number; NIHSS: National Institutes of Health Stroke Scale score; TOAST: Trial of ORG 10172 in Acute Stroke Treatment; P: significant value (0.05)
| Low Lp (a), 6502 (69%) | High Lp (a), 2948 (31%) | P | |
|---|---|---|---|
| Age | 62 (54-70) | 63 (55-70) | 0.058 |
| Gender, Male | 4469(69) | 1995 (68) | 0.309 |
| BMI | 24 (22-26) | 24 (22-26) | ＜0.001 |
| Smoking | 2051 (32) | 932 (32) | 0.980 |
| Drinking | 910 (14) | 436 (15) | 0.300 |
| Hypertension | 4062 (63) | 1848 (63) | 0.856 |
| Diabetes | 1571 (24) | 677 (23) | 0.286 |
| Dyslipidemia | 546 (8) | 264 (9) | 0.380 |
| History of stroke | 1420 (22) | 725 (25) | ＜0.001 |
| Coronary disease | 672 (10) | 349 (12) | 0.024 |
| Arterial fibrillation | 483 (7) | 201 (7) | 0.308 |
| NIHSS | 3 (1-6) | 3 (1-6) | ＜0.001 |
| TOAST | ＜0.001 | ||
| Atherosclerosis | 1538 (24) | 803 (27) | |
| Cardioembolic | 454 (7) | 164 (6) | |
| Small Vessel Disease | 1396 (21) | 572 (19) | |
| Determent causse | 64 (1) | 40 (1) | |
| Indeterminate causse | 3050 (47) | 1369 (46) | |
| Inpatient treatment | |||
| Antiplatelet | 6265 (96) | 2844 (96) | 0.722 |
| Hypertension lowering | 3083 (47) | 1321 (45) | 0.024 |
| Statin | 6251 (96) | 2834 (96) | 0.848 |
Abbreviation: BMI: Body Mass Index; n: patients’ number; NIHSS: National Institutes of Health Stroke Scale score; TOAST: Trial of ORG 10172 in Acute Stroke Treatment; P: significant value (0.05).
| Low hs-CRP, 6060 (64%) | High hs-CRP, 3390 (36%) | P | |
|---|---|---|---|
| Age | 61 (54-69) | 65 (57-73) | ＜0.001 |
| Gender | 4205 (69) | 2259 (67) | ＜0.001 |
| BMI | 24 (22-26) | 24 (22-27) | 0.067 |
| Smoking | 1972 (32) | 1011 (29) | ＜0.001 |
| Drinking | 893 (15) | 453 (13) | 0.056 |
| Hypertension | 3704 (61) | 2206 (65) | ＜0.001 |
| Diabetes | 1371 (22) | 877 (25) | ＜0.001 |
| Dyslipidemia | 507 (8) | 303 (8) | 0.424 |
| History of stroke | 1316 (22) | 829 (24) | ＜0.001 |
| Coronary disease | 567 (9) | 454 (13) | ＜0.001 |
| Arterial fibrillation | 316 (5) | 368 (11) | ＜0.001 |
| NIHSS | 3 (1-5) | 4 (2-8) | ＜0.001 |
| TOAST | |||
| Atherosclerosis | 1340 (22) | 1001 (30) | |
| Cardioembolic | 313 (5) | 305 (9) | |
| Small Vessel Disease | 1450 (24) | 518 (15) | |
| Determent | 65 (1) | 39 (1) | |
| Indeterminant | 2892 (48) | 1527 (45) | |
| Inpatient treatment | |||
| Antiplatelet | 5872 (97) | 3237 (96) | ＜0.001 |
| Hypertension lowering | 2756 (45) | 1648 (49) | ＜0.001 |
| Statin | 5842 (96) | 3243 (95) | 0.020 |
Abbreviation: BMI: Body Mass Index; n: patients’ number; NIHSS: National Institutes of Health Stroke Scale score; TOAST: Trial of ORG 10172 in Acute Stroke Treatment; P: significant value (0.05).
Table 2 shows the association of the plasma D-dimer, Lp (a), and hs-CRP levels with functional disability. In model 1 (unadjusted), higher baseline levels of each D-dimer, Lp (a), and hs-CRP plasma were significantly associated with a 1.7-fold (OR: 1.72, 95% CI (1.45-2.05); P＜0.001), 1.4-fold (OR: 1.40, 95% CI (1.17-1.50); P＜0.001), and 2.6-fold (OR: 2.61, 95% CI (2.31-2.94); P＜0.001); respectively, regarding the increase in the risk of apoor outcome after 1 year. These associations remained significant after further adjustment in models 2 and 3.
| Event, n (%) | Model 1 | Model 2 | Model 3 | |||||||
| OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | ||
| D-dimer | ||||||||||
| Low | 165 (9) | Ref | - | - | - | - | - | - | - | - |
| High | 1101 (15) | 1.72 | 1.45 - 2.05 | ＜0.001 | 1.21 | 1.00 – 1.46 | 0.045 | 1.20 | 1.00 – 1.45 | 0.048 |
| Lp (a) | ||||||||||
| Low | 802 (12) | Ref | - | - | - | - | - | - | - | - |
| High | 464 (16) | 1.40 | 1.17 – 1.50 | ＜0.001 | 1.22 | 1.07 – 1.41 | 0.003 | 1.23 | 1.07 – 1.41 | 0.002 |
| hs-CRP | ||||||||||
| Low | 557 (9) | Ref | - | - | - | - | - | - | - | - |
| High | 709 (21) | 2.61 | 2.31 – 2.94 | ＜0.001 | 1.49 | 1.30 – 1.71 | ＜0.001 | 1.48 | 1.29 -1.70 | ＜0.001 |
| Combined D-dimer, Lp(a), and hs-CRP | ||||||||||
| Groupe 1 | 67 (7) | Ref | Ref | Ref | ||||||
| Groupe 2 | 302 (9) | 1.39 | 1.06- 1.83 | 0.017 | 1.12 | 0.84 - 1.50 | 0.415 | 1.13 | 0.84- 1.51 | 0.393 |
| Groupe 3 | 28 (7) | 1.09 | 0.69- 1.73 | 0.688 | 1.09 | 0.63 – 1.66 | 0.906 | 1.05 | 0.65- 1.70 | 0.817 |
| Groupe 4 | 44 (13) | 2.19 | 1.46- 3.28 | ＜0.001 | 2.19 | 0.96 – 2.31 | 0.071 | 1.51 | 0.98- 2.35 | 0.061 |
| Groupe 5 | 160 (11) | 1.78 | 1.32- 2.40 | ＜0.001 | 1.39 | 1.01- 1.90 | 0.039 | 1.40 | 1.02- 1.92 | 0.036 |
| Groupe 6 | 389 (20) | 3.55 | 2.71- 4.66 | ＜0.001 | 1.65 | 1.23 – 2.21 | ＜0.001 | 1.64 | 1.22- 2.21 | ＜0.001 |
| Groupe 7 | 26 (16) | 2.76 | 1.69- 4.50 | ＜0.001 | 1.80 | 1.06 – 3.06 | 0.029 | 1.77 | 1.04- 3.02 | 0.033 |
| Groupe 8 | 250 (24) | 4.51 | 3.39- 6.01 | ＜0.001 | 2.00 | 1.46 -2.72 | ＜0.001 | 2.01 | 1.47- 2.74 | ＜0.001 |
Abbreviations: OR: Odds ratio; CI: confidante Intervale; p: significant value (＜0.05) ; Ref : reference; Group 1: low D-dimer + low Lp (a)+ Low hs-CRP ; Group 2: high D-dimer + low Lp (a)+ Low hs-CRP; Group 3: low D-dimer + high Lp (a)+ Low hs-CRP; Group 4: Low D-dimer + low Lp (a)+ high hs-CRP; Group 5: high D-dimer + high Lp (a)+ Low hs-CRP; Group 6: high D-dimer + Low Lp (a)+ high hs-CRP; Group 7: Low D-dimer + high Lp (a)+ high hs-CRP; Group 8: high D-dimer + high Lp (a)+ high hs-CRP; Model 1: unadjusted analysis; Model 2: Adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, TOAST; Model 3: adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, TOAST, inpatient treatment of antiplatelet, hypertension lowering and lipid-lowering.
The effect of the combined biomarkers on functional disability was also assessed in patients divided into eight groups based on their biomarker levels. The group with high plasma of D-dimer, Lp (a), and hs-CRP levels had a significantly higher frequency (24%) of functional disability compared to the other groups. Patients in group 8, who had high plasma D-dimer-high Lp (a)-high hs-CRP levels, had the highest risk of functional disability (OR: 4.51; 95% CI (3.39-6.01); P＜0.001). The association remained significant even after further adjustment, with a two-fold increase in the risk of functional disability in models 2 (OR, 2.00; 95% CI (1.46-2.72); P＜0.001) and model 3 (OR: 2.01; 95% CI (1.47-2.74); P＜0.001). Additionally, group 5 (high D-dimer-high Lp (a)), group 6 (high D-dimer-high hs-CRP), and group 7 (high Lp (a)-high hs-CRP) were significantly associated with an increased risk of functional disability in model 2.
Fig.2 (a) displays a forest plot graph illustrating the increasing risk of poor outcomes based on various groups of combined biomarkers in Model 3. The risk of poor outcomes increased by 1.4-fold in Group 5, 1.6-fold in Group 6, and 1.7-fold in Group 7.
(a): Odds ratio of a poor functional outcome; (b): Hazard ratio of mortality
Supplementary Table 4 in the supplementary indicates no interaction between biomarkers and poor functional outcomes one year after AIS.
| Predictors biomarkers | P |
|---|---|
| D-Dimer＊Lp (a) | 0.354 |
| D-Dimer＊ hs-CRP | 0.510 |
| Lp (a)＊ hs-CRP | 0.612 |
| D-Dimer＊ Lp (a)＊ hs-CRP | 0.981 |
Abbreviation: Lp (a): Lipoprotein (a); hs-CRP: high-sensitivity C-reactive protein; P: significant value (＜0.05); ＊: Interaction
Table 3 summarizes the effect of individual biomarkers on the risk of mortality after one year using a multivariate Cox regression analysis. Model 1 showed that high levels of D-dimer, Lp(a), and hs-CRP were each significantly associated with an increased risk of mortality (HR: 2.20; 95% CI (1.54-3.16); P＜0.001), (HR: 1.35; 95% CI (1.07-1.68); P＜0.001), and (HR: 3.18; 95% CI (2.54-3.99); P＜0.001), respectively. After adjusting for multiple factors, high D-dimer plasma levels as well as high plasma hs-CRP levels remained associated with the risk of mortality after 1 year in models 2 and 3. In the Kaplan-Meier curve presented in Fig.3 (a–c), the cumulative probability of mortality was associated with high D-dimer, Lp (a), and hs-CRP levels.
| Event rate (%) | Model 1 | Model 2 | Model 3 | |||||||
| HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | ||
| D-dimer | ||||||||||
| Low | 33 (2) | Ref | - | - | Ref | - | - | Ref | - | - |
| High | 296 (4) | 2.20 | 1.54 –3.16 | ＜0.001 | 1.53 | 1.05 - 2.23 | 0.025 | 1.50 | 1.05 – 2.22 | 0.026 |
| Lp (a) | ||||||||||
| Low | 205 (3) | Ref | - | - | Ref | - | - | Ref | - | - |
| High | 124 (4) | 1.35 | 1.07 –1.68 | ＜0.001 | 1.25 | 0.99 - 1.59 | 0.055 | 1.26 | 0.99 – 1.60 | 0.054 |
| hs-CRP | ||||||||||
| Low | 120 (2) | Ref | - | - | Ref | - | - | Ref | - | - |
| High | 209 (6) | 3.18 | 2.54 – 3.99 | ＜0.001 | 1.95 | 1.52 –2.48 | ＜0.001 | 1.93 | 1.52 – 2.47 | ＜0.001 |
| Combined D-dimer, Lp(a), and hs-CRP | ||||||||||
| Groupe 1 | 12 (1) | Ref | - | - | Ref | - | - | Ref | - | - |
| Groupe 2 | 63 (2) | 1.59 | 0.85 - 2.95 | 0.136 | 1.28 | 0.68– 2.40 | 0.443 | 1.29 | 0.68- 2.43 | 0.425 |
| Groupe 3 | 4 (1) | 0.87 | 0.28- 2.70 | 0.813 | 0.84 | 0.26 - 2.66 | 0.774 | 0.88 | 0.27- 2.77 | 0.827 |
| Groupe 4 | 11 (3) | 2.86 | 1.26 - 6.5 | 0.012 | 2.11 | 0.90 – 4.93 | 0.083 | 2.17 | 0.93- 5.08 | 0.072 |
| Groupe 5 | 41 (3) | 2.44 | 1.29- 4.66 | 0.006 | 1.96 | 1.01– 3.79 | 0.044 | 1.98 | 1.02- 3.55 | 0.042 |
| Groupe 6 | 119 (6) | 5.33 | 2.94- 9.65 | ＜0.001 | 2.66 | 1.44– 4.93 | ＜0.001 | 2.60 | 1.42- 4.83 | ＜0.001 |
| Groupe 7 | 6 (4) | 3.23 | 1.21- 8.60 | 0.019 | 2.31 | 0.82 -6.47 | 0.111 | 2.22 | 0.78- 5.53 | 0.131 |
| Groupe 8 | 73 (7) | 6.14 | 3.33-11.30 | ＜0.001 | 2.90 | 1.53- 5.47 | ＜0.001 | 2.93 | 1.55- 5.33 | ＜0.001 |
Abbreviation: OR: Odds ratio; CI: confidante Intervale; P: significant value (＜0.05); Ref: reference;
Group 1: low D-dimer + low Lp (a)+ Low hs-CRP; Group 2: high D-dimer + low Lp (a)+ Low hs-CRP; Group 3: low D-dimer + high Lp (a)+ Low hs-CRP; Group 4: Low D-dimer + low Lp (a)+ high hs-CRP; Group 5: high D-dimer + high Lp (a)+ Low hs-CRP; Group 6: high D-dimer + Low Lp (a)+ high hs-CRP; Group 7: Low D-dimer + high Lp (a)+ high hs-CRP; Group 8: high D-dimer + high Lp (a)+ high hs-CRP; Model 1: unadjusted analysis; Model 2: Adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, TOAST; Model 3: adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, TOAST, inpatient treatment of antiplatelet, hypertension lowering and lipid-lowering.
(a): Kaplan-Meier curve of the cumulative probabilities of mortality according to the plasma D-Dimer level; (b): Kaplan-Meier curve of cumulative probabilities of mortality according
In model 1, the risk of one-year mortality was significantly associated with high D-dimer and high Lp(a) (group 5) (HR: 2.44; 95 % CI (1.29 -4.66); P=0.006), high Lp (a), and high hs-CRP (group 7) (HR: 3.23; 95 % CI (1.21-8.60); P=0.019). In addition, high D-dimer and high hs-CRP were significantly associated with a 5.3- and 6.1-fold increased risk of death at 1 year (HR: 5.33; 95 % CI (2.94-9.65); P＜0.001) and (HR: 6.14; 95 % CI (3.33-11.30); P＜0.001), respectively, in group 6 with high D-dimer and high hs-CRP and group 8 with combined high D-dimer-high Lp (a)-high hs-CRP.
After adjusting the data for the analysis (Model 2), the risk of mortality increased 2.9-fold in the group with three high biomarker levels (group 8), 2.9-fold (HR: 2.90; 95%CI (1.53-5.47); P＜0.001), 2.6-fold in the group with high D-dimer and the high hs-CRP (group 6), and 1.9-fold in the group with high D-dimer and high Lp (a) (group 5), (HR: 1.96; 95%CI (1.01-3.79); P=0.044). This association was still significant in model 3, with an increased risk of mortality in AIS patients by 2.9-fold (HR: 2.93; 95% CI (1.55-5.33); P＜0.001), 2.6-fold (HR: 2.60; 95% CI (1.42-4.83); P＜0.001), and 1.9-fold (HR: 1.98; 95% CI (1.02-3.55); P=0.042) respectively in groups 8,6 and 5. A forest plot displaying the HR adjustment for this model is shown in Fig.2 (b).
Supplementary Table 5 in the supplementary materials indicates no interaction between the biomarkers and mortality one year after AIS.
| Predictors biomarkers | P |
|---|---|
| D-Dimer*Lp (a) | 0.655 |
| D-Dimer* hs-CRP | 0.809 |
| Lp (a)* hs-CRP | 0.082 |
| D-Dimer* Lp (a)* hs-CRP | 0.328 |
Abbreviation: Lp (a): Lipoprotein (a); hs-CRP: high-sensitivity C-reactive protein; P: significant value (＜0.05); ＊: Interaction
The addition of combined high levels of D-dimer, hs-CRP, and Lp (a) plasma to the basic model significantly improved the discrimination and reclassification for the prediction of functional disability (C-statistic 0.80 vs 0.79, P=0.001; NRI=0.2 (0.188,0.305), 0.24, P＜0.001; IDI= 0.004 (0.002,0.006), P＜0.001) and mortality (C-statistic 0.79 vs. 0.78, P=0.01; NRI=0.36 (0.261,0.469), P＜0.001; IDI=0.006 (0.003,0.008), P＜0.001); (Table 4).
| C-Statistic | P | NRI | 95%CI | P | IDI | 95%CI | P | |
|---|---|---|---|---|---|---|---|---|
| Poor outcome | ||||||||
| Basic Model | 0.79 | Ref | Ref | Ref | Ref | Ref | ||
| Basic model+ D- | ||||||||
| Dimer+Lp (a)+hs-CRP | 0.80 | ＜0.001 | 0.24 | (0.188-0.305) | ＜0.001 | 0.004 | (0.002-0.006) | ＜0.001 |
| Mortality | ||||||||
| Basic Model | 0.78 | Ref | Ref | Ref | Ref | Ref | ||
| Basic model+ D- | ||||||||
| Dimer+Lp (a)+hs-CRP | 0.79 | 0.01 | 0.36 | (0.2611-0.469) | ＜0.001 | 0.006 | (0.003-0.008) | ＜0.001 |
Abbreviation: OR: Odds ratio; CI: confidante Intervale; P: significant value (＜0.05); Ref: reference; NRI: Net Reclassification Improvement; IDI: Integrated Discrimination Improvement; DD: D-Dimer; Lp(a): Lipoprotein (a); hs-CRP: high-sensitivity C-reactive protein; Basic Model: including Age, Gender, BMI, medical history of stroke, of coronary heart disease, of arterial fibrillation, hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, TOAST classification.
The relationship between biomarkers (individual and combined) and stroke prognosis for each stroke subtype, as defined according to the TOAST classification, is presented in Supplementary Tables 6 and 7.
| Large artery atherosclerosis | Cardioembolism | Small artery occlusion | Other determined causes | Undetermined causes | ||||||||||||||||
| n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | |
| D-dimer | ||||||||||||||||||||
| Low | 54 (12) | Ref | - | - | 17 (19) | Ref | - | - | 19 (4) | Ref | - | - | 0 (0) | Ref | - | - | 75 (8) | Ref | - | - |
| High | 383 (20) | 1.22 | 0.84-1.71 | 0.45 | 99 (18) | 0.67 | 0.35-1.27 | 0.22 | 99 (6) | 1.32 | 0.78-2.25 | 0.29 | 17 (19) | - | - | 0.87 | 503 (14) | 1.23 | 0.93-2.63 | 0.13 |
| Lp (a) | ||||||||||||||||||||
| Low | 265 (17) | Ref | - | - | 83 (18) | Ref | - | - | 78 (6) | Ref | - | - | 7 (11) | Ref | 369 (12) | Ref | - | - | ||
| High | 172 (21) | 1.27 | 0.99-1.62 | 0.06 | 33 (20) | 1.05 | 0.64-1.74 | 0.82 | 40 (7) | 1.22 | 0.80-1.87 | 0.34 | 10 (25) | - | - | 0.08 | 209 (15) | 1.22 | 1.00-1.50 | 0.05 |
| hs-CRP | ||||||||||||||||||||
| Low | 151 (11) | Ref | - | - | 43 (14) | Ref | - | - | 65 (4) | Ref | - | - | 6 (9) | Ref | - | - | 292 (10) | Ref | - | - |
| High | 286 (28) | 1.86 | 1.45-2.38 | ＜0.001 | 73 (24) | 1.35 | 0.85-2.17 | 0.20 | 53 (10) | 1.77 | 1.18-2.66 | 0.00 | 11 (28) | - | - | 0.85 | 286 (19) | 1.25 | 1.02-1.53 | 0.02 |
| Combined D-dimer, Lp(a), and hs-CRP | ||||||||||||||||||||
| Groupe 1 | 19 (9) | Ref | - | - | 7 (16) | Ref | - | - | 9 (4) | Ref | - | - | 0 (0) | Ref | - | - | 32 (6) | Ref | - | - |
| Groupe 2 | 80 (11) | 0.98 | 0.56-1.71 | 0.95 | 24 (13) | 0.46 | 0.17-1.24 | 0.12 | 35 (4) | 1.08 | 0.52-2.35 | 0.83 | 3 (8) | - | - | 0.89 | 160 (10) | 1.13 | 0.91-1.41 | 0.14 |
| Groupe 3 | 10 (10) | 1.02 | 0.44-2.38 | 0.94 | 2 (13) | 0.47 | 0.07-3.01 | 0.42 | 2 (2) | 0.68 | 0.13-3.38 | 0.64 | 0 (0) | - | - | 0.96 | 14 (8) | 1.36 | 0.69-2.69 | 0.36 |
| Groupe 4 | 10 (12) | 0.79 | 0.33-1.89 | 0.59 | 7 (25) | 1.17 | 0.32-4.28 | 0.81 | 6 (9) | 1.65 | 0.53-5.16 | 0.38 | 0 (0) | - | - | 0.99 | 21 (14) | 2.18 | 1.15-4.12 | 0.01 |
| Groupe 5 | 42 (13) | 1.10 | 0.60-2.03 | 0.73 | 10 (14) | 0.58 | 0.18-1.82 | 0.35 | 19 (6) | 1.37 | 0.59-3.17 | 0.45 | 3 (18) | - | - | 0.87 | 86 (13) | 1.75 | 1.11-2.76 | 0.01 |
| Groupe 6 | 156 (28) | 1.84 | 1.06-3.19 | 0.02 | 45 (22) | 0.67 | 0.26-1.75 | 0.41 | 28 (10) | 1.86 | 0.83-4.17 | 0.12 | 4 (20) | - | - | 0.90 | 156 (18) | 1.64 | 1.05-2.54 | 0.02 |
| Groupe 7 | 15 (29) | 3.00 | 1.32-6.83 | 0.00 | 1 (25) | 0.39 | 0.02-6.23 | 0.51 | 2 (7) | 1.16 | 0.19-6.82 | 0.86 | 0 (0) | - | - | 0.98 | 8 (11) | 1.19 | 0.48-2.94 | 0.70 |
| Groupe 8 | 105 (33) | 2.17 | 1.22-3.84 | 0.00 | 20 (27) | 0.79 | 0.27-2.25 | 0.66 | 17 (12) | 2.65 | 1.09-6.47 | 0.03 | 7 (41) | - | - | 0.88 | 101 (22) | 1.95 | 1.23-3.11 | ＜0.001 |
Abbreviation: CI: Confidence intervals; hs-CRP: high-sensitivity C-reactive protein; Lp (a): Lipoprotein (a); n: Number; OR adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, inpatient treatment of antiplatelet, hypertension lowering and lipid-lowering; P: significant value (＜0.05); Ref: reference; Group 1: low D-dimer + low Lp (a)+ Low hs-CRP ; Group 2: high D-dimer + low Lp (a)+ Low hs-CRP; Group 3: low D-dimer + high Lp (a)+ Low hs-CRP; Group 4: Low D-dimer + low Lp (a)+ high hs-CRP; Group 5: high D-dimer + high Lp (a)+ Low hs-CRP; Group 6: high D-dimer + Low Lp (a)+ high hs-CRP; Group 7: Low D-dimer + high Lp (a)+ high hs-CRP; Group 8: high D-dimer + high Lp (a)+ high hs-CRP
| Large artery atherosclerosis | Cardioembolism | Small artery occlusion | Other determined etiology | Undetermined etiology | ||||||||||||||||
| n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | n (%) | OR | 95% CI | P | |
| D-dimer | ||||||||||||||||||||
| Low | 12 (3) | Ref | - | - | 4 (4) | Ref | - | - | 0 (0) | Ref | - | - | 0 (0) | Ref | - | - | 17 (2) | Ref | - | - |
| High | 93 (5) | 1.26 | 0.68-2.34 | 0.45 | 38 (7) | 0.91 | 0.32-2.72 | 0.91 | 23 (1) | - | - | 0.98 | 7 (8) | - | - | 0.99 | 135 (4) | 1.21 | 0.72-2.04 | 0.45 |
| Lp (a) | ||||||||||||||||||||
| Low | 57 (4) | Ref | - | - | 26 (6) | Ref | - | - | 18 (2) | Ref | - | - | 4 (6) | Ref | - | - | 100 (3) | Ref | - | - |
| High | 48 (6) | 1.54 | 1.04-2.27 | 0.02 | 16 (10) | 1.41 | 0.73-2.73 | 0.30 | 5 (1) | 0.56 | 0.20-1.59 | 0.28 | 3 (7) | - | - | 0.28 | 52 (4) | 1.08 | 0.77-1.51 | 0.65 |
| hs-CRP | ||||||||||||||||||||
| Low | 29 (2) | Ref | - | - | 13 (4) | Ref | - | - | 15 (1) | Ref | - | - | 3 (5) | Ref | - | - | 60 (2) | Ref | - | - |
| High | 76 (7) | 1.86 | 1.45-2.38 | ＜0.001 | 29 (10) | 1.76 | 0.88-3.51 | 0.10 | 8 (1) | 1.12 | 0.46-2.74 | 0.79 | 4 (10) | - | - | 0.76 | 92 (6) | 1.90 | 1.35-2.68 | ＜0.001 |
| Combined D-dimer, Lp(a), and hs-CRP | ||||||||||||||||||||
| Groupe 1 | 3 (1) | Ref | - | - | 2 (5) | Ref | - | - | 0 (0) | Ref | - | - | 0 (0) | Ref | - | - | 7 (1) | Ref | - | - |
| Groupe 2 | 10 (1) | 0.80 | 0.22-2.94 | 0.74 | 6 (3) | 0.36 | 0.07-1.89 | 0.23 | 12 (1) | - | - | 0.99 | 2 (6) | - | - | 0.99 | 33 (2) | 1.10 | 0.48-2.52 | 0.80 |
| Groupe 3 | 2 (2) | 1.29 | 0.21-7.75 | 0.78 | 1 (7) | 1.09 | 0.09-12.7 | 0.94 | 0 (0) | - | - | 0.99 | 0 (0) | - | - | 0.99 | 1 (1) | 0.42 | 0.05-3.47 | 0.42 |
| Groupe 4 | 4 (5) | 2.5 | 0.56-11.39 | 0.22 | 1 (4) | 0.55 | 0.04-4.28 | 0.63 | 0 (0) | - | - | 0.99 | 0 (9) | - | - | 0.99 | 6 (4) | 2.80 | 0.94-8.37 | 0.06 |
| Groupe 5 | 14 (4) | 2.34 | 0.66-8.19 | 0.18 | 4 (6) | 0. 80 | 0.13-4.77 | 0.81 | 3 (1) | - | - | 0.99 | 1 (6) | - | - | 0.99 | 19 (3) | 1.56 | 0.65-3.75 | 0.31 |
| Groupe 6 | 40 (7) | 2.85 | 0.86-9.41 | 0.08 | 17 (9) | 0.93 | 0.20-4.26 | 0.93 | 6 (2) | - | - | 0.99 | 2 (10) | - | - | 0.99 | 54 (6) | 2.28 | 1.02-5.10 | 0.04 |
| Groupe 7 | 3 (6) | 2.94 | 0.59-14.69 | 0.18 | 0 (0) | - | - | 0.98 | 0 (0) | - | - | 0.99 | 0 (0) | - | - | 0.99 | 3 (4) | 2.47 | 0.63-9.66 | 0.19 |
| Groupe 8 | 29 (9) | 3.48 | 1.03-11.69 | 0.04 | 11 (15) | 1.09 | 0.22-5.28 | 0.90 | 2 (1) | - | - | 0.99 | 2 (12) | - | - | 0.99 | 29 (6) | 2.03 | 0.87-4.73 | ＜0.001 |
Abbreviation: OR: adjusted for Age, Gender, BMI, History of stroke, History of coronary heart disease, history of arterial fibrillation, Hypertension, dyslipidemia, diabetes, smoking, drinking, NIHSS, inpatient treatment of antiplatelet, hypertension lowering and lipid-lowering; CI: Confidence intervals; hs-CRP: high-sensitivity C-reactive protein; Lp (a): Lipoprotein (a); n: Number; P: significant value (＜0.05); Ref: reference; Group 1: low D-dimer + low Lp (a)+ Low hs-CRP ; Group 2: high D-dimer + low Lp (a)+ Low hs-CRP; Group 3: low D-dimer + high Lp (a)+ Low hs-CRP; Group 4: Low D-dimer + low Lp (a)+ high hs-CRP; Group 5: high D-dimer + high Lp (a)+ Low hs-CRP; Group 6: high D-dimer + Low Lp (a)+ high hs-CRP; Group 7: Low D-dimer + high Lp (a)+ high hs-CRP; Group 8: high D-dimer + high Lp (a)+ high hs-CRP
This multicenter cohort study of 9450 patients revealed several significant findings. First, high plasma D-dimer, Lp (a), and hs-CRP levels at admission were significantly associated with an increased risk of functional disability and mortality one year after the onset of AIS or AIT. Second, this risk increased with an increasing number of combined biomarkers. Third, the predictive model was improved by combining high plasma D-dimer, Lp (a), and hs-CRP levels. The mechanism responsible for reduced functional disability after stroke remains unclear.
A complex interplay of molecular and cellular mechanisms is involved in this pathway, thus leading to poor outcomes and mortality. Therefore, a better understanding of this pathway may help improve therapeutic targeting. Previous studies have shown a significant role of thrombosis, lipid, and inflammatory markers in the development and prognosis of cardiovascular and stroke diseases. However, an assessment of the potential synergistic effects between these markers is still limited. The first part of our study investigated the individual effects of thrombotic, lipid, and inflammatory biomarkers on poor outcomes. Given the absence of approved indicators for stratifying thrombosis risk, we chose to describe the risk of thrombosis by analyzing D-dimer levels. D-dimer is a product of fibrin degradation, resulting from the cleavage of cross-linked fibrin by plasmin during fibrinolysis. Elevated D-dimer levels indicate an increased coagulation activity, thrombin generation, and fibrin formation as well as potential plaque instability and thrombin activation. Previous studies have shown that high D-dimer levels are associated with an increased risk of major thrombotic events and recurrent stroke, which in turn may worsen stroke severity and lead to an increase in functional disability and mortality17, 18). Our findings are consistent with these studies, as we found that plasma D-dimer levels greater than 0.5 µg/mL at admission increased the risk of functional disability by 1.2-fold and the risk of mortality at 1 year by 1.5-fold in patients with AIS. In addition, we selected Lp (a) as a lipid marker, which has garnered considerable attention in several large observational clinical genetic studies. Elevated plasma Lp (a) levels are an independent and causal risk factor for atherosclerotic cardiovascular diseases, mediated by mechanisms associated with increased atherogenesis, inflammation, and thrombosis19-21). Our findings confirmed a significant association between Lp (a) levels above 30 g/dL and increased functional disability, but not mortality. As the most common and effective biomarker for determining the degree of inflammation, we chose hs-CRP as an inflammatory biomarker. Previous meta-analyses have shown that hs-CRP can be used to predict the prognosis of patients with various diseases, notably all stroke types22-24). Our results are consistent with these findings, as a high plasma level of hs-CRP increased the risk of functional disability by 1.4-fold and the risk of mortality by 1.9-fold.
The second part of this study investigated the combined effects of D-dimer, Lp (a), and hs-CRP markers on adverse outcomes after AIS. However, this combination has not yet been clearly established. Our study results showed that the combination of high levels of D-dimer, Lp(a), and hs-CRP was associated with the highest risk compared to low plasma levels of the three biomarkers. Likewise, a study by Li et al. reported a 2.7-fold increased risk of all-cause mortality and a 4.1-fold increased risk of cardiovascular death after heart disease with high levels of D-dimer, hs-CRP, and Lp(a)25). In addition, Tang et al. demonstrated that the combined effect of elevated plasma levels of D-dimer and CRP was a better predictor of adverse clinical outcomes than each biomarker alone in patients diagnosed with acute type A aortic dissection10). On the other hand, a multicenter atherosclerosis study examining the association between inflammation and lipid markers found that the combination of hs-CRP and Lp(a) was independently associated with a 1.6- fold increase in the risk of major cardiovascular events and a 1.3-fold increase in the risk of all-cause mortality12).
The mechanism underlying the interaction between D-dimer, Lp(a), and hs-CRP levels remains unclear. Previous studies have suggested that the interplay between the thrombotic and inflammatory pathways creates a positive feedback loop, thus leading to thrombosis and subsequent inflammation. Inflammation can trigger the coagulation cascade, whereas thrombin and reactive platelets generated during thrombosis can exacerbate inflammation by inducing cytokine release and promoting platelet-inflammatory cell-endothelial cell interactions26). Furthermore, elevated plasma D-dimer levels are associated with interleukin-6, cytokine, and monocyte activation27). In addition, several studies have shown a positive correlation between D-dimer and inflammatory markers, particularly hs-CRP and CRP28-30). In contrast to these previous studies, which suggested an interaction between thrombotic and inflammatory pathways, our current findings indicate that the associations of D-dimer, Lp(a), and hs-CRP with post-stroke outcomes were independent and additive. We did not find any evidence of confounding or interactions between these biomarkers in their analysis, which supports their use as separate predictors of stroke prognosis. Jackson et al. proposed that combined antithrombotic and anti-inflammatory strategies may be effective and safe31). Thus, there is a growing recognition of the need for a deeper understanding of targeted therapies that specifically target thrombo-inflammation to reduce the risk of atherosclerosis and thrombosis while preserving the immune system function9). Recent observations have underscored the importance of specific treatments targeting thromboinflammation, particularly in light of the adverse events associated with certain anti-inflammatory drugs post-myocardial infarction. Clinical studies, such as COMPASS32) and VOYAGER PAD33), have demonstrated the specific and partial targeting of the anti-inflammatory pathway through antithrombotic therapy34). Additionally, previous studies have explored the correlation between lipids and inflammation processes. The American College of Cardiology/American Heart Association Primary Prevention Guideline 2019 and Cholesterol Guideline 2018 35) recommended the use of both Lp (a) and hs-CRP as risk factors for clinical decisions regarding statin therapy initiation.
The simultaneous analysis of these three biomarkers may help identify patients at high risk of a poor prognosis after AIS and to initiate intensive treatment to ameliorate side effects. Future research, including interventional studies, is recommended to examine the efficacy of such drugs on the stroke prognosis in patients using these specific biomarkers. This knowledge has the potential to guide more targeted and effective treatment approaches for patients with stroke, thereby ultimately improving their overall prognosis and quality of life.
Although our study has significant clinical implications, it is important to recognize that some limitations were also associated with this study. First, our study focused only on the D-dimer, Lp (a), and hs-CRP levels. Therefore, future studies involving larger markers of coagulation, lipids, and inflammation should be conducted to elucidate the molecular mechanisms underlying stroke outcomes. Furthermore, the exclusion of a substantial number of patients due to missing laboratory analysis data or mRS measurements, as well as the absence of serial biomarker measurements, may have introduced some deviation in the analysis of the associations between biomarker levels and outcomes. Therefore, future studies should investigate the impact of longitudinal biomarker measurements on the stroke prognosis.
In summary, our study emphasizes the synergistic effect of high plasma D-dimer, Lp(a), and hs-CRP levels in increasing the annual risk of functional disability and mortality in patients with AIS. To enhance the functional outcomes, there is a need for greater emphasis on combined antithrombotic, antilipidemic, and anti-inflammatory therapies in addition to conventional secondary prevention strategies for stroke.
We thank the participants and staff of the Third China National Stroke Registry (CNSR-III) study.
Ethics Approval and Consent to ParticipateThe protocol and data collection methods of the CNSR-III were approved by the ethics committee of Beijing Tiantan Hospital (IRB approval No: KY2015-001-01) and all participating centers and other participating hospitals. Written informed consent was obtained from each participant or a legally authorized representative (primarily spouse, parents, adult children, or otherwise indicated). The CNSR III study was performed in accordance with the principles of the Declaration of Helsinki.
Competing InterestsAll authors declare no conflicts of interest.
This study was supported by grants from the National Natural Science Foundation of China (U20A20358), Capital’s Funds for Health Improvement and Research (2020-1-2041), and Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (2019-I2M-5-029).
LM, the first author, conducted this study, interpreted the data, and drafted the manuscript. AJ conducted the statistical analysis and interpreted the data. YP interpreted data and reviewed the manuscript. JL performed experiments and interpreted the data. YJ, the author, collected and verified the data. XM interpreted data and supervised the study. YW, the corresponding author, designed and supervised the study this study and revised the manuscript.
Lamia M’barek : 0000-0003-1800-0108; Aoming Jin: 0000-0002-7221-5444; Yuesong Pan: 0000-0003-3082-6789; Jinxi Lin: 0009-0005-0995-0541; Yong Jiang: 0000-0001-7700-6054; Xia Meng: 0000-0001-5223-7390 ; Yongjun Wang: 0000-0002-9976-2341.
