Abstract
Background
Statins had a higher risk of statin-induced side effects for acute ischemic stroke patients with SLCO1B1 variants. Ezetimibe might reach the low-density lipoprotein cholesterol (LDL-C) level to recommend a range with a low risk of side effects. Which effect was better needs validation?
Methods
Data from acute ischemic stroke patients with SLCO1B1 variants administered. The stroke recurrence within 1 year and recommended LDL-C level 30 days after admission were used to evaluate the primary outcome. The secondary outcome included unfavorable functional outcomes and side effects of statins.
Results
The mean age of the 158 patients was 66.59 ± 13.268 years, and 65 were female (41.1%). The ezetimibe group had a lower stroke recurrence (6.3% vs. 17.9%, p = 0.036), a higher rate of patients reaching the recommended LDL-C level (81.0% vs. 48.4%, p < 0.001), and a lower rate of side effects of statins (3.2% vs. 13.7%, p = 0.027). Regression analysis showed the ezetimibe was related to stroke recurrence (OR = 0.276, p = 0.039), the recommended LDL-C level (OR = 2.312, p < 0.001), and side effects of statins (OR = 0.194, p = 0.038), but not to unfavorable functional outcomes at 90 days after admission (OR = 0.960, p = 0.911). The patients had a significant difference in base character and outcome events between the hierarchical clustering groups.
Conclusions
The ezetimibe group had a lower rate of recurrence and side effects, a higher rate of recommended LDL-C level, and a similar unfavorable functional outcome. For male, younger stroke patients with SLCO1B1 variants, higher blood pressure, and blood lipid using ezetimibe are related to a better prognosis than other patients.
Supplementary Information
The online version contains supplementary material available at 10.1186/s40001-025-03738-2.
Keywords: Ezetimibe, Low-dose statins, Ischemic stroke, SLCO1B1 variants, Machine learning
Introduction
Statins are usually recommended for acute ischemic stroke patients [1]. Some studies have suggested that the treatment of statins is related to a better neurological function and prognosis for acute ischemic stroke patients [2, 3]. However, it is also controversial whether high-intensity statins are associated with side effects in patients with ischaemic stroke [4, 5].
Especially for acute ischemic patients with SLCO1B1 variants, taking statins had a higher incidence rate of statin-induced side effects, such as elevated transaminase or myodynia [6]. Low-dose statins could be a choice for these patients [7]. Our previous study has shown that half of the acute ischemic strokes taking low-dose statins can reach the guideline-recommended LDL-C level [8]. A study also showed that ischemic stroke with a lower LDL-C level had a better prognosis [9].
Ezetimibe, another lipid-lowering drug, can avoid the side effects of statins. Although ezetimibe does not exhibit the neuroprotective effects associated with statins, its unique mechanism of action offers a practical lipid-lowering alternative for stroke patients who are insensitive to statin therapy [10–12].
This study compares which medication of ezetimibe or low-dose statins were better for patients. In addition, we want to know which medication was better for some patients. The clustering analysis methods of unsupervised machine learning methods could classify patients by clinical and other characteristics [13]. The appropriate classes of patients might have a better prognosis by taking some medicine or methods [13]. Therefore, the class of clustering analysis could provide a better clinical treatment strategy.
This study explored which ezetimibe or low-dose statins are better for ischemic stroke patients with SLCO1B1 variants and whether this ezetimibe could reduce stroke recurrence and achieve recommended LDL-C levels. We further analyzed whose patients might benefit from ezetimibe or low-dose statins by unsupervised machine learning.
Methods
Study subjects
The study is a cohort study. We recruited acute ischemic stroke patients with SLCO1B1 variants in the Neurology Department and Rehabilitation Department of the Affiliated Hospital of Youjiang Medical University for Nationalities. The patients were enrolled from January 1, 2020, to July 31, 2022, and followed up to July 31, 2023. All patients received neuroimaging examinations and met the WHO stroke diagnostic criteria.
The patients who took ezetimibe were assigned to the ezetimibe group. In addition, those who took low-dose statins were assigned to the statin group. The patients were prescribed statins or ezetimibe for their different conditions. Low-dose statins were defined as atorvastatin 10 mg, simvastatin 5 mg, and rosuvastatin 5 mg daily after onset [7]. Considering a 10% loss to follow-up, we initially recruited more than 100 patients based on a previous article [11].
The inclusion criteria were as follows: (1) aged 18 years or older; (2) having received a conventional therapy, such as thrombolysis or antiplatelet drug after admission; and (3) beginning to take statins or ezetimibe within 48 h after onset.
The exclusion criteria were as follows: (1) mRS ≥ 2 before onset; (2) patients have a history of hyperlipidemia or taking statins before admission; (3) having taken other lipid-lowering drugs, such as fenofibrate; and (4) intracerebral hemorrhage, subarachnoid hemorrhage, thrombus of lower extremity veins or severe systemic disease.
The study was performed by the Declaration of Helsinki and the ethical standards of the institutional and/or national research committee. The study was approved by the Ethics Committee of the Affiliated Hospital of Youjiang Medical University for Nationalities. Informed consent was obtained from all individual participants included in the study.
Genotyping
We use an ABI 7500 solid Time fluorescence quantitative PCR instrument (manufacturer: Thermo Fisher, USA) and human SLCO1B1 gene testing kit (manufacturer: Wuhan Youzhiyou Medical Technology Co., Ltd). Tested according to reagent and instrument instructions for SLCO1B1. The SLCO1B1 variants included SLCO1B1 388A > G and SLCO1B1 521T>C.
Study procedures
After admission, the patient’s demographic data (such as age, gender, and subtype of stroke) and laboratory data, such as TC (total cholesterol), TG (triglyceride), HDL-C (high-density lipoprotein cholesterol), and LDL-C, from electronic clinical records we collected. The patients’ NIHSS and mRS scores were evaluated. Information on smoking and drinking history, history of diseases such as stroke, hypertension, diabetes, and atrial fibrillation, and history of using antiplatelet, antihypertensive, antidiabetics or insulin, and anticoagulant drugs were collected using the structured questionnaires completed by patients or their relatives.
Endpoints
The primary outcome was the rate of stroke recurrence and the recommended LDL-C level. The secondary outcome was the unfavorable functional outcome. The stroke recurrence was ischemic stroke occurrence within 1 year after onset. The recommended LDL-C level was determined 30 days after admission, defined as LDL-C level < 1.8 mmol/L or half of the level at admission. The unfavorable functional outcome was defined as mRs > 2 90 days after admission. The side effects of statins included elevated transaminase and myosalgia within 1 year.
At 30 days after admission, the patient’s LDL-C level was collected. At 90 days after admission, the patients’ mRS score was obtained via face-to-face or telephone interview. At 90 days, 120 days, 180 days, and 1 year after admission, the recurrence and side effect was obtained via face-to-face or telephone interview. All events were determined by two experienced neurologists blind to the patient’s condition and grouping. All outcomes were determined by two experienced neurologists blind to the patient’s condition and grouping.
Statistical analysis
We used the chi-square test for categorical variables to compare the baseline characteristics and outcomes between the two groups. We used a t test for normally distributed continuous variables and a non-parametric test for abnormally distributed continuous variables. The continuous variables in accord with normal distribution were expressed as mean ± SD (standard deviation), and continuous variables with abnormal distribution were expressed as the median and frequencies. Categorized data and ranked data were expressed as numbers and percentages. The statistical significance was set as P < 0.05.
We analyzed the relationship between risk factors and outcome events using univariable and multivariable logistic regression methods. The correlation was expressed as the odds ratios (ORs), 95% Cis, and p values.
Classify data by unsupervised machine learning
We analyzed different characters of patients through cluster analysis. We standardize data by the StandardScaler module in the sklearn library. We classify data by K-means methods (K-means module and sklearn library) and hierarchical clustering methods (agglomerative clustering module and sklearn library). To choose the best number of groups, we draw a line chart by Silhouette score for k-means methods and a hot map for Hierarchical Clustering methods. The line or hot map could suggest a better number of groups. Then, we could analyze the data by the groups. We compare baseline data and outcome events by the above difference group. The grouping was better when the data and events between groups had a significant difference. Then, we further analyzed which medicine was more appropriate for patients with different characters through the grouping.
SPSS 23.0 for Windows and Python 3.80 was used to process the data.
Results
Study subjects
Initially, we recruited 178 eligible patients for the study. Eleven patients were lost to follow-up, and nine withdrew from the study. Finally, we acquired the data of 158 patients (63 in the ezetimibe group and 95 in the statin group).
Baseline characteristics
The mean age of the 158 patients was 66.59 ± 13.268 years, and 41.1% were female. The median and interquartile ranges of the NIHSS score at admission were 12 [8–17].
Compared with the statin group, the ezetimibe group had a higher rate of antihypertensive drug use and a lower diastolic blood pressure value at admission. No significant difference in other baseline data was found between the two groups (Table 1).
Table 1.
Baseline characteristic and outcome data between Ezetimibe and statins group
| variables | Ezetimibe group (N = 63) | Statins group (N = 95) | P* |
|---|---|---|---|
| Baseline characteristic | |||
| Age, years | 67.11 (12.647) | 66.25 (13.721) | 0.692 |
| Female, % | 21 (33.3) | 44 (46.3) | 0.104 |
| Admission NIHSS score | 13 (10–18) | 11 (5–17) | 0.174 |
| Antiplatelet Drug in Hospital, % | 56 (88.9) | 81 (85.3) | 0.511 |
| Systolic Blood Pressure at admission, mmHg | 138.68 (26.402) | 147.64 (22.903) | 0.069 |
| Diastolic Blood Pressure at admission, mmHg | 78.33 (14.123) | 84.72 (14.676) | 0.007 |
| Smoking, % | 32 (33.7) | 26 (41.3) | 0.333 |
| Hypertension, % | 36 (57.1) | 50 (52.6) | 0.577 |
| Diabetes Mellitus, % | 9 (14.3) | 15 (15.8) | 0.797 |
| Coronary heart disease, % | 10 (15.9) | 12 (12.6) | 0.564 |
| Antihypertensive, % | 33 (52.4) | 27 (28.4) | 0.002 |
| Platelet, mmol/l | 183.00 (63.390) | 178.21 (52.666) | 0.607 |
| INR | 1.01 (0.154) | 0.98 (0.120) | 0.102 |
| ALT, mmol/l | 24.89 (19.909) | 22.93 (13.412) | 0.494 |
| Creatinine, mmol/l | 81.92 (25.076) | 74.39 (24.094) | 0.059 |
| Glucose, mmol/l | 8.06 (2.240) | 7.93 (3.021) | 0.783 |
| Triglyceride, mmol/l | 1.48 (1.032) | 1.57 (1.263) | 0.655 |
| Total cholesterol, mmol/l | 4.24 (1.029) | 4.46 (0.957) | 0.167 |
| LDL-C, mmol/l | 2.58 (0.815) | 2.71 (0.812) | 0.197 |
| HDL-C, mmol/l | 1.26 (0.353) | 1.31 (0.332) | 0.392 |
| Outcome data | |||
| Recurrence, % | 4 (6.3) | 17 (17.9) | 0.036 |
| Standard LDL-C at 15 days, % | 51 (81.0) | 46 (48.4) | < 0.001 |
| UFO, % | 36 (57.1) | 58 (61.1) | 0.624 |
| Side effects of statins, % | 2 (3.2) | 13 (13.7) | 0.027 |
INR international normalized ratio, ALT glutamic-pyruvic transaminase, LDL-C low-density lipoprotein cholesterol, HDL-C high-density lipoprotein cholesterol, UFO Unfavorable functional outcome
P* was calculated by ANOVA, Chi-square test, or Mann–Whitney U test as appropriate
Primary outcome
The chi-square test showed that the ezetimibe group had a lower rate of patients with stroke recurrence (6.3% vs. 17.9%, p = 0.036). The ezetimibe group had a higher rate of patients reaching the recommended LDL-C at 30 days after admission than the statin group (81.0% vs. 48.4%, p < 0.001) (Table 1).
The univariate logistic regression analysis found that older age, higher systolic blood pressure at admission, higher NIHSS scores at admission, taking ezetimibe, and taking antiplatelet were related to stroke recurrence. Higher diastolic blood pressure, taking ezetimibe, antiplatelet, antihypertensive, higher total cholesterol or LDL-C value at admission, and higher creatinine value at admission were related to the rate of recommended LDL-C level.
By the multivariable logistic regression analysis, we found that higher NIHSS scores at admission (OR = 1.100, p = 0.008), taking ezetimibe (OR = 0.276, p = 0.039) and taking antiplatelet (OR = 0.162, p = 0.002) were still related to the stroke recurrence (Table 2). The taking of ezetimibe (OR = 2.312, p < 0.001) and higher LDL-C value (OR = 0.348, p < 0.001) at admission were still associated with the recommended LDL-C level (Table 2).
Table 2.
Logistic regression analysis for outcome events
| Risk factor | OR (95%CI) | P* |
|---|---|---|
| Recurrence events | ||
| Using ezetimibe | 0.276 (0.081–0.936) | 0.039 |
| NIHSS at admission | 1.100 (1.025–1.180) | 0.008 |
| Antiplatelet medicine in hospital | 0.162 (0.053–0.499) | 0.002 |
| Standard LDL-C at 15 days | ||
| LDL-C at admission | 0.348 (0.210–0.578) | < 0.001 |
| Using ezetimibe | 2.312 (1.621–3.676) | < 0.001 |
| Unfavorable functional outcome | ||
| NIHSS at admission | 1.141 (1.077–1.210) | < 0.001 |
| Blood glucose at admission | 1.188 (1.034–1.365) | 0.015 |
| Using ezetimibe | 0.960 (0.469–1.965) | 0.911 |
| Side effects | ||
| Using ezetimibe | 0.194 (0.041–0.911) | 0.038 |
P* was calculated by multivariable logistic regression analysis, LDL-C low-density lipoprotein cholesterol
Second outcome
The chi-square test revealed that the ezetimibe and statin groups had a similar unfavorable functional outcome rate (42.9% vs. 38.9%, p = 0.624) 90 days after admission. The ezetimibe group had a lower rate of side effects of statins than the statins group (3.2% vs. 13.7%, p = 0.027) (Table 1).
Through the univariate logistic regression analysis, we found that older age, females, higher NIHSS scores at admission, taking ezetimibe, taking antiplatelet, taking hypoglycemic, history of diabetes, and higher value of blood glucose at admission were related to the unfavorable functional outcome. In addition, older age, females, higher NIHSS scores at admission, taking ezetimibe, and higher value of LDL at admission were related to the rate of side effects of statins.
By the multivariable logistic regression analysis, we found that higher NIHSS scores at admission (OR = 1.141, p < 0.001) and higher values of blood glucose (OR = 1.188, p = 0.015) at admission were still related to the unfavorable functional outcome. Taking ezetimibe (OR = 0.194, p = 0.038) was still associated with the rate of side effects of statins (Table 2).
Classify data results by unsupervised machine learning
The Silhouette score suggested that the 2 group was the best choice for k-means methods (Fig. 1). The hot map also suggested that the two groups were the best choice for the Hierarchical Clustering methods (Fig. 2). First, we group patients into two groups through k-means (k-means-1 group and k-means-2 group). Then, we group patients through hierarchical clustering methods (hierarchical-1 group and hierarchical-2 group).
Fig. 1.
Line chart of Silhouette score for k-means methods
Fig. 2.
Hot map for hierarchical clustering methods
We compare outcome event rates in different classified groups. The k-means-1 group had a lower rate of recommended LDL-C (p = 0.001) than the k-means-2 group; the other event outcomes did not differ significantly (Supplemental Table 1). The hierarchical-2 group had a lower rate of recurrent events (p = 0.036), a higher rate of recommended LDL-C (p = 0.001), and a higher rate of FFO (p = 0.001) than the hierarchical-1 group (Table 3). Therefore, the hierarchical clustering method was the better grouping method for data in the study. The use of ezetimibe had a higher rate in the hierarchical-2 group (88.9%) and k-means-2 group (53.9%). In addition, the patients of the hierarchical-2 group are obviously from ezetimibe group (Table 3, Supplemental Table 1). In logistic regression analysis, the hierarchical clustering group relates to recurrence events (OR = 0.311 p = 0.045), rate of recommended LDL-C (OR = 3.150, p < 0.001), and UFO (OR = 0.333, p = 0.001) (Table 4).
Table 3.
Outcomes between hierarchical clustering group
| outcome | Hierarchical-1 Group (n = 95) |
Hierarchical-2 Group (n = 63) |
P* |
|---|---|---|---|
| Recurrence | 17 (17.9) | 4 (6.3) | 0.036 |
| Standard LDL-C | 47 (49.5) | 50 (79.4) | < 0.001 |
| UFO | 41 (43.2) | 23 (14.8) | 0.001 |
| Side effects | 12 (12.6) | 3 (4.8) | 0.098 |
| Ezetimibe group | 7 (7.4) | 56 (88.9) | < 0.001 |
UFO Unfavorable functional outcome
P* was calculated by the Chi-square test. LDL-C low-density lipoprotein cholesterol
Table 4.
Logistic regression analysis for hierarchical clustering group and outcome
| Risk factor | OR (95%CI) | P* |
|---|---|---|
| Recurrence events | 0.311 (0.099–0.973) | 0.045 |
| Standard LDL-C at 30 days | 3.150 (1.602–6.193) | 0.001 |
| UFO | 0.333 (0.172–0.644) | 0.001 |
| Side effects | 0.346 (0.093–1.279) | 0.112 |
UFO Unfavourable functional outcome
P* was calculated by univariate logistic regression analysis, LDL-C low-density lipoprotein
To compare the character of patients between two hierarchical clustering groups, we found that the patients of the hierarchical-2 group had fewer female patients, fewer older patients, higher blood pressure at admission, lower NIHSS and mRS scores at admission, higher rate of using antiplatelet and antihypertensive medicine; lower value of INR, platelet, ALT and higher value of blood lipid at admission than hierarchical-1 group (Table 5).
Table 5.
Baseline characteristic between hierarchical clustering group
| variables | Hierarchical group 1 (n = 95) | Hierarchical group 2 (n = 63) | P* |
|---|---|---|---|
| Age, years | 74.08 (9.495) | 59.66 (12.510) | < 0.001 |
| Female, % | 43 (45.2) | 22 (34.9) | < 0.001 |
| Admission NIHSS score | 14 (10–18) | 11 (5–16) | 0.174 |
| Antiplatelet Drug in Hospital, % | 76 (80.0) | 61 (96.8) | 0.022 |
| Systolic Blood Pressure at admission, mmHg | 139.14 (25.753) | 148.63 (22.845) | 0.015 |
| Diastolic Blood Pressure at admission, mmHg | 78.22 (12.931) | 85.83 (15.449) | 0.001 |
| Smoking, % | 15 (15.7) | 43 (68.3) | < 0.001 |
| Hypertension, % | 45 (47.4) | 41 (65.1) | 0.029 |
| Diabetes Mellitus, % | 9 (9.5) | 15 (23.8) | 0.014 |
| Coronary heart disease, % | 19 (20.0) | 3 (4.7) | < 0.001 |
| Antihypertensive, % | 24 (25.3) | 36 (57.1) | < 0.001 |
| Platelet, mmol/l | 189.28 (60.449) | 170.24 (51.697) | 0.036 |
| INR | 0.94 (0.068) | 1.05 (0.165) | < 0.001 |
| ALT, mmol/l | 27.26 (17.401) | 19.88 (14.122) | 0.004 |
| Creatinine, mmol/l | 80.25 (24.838) | 74.74 (19.758) | 0.124 |
| Glucose, mmol/l | 8.20 (2.871) | 7.78 (2.593) | 0.331 |
| Triglyceride, mmol/l | 1.35 (0.894) | 1.71 (1.369) | 0.049 |
| Total cholesterol, mmol/l | 4.04 (0.978) | 4.68 (0.903) | < 0.001 |
| LDL-C, mmol/l | 2.35 (0.811) | 2.92 (0.696) | < 0.001 |
| HDL-C, mmol/l | 1.36 (0.373) | 1.22 (0.291) | 0.007 |
INR international normalized ratio, ALT glutamic-pyruvic transaminase, LDL-C low-density lipoprotein cholesterol, HDL-C high-density lipoprotein cholesterol
P* was calculated by ANOVA, Chi-square test, or Mann–Whitney U test as appropriate
Discussion
The present study aimed to explore the association of ezetimibe or low-dose statins with the prognosis and rate of recommended LDL-C levels for ischemic stroke patients with SLCO1B1 variants. We found the ezetimibe group had a lower rate of stroke recurrence within 1 year. The ezetimibe was associated with a lower rate of stroke recurrence. The ezetimibe and statin groups did not significantly differ in unfavorable functional outcomes 90 days after admission. The patients in the ezetimibe group had a higher rate of recommended LDL-C than those in the statin group, and the difference was statistically significant. The ezetimibe was associated with a higher rate of recommended LDL-C level. The ezetimibe group had a lower rate of statin-induced side effects. By classifying the group, The hierarchical-2 group had a better prognosis. For using ezetimibe, patients in the classified group that were male and younger had higher blood pressure and lower NIHSS scores; more using antiplatelet and antihypertensive medicine, lower value of INR, platelet, ALT and higher value of blood lipid at admission had a better prognosis.
The patients of the ezetimibe group had a lower stroke recurrence rate. The results were reasonable, for the ezetimibe group had a higher rate of patients with recommended LDL-C levels [7–9]. The second reason for the results might be that the patients on statins had worse compliance with medicine for side effects. The patients with SLCO1B1 variants had a higher rate of side effects for a longer using statin [6, 7]. Therefore, for a longer time, more patients might have a stroke recurrence for not taking statins continuously.
Our data suggested that the ischemic stroke patients with SLCO1B1 variants using ezetimibe and low-dose statins had a similar prognosis. No significant difference in unfavorable functional outcomes was found between the two groups. Our findings are consistent with the results of previous studies, which have suggested that ezetimibe and statins had similar effects in preventing stroke [11, 12]. Nonetheless, more RCT research is needed to confirm this conclusion.
The ezetimibe group had a lower rate of statin-induced side effects than the statin group, indicating the safety of the ezetimibe. These results confirm the ezetimibe’s safety in treating acute ischemic stroke patients with SLCO1B1 variants.
However, the LDL-C level was reduced in both groups after admission. A higher rate of patients in the ezetimibe group reached the recommended LDL-C level. Some studies have suggested that a recommended LDL-C level is related to the occurrence and prognosis of ischemic stroke patients [14, 15]. Our data showed a higher rate of patients in the ezetimibe group having a recommended LDL-C level 30 days after admission. The results were consistent with the results of other studies [11, 12].
Unsupervised machine learning methods are approaches used to identify unknown features or influencing factors within unlabeled data [13]. These methods can be applied to classify clinical patients in terms of prognosis, diagnosis, or treatment tendencies when clear influencing factors are absent [13]. In our previous research, we successfully utilized unsupervised machine learning to guide treatment strategy selection for patients who experienced adverse reactions to statins or antiplatelet agents [16, 17].In this study, the application of unsupervised machine learning methods to data analysis similarly resulted in analogous groupings and the identification of relevant factors. The unsupervised machine learning analysis showed that the male, younger, and lower NIHSS score patients taking ezetimibe had a better prognosis. The male stroke patients had a better prognosis, consistent with previous studies results, including this study [8]. Patients with higher blood lipids might benefit from the ezetimibe. Therefore, these patients had a better prognosis. The lower blood pressure might have a worse condition for their worse prognosis.
Recent studies have raised concerns about the safety of using low-dose statins in patients with ischemic stroke, leading to new conclusions regarding appropriate statin dosage and patient prognosis [18–22]. Consequently, it is crucial to pay closer attention to safety considerations when prescribing statins in clinical practice. This also underscores the importance of alternative lipid-lowering agents, such as ezetimibe, for patients with stroke. Our results on the effectiveness and safety of ezetimibe suggest this ezetimibe is a feasible therapy for acute ischemic stroke with SLCO1B1 variants, especially for male patients. Hopefully, the ezetimibe could achieve better safety in these patients. Therefore, ezetimibe may be a better therapeutic method for them.
This study has several limitations. First, fewer patients were recruited in the ezetimibe group. Hence, we used different statistical methods to verify the validity of the results. Second, the number of recurrences and other complications were fewer. Therefore, we further verified the results with multivariable logistic regression analysis. All the results further strengthen the statistical power of our findings. Therefore, the study conclusions should be validated in a cohort comprising a larger number of patients to obtain more robust statistical results.
In conclusion, for acute ischemic stroke patients with SLCO1B1 variants, ezetimibe is related to a lower rate of recurrence statin-induced side effects than low-dose statin group. The two groups had a similar rate of unfavorable functional outcomes. The ezetimibe group is related to a higher rate of recommended LDL-C level at 30 days after admission. For male, younger stroke patients with SLCO1B1 variants, higher blood pressure, and blood lipid using ezetimibe are related a better prognosis than other patients.
Supplementary Information
Acknowledgements
We thank MD. Jiaona Lan for offering suggestions and guides for the paper.
Author contributions
Dr. Chaohua Cui and MD Tonghua Long conceived and designed the study. MD Haoye Guan and MD Qiulian Yin screened and collected data and analzed the data. Dr. Chaohua Cui and MD Tonghua Long write the manuscript. All authors reviewed and revised the manuscript.
Funding
This study was supported by the Specific Research Project of Guangxi for Research Bases and Talents (AD23026241). The funds provided financial support for this study.
Declarations
Ethics approval and consent to participate
The study was performed in accordance with the Declaration of Helsinki and the ethical standards of the institutional and national research committees and approved by the Ethics Committee of the Affiliated Hospital of Youjiang Medical University for Nationalities (2019(013)). Informed consent was secured from all participants. Informed consent was obtained from all individual participants included in the study.
Consent for publication
All patients had given informed consent for publication.
Competing interest
The authors declare no competing interests.
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Chaohua Cui and Tonghua Long contributed equally to this article.
References
- 1.Marti-Fabregas J, Gomis M, Arboix A, et al. Favorable outcome of ischemic stroke in patients pretreated with statins. Stroke. 2004;35:1117–21. 10.1161/01.STR.0000125863.93921.3f. [DOI] [PubMed] [Google Scholar]
- 2.Tsivgoulis G, Katsanos AH, Sharma VK, et al. Statin pretreatment is associated with better outcomes in large artery atherosclerotic stroke. Neurology. 2016;86:1103–11. 10.1212/wnl.0000000000002493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Powers. 2018 Guidelines for the Early Management of Patients With Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association (vol 49, pg e46, 2018). Stroke. 2018;49(3):e46–e110. 10.1161/STR.0000000000000158. [DOI] [PubMed]
- 4.Amarenco P, Bogousslavsky J, Callahan A, et al. High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med. 2006;355:549–59. [DOI] [PubMed] [Google Scholar]
- 5.Chung CM, Lin MS, Liu CH, Lee TH, Chang ST, Yang TY, et al. Discontinuing or continuing statin following intracerebral hemorrhage from the view of a national cohort study. Atherosclerosis. 2018;278:15–22. [DOI] [PubMed] [Google Scholar]
- 6.Bigossi M, Maroteau C, Dawed AY, Taylor A, Srinivasan S, Melhem AL, et al. A gene risk score using missense variants in slco1b1 is associated with earlier onset statin intolerance. Eur Heart J Cardiovasc Pharmacother. 2023;9:536–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Vladutiu GD, Isackson PJ. slco1b1 variants and statin-induced myopathy. N Engl J Med. 2009;360:304–3046. [DOI] [PubMed] [Google Scholar]
- 8.Liao JK. Safety and efficacy of statins in Asians. Am J Cardiol. 2007;99:410–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cui CH, Li YB, Bao JJ, et al. Low dose statins improve the prognosis of ischemic stroke patients with intravenous thrombolysis. BMC Neurol. 2021;21(1):220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Amarenco P, Kim JS, Labreuche J, Charles H, Abtan J, Bejot Y, et al. A comparison of two ldl cholesterol targets after ischemic stroke. N Engl J Med. 2020;382:9–19. [DOI] [PubMed] [Google Scholar]
- 11.Nakamura H, Arakawa K, Itakura H, et al. Primary prevention of cardiovascular disease with pravastatin in Japan (MEGA study): a prospective randomised controlled trial. Lancet. 2006;368:1155–63. 10.1016/s0140-6736(06)69472-5. [DOI] [PubMed] [Google Scholar]
- 12.Bohula EA, Wiviott SD, Giugliano RP, Blazing MA, Park JG, Murphy SA, et al. Prevention of stroke with the addition of ezetimibe to statin therapy in patients with acute coronary syndrome in improve-it (improved reduction of outcomes: Vytorin efficacy international trial). Circulation. 2017;136:2440. [DOI] [PubMed] [Google Scholar]
- 13.Liu CH, Chen TH, Lin MS, Hung MJ, Chung CM, Cherng WJ, et al. Ezetimibe-simvastatin therapy reduce recurrent ischemic stroke risks in type 2 diabetic patients. J Clin Endocrinol Metab. 2016;101:2994–3001. [DOI] [PubMed] [Google Scholar]
- 14.Cui CH, Li YC, Liu SH, Wang P, Huang ZH. The unsupervised machine learning to analyze the use strategy of statins for ischaemic stroke patients with elevated transaminase. Clin Neurol Neurosurg. 2023. 10.1016/j.clineuro.2023.107900. [DOI] [PubMed] [Google Scholar]
- 15.Wang AX, Dai LY, Zhang N, Meng X, Lin JX, Zuo YT, et al. Oxidized low-density lipoprotein and low-density lipoprotein have combined utility in predicting outcomes of minor stroke and tia. Stroke. 2019;50
- 16.Wang A, Zhang X, Li S, Zhao X, Liu L, Johnston SC, et al. Oxidative lipoprotein markers predict poor functional outcome in patients with minor stroke or transient ischaemic attack. Eur J Neurol. 2019;26:1082–90. [DOI] [PubMed] [Google Scholar]
- 17.Cui CH, Li YC, Liu SH, et al. The unsupervised machine learning to analyze the use strategy of statins for ischaemic stroke patients with elevated transaminase. Clin Neurol Neurosurg. 2023;232:7. [DOI] [PubMed] [Google Scholar]
- 18.Cui CH, Li CH, Hou M, et al. The machine learning methods to analyze the using strategy of antiplatelet drugs in ischaemic stroke patients with gastrointestinal haemorrhage. BMC Neurol. 2023;23:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chen ZM, Mo JL, Yang KX, et al. Beyond low-density lipoprotein cholesterol levels: Impact of prior statin treatment on ischemic stroke outcomes. Innovation (Camb). 2024;5:100713. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chen ZM, Gu HQ, Mo JL, et al. U-shaped association between low-density lipoprotein cholesterol levels and risk of all-cause mortality mediated by post-stroke infection in acute ischemic stroke. Sci Bull (Beijing). 2023;68:1327–35. [DOI] [PubMed] [Google Scholar]
- 21.Xu J, Chen Z, Wang M, et al. Low LDL-C level and intracranial haemorrhage risk after ischaemic stroke: a prospective cohort study. Stroke Vasc Neurol. 2023;8:127–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chen Z, Mo J, Xu J, et al. Risk profile of ischemic stroke caused by small-artery occlusion vs. deep intracerebral hemorrhage. Front Neurol. 2019;10:1213. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.