Impact of homocysteine levels on clinical outcome in patients with acute ischemic stroke receiving intravenous thrombolysis therapy

Background The purpose of this study was to retrospectively assess the potential correlation between clinical outcomes and homocysteine (Hcy) levels in acute ischemic stroke (AIS) patients after recombinant tissue plasminogen activator (rtPA) treatment. Methods AIS patients treated by rtPA were enrolled between September 2018 and March 2019 in the Stroke Center (Department of Neurology and Neurosurgery), Shanghai Tenth People’s Hospital, Tongji University School of Medicine. Demographics, baseline and clinical characteristics, and modified Rankin Scale (mRS) score after three months from the onset were retrospectively analyzed. Then we compared data about demographics, baseline and clinical characteristics between patients with favorable (mRS score 0–2) and unfavorable (mRS score 3–6) outcomes. Results Among 141 patients, 36 patients had poor outcome, for an incidence of 25.53%. Univariate analysis showed that higher Hcy levels (OR = 1.07, 95% CI [1.02–1.12]), older age (OR = 1.06, 95% CI [1.02–1.10]), longer door to needle time (DNT) (OR = 1.03, 95% CI [1.01–1.05]), higher D-Dimer levels (OR = 1.33, 95% CI [1.03–1.71]), and higher National Institutes of Health Stroke Scale (NIHSS) score before treatment (OR = 1.21, 95% CI [1.08–1.35]) were each associated with poor outcome. Also, without internal carotid artery plaque (OR = 0.30, 95% CI [0.10–0.92]) showed a protective effect on patients’ clinical outcome. Patients with higher levels of Hcy decline also showed an increased risk of poor outcome for AIS patients obtaining rtPA treatment (Non-adjusted: OR = 1.07, 95% CI [1.02–1.12]; Adjust model I adjusts for demographics (age, male): OR = 1.06, 95% CI [1.02–1.11]; Adjust model II adjusts for hospital care factors (onset to treatment, DNT): OR = 1.08, 95% CI [1.03–1.13]; Adjust model III adjusts for health and stroke factors (INR, D-Dimer, HGB, NIHSS score before treatment, smoking, drinking, hypertension, diabetes, coronary disease, hyperlipidemia, previous stroke, atrial fibrillation, hemorrhagic transformation, internal carotid artery plaque): OR = 1.06, 95% CI [1.02–1.11]). The results are very stable in all three models constructed. Conclusion The results of this study indicate that increased Hcy level independently predicts unfavorable outcome in AIS patients accepting thrombolytic therapy. However, the contribution of Hcy to the outcome, although significant, is relatively small and perhaps not clinically significant when all the other confounders are considered.


INTRODUCTION
Acute ischemic stroke (AIS) is one of the principal causes of disability and death, which causes great economic and mental burden to patients' families (Merlino et al., 2019;Rosenbaum Halevi et al., 2019). Intravenous thrombolytic therapy and recombinant tissue plasminogen activator (rtPA) have been broadly used in AIS patients within 4.5 h after symptom onset (Valdes Hernandez et al., 2014).
Previous studies have shown that those who have hyperhomocysteinemia (Hhcy) have a higher risk of the development of AIS, and Hhcy is correlated with poor prognosis in AIS patients (Davis Armstrong et al., 2018;Zaric et al., 2019). Besides, the increased risk of cardiovascular disease could also be partially attributed to Hhcy (Borowczyk et al., 2019). Its possible underlying mechanism is that elevated serum homocysteine (Hcy) may lead to endothelial dysfunction (Esse et al., 2019;Wu et al., 2019;Yan et al., 2019), neurotoxicity (Moretti & Caruso, 2019) and up-regulation of thrombosis formation factors (Diao et al., 2019;Jin et al., 2018). However, the likelihood of applying Hcy to prognosticate AIS patients' clinical outcomes after rtPA therapy hasn't been thoroughly investigated. Therefore, we designed this retrospective study to assess the potential relationship between clinical outcomes and Hcy in patients with AIS after rtPA therapy.

Patients
The study was designed as a retrospective study. Patients with AIS taking rtPA treatment were enrolled between September 2018 and March 2019 in a single Stroke Center (Department of Neurology and Neurosurgery, Shanghai Tenth People's Hospital, Tongji University School of Medicine). The study population included 205 consecutive patients admitted with AIS within 4.5 h of their symptom onset.

Exclusion criteria
Evidence of severe infection, cardiopulmonary disease, cancer, hepatic or renal disease, or multiple organ dysfunction. Patients with mental disorders, severe cognitive dysfunction, a history of mental problems and abnormal behavior were also excluded.

Data and design
The institutional ethics committee of Shanghai Tenth People's Hospital has approved this retrospective study. All investigations and methods were performed in accordance with Shanghai Tenth People's Hospital's guidelines and regulations. The consent form was not required due to the retrospective nature and de-identified nature of this retrospective study. The standard cardiological and neurological examinations were performed immediately on the patients' arrival in the emergency room. Before the initiation of therapy, the information about vital signs, blood chemistry and computed tomography scans was obtained. Information about previous and concomitant diseases were recorded. National Institutes of Health Stroke Scale (NIHSS) score was assessed (Brott et al., 1989). Adverse events of hemorrhage, including symptomatic intracerebral hemorrhage (sICH), were identified.
All patients underwent intravenous (i.v.) thrombolytic therapy using rtPA at a dose of 0.9 mg/kg. And those who were followed up for at least three months and had the available modified Rankin Scale (mRS) score (Van Swieten et al., 1988) at three months from symptom onset were involved in the final retrospective analysis. Then we compared data about demographics, baseline and clinical characteristics between patients with favorable (mRS score 0-2) and unfavorable (mRS score 3-6) outcomes.

Statistical analysis
Applying the Kolmogorov-Smirnov test to test if the metrological data followed the normal distribution or not. Mean ± standard deviation (SD) was used to express continuous variables that follow the normal distribution, and the median (quartiles) was used to indicate variables that do not follow a normal distribution. Frequencies or percentages were used to indicate categorical variables.
Chi-Square tests and Kruskal-Wallis H test or one-way ANOVA test were applied to detect any statistical difference between the means and proportions of the three groups.
Univariate analysis of baseline characteristics and clinical outcome were carried out. In this process, variables were added to the model with recoding into tertiles/binary or without recoding as needed.
Multiple logistic regression models were utilized to assess the association between serum Hcy levels and clinical outcome. Both non-adjusted and multivariate-adjusted models (variables adjusted for demographics (age, male)); hospital care factors (onset to treatment, door to needle time (DNT)); health and stroke factors (INR, D-Dimer, hemoglobin (HGB), NIHSS score before treatment, smoking, drinking, hypertension, diabetes, coronary disease, hyperlipidemia, previous stroke, atrial fibrillation, hemorrhagic transformation, internal carotid artery plaque) were applied. In this step, multivariate logistic regression models with the conditional forward selection method were used to reduce confounding factors' contribution and evaluate the independent contribution of Hcy to clinical outcome after adjusting different factors. All analyses were performed with the SPSS 22.0. A two-sided significance level of 0.05 was applied to assess statistical significance.

General patient characteristics
Among 205 consecutive patients with AIS, 141 (68.78%) patients' follow-up data could be analyzed. Among them, 36 patients had poor outcomes, for an incidence of 25.53%.
The data about baseline clinical findings, demographic characteristics and medical history records were summarized in Table 1.

Subgroup analysis
We further explored the role of other covariables on the association between Hcy levels and outcome. The impact of Hcy levels on outcome exhibited a significant difference in DNT subgroups (P = 0.037), whereas no difference in the following subgroups: age, gender, onset to treatment, INR, D-Dimer, smoking, drinking, hypertension, diabetes, coronary disease, hyperlipidemia, previous stroke, atrial fibrillation, hemorrhagic transformation, internal carotid artery plaque, HGB and NIHSS score before treatment (all P value for interaction is more than 0.05) ( Table 3).  The results are very stable in all three models constructed (Table 4).

DISCUSSION
Currently, intravenous thrombolysis (IVT) with alteplase is the standard treatment approved for AIS patients within 4.5 h of the onset of their symptoms (Valdes Hernandez et al., 2014). In this study, we found that high level of Hcy, as a predictor for poor prognosis at three months in AIS patients receiving rtPA therapy, has important clinical implications. We also demonstrated that the risk of poor prognosis in AIS patients taking IVT increased by approximately 10% for every 1 mmol/L increase in serum Hcy concentration.
In the past decade, an enormous amount of epidemiological evidence supports the correlation between high Hcy levels and increased risk of the development of AIS (Cheng et al., 2018;Lu et al., 2018b), and some researchers have discovered that increased serum Hcy levels are related to increased hematoma volume (Hacke et al., 2008;Zhou et al., 2015). Previous studies also demonstrated that increased Hcy levels are correlated with functional disability in the acute phase of stroke (Mizrahi et al., 2005;Song et al., 2009). A prospective multicenter research conducted by Kwon et al. unveiled that the risk of early neurological deterioration increased along with the increase of Hcy levels in patients with ischemic stroke (Kwon et al., 2014). However, whether high Hcy levels can be regarded as an independent risk factor for unfavorable clinical outcome in ischemic stroke patients accepting IVT has not been well addressed. Though a small number of studies have already been published, the prognostic value of Hcy levels in patients with AIS after IVT remains controversial. Some investigations revealed that Hhcy was correlated with unfavorable outcomes in patients with ischemic stroke (Ling et al., 2018;Luo et al., 2019;Yao et al., 2016), whereas some other studies indicated that there was no significant correlation between Hhcy and ischemic stroke patients' clinical outcome (Ribo et al., 2004a(Ribo et al., , 2004b. To further confirm the association between Hcy levels and clinical outcome of AIS patients treated with thrombolytic therapy, we conducted this retrospective study. Consistent with the previous retrospective studies conducted by Ling et al. (2018), Luo et al. (2019) and Yao et al. (2016), we found that there exists a correlation between Hcy levels and poor prognosis after acute thrombolytic therapy in AIS patients in this retrospective study. And we indicated that higher Hcy levels have a negative impact on prognosis. The result of our study implements new proof that Hhcy has a negative impact on ischemic stroke patients' clinical outcome.
The underlying mechanism may be due to impaired vascular wall integrity and disturbance of cerebrovascular permeability resulting from increased levels of Hcy, which may lead to endothelial dysfunction, damage to elastic structures and damage to the basal layer of cerebral arterioles and microvessels Mach et al., 1997).
Some studies have consistently shown that hyperhomocysteinemia is an independent risk factor for atherosclerosis' development, suggesting that raised plasma levels of Hcy are relevant to endothelial dysfunction (Borowczyk et al., 2019;Lu et al., 2018a;Wang et al., 2017). Besides, high Hcy levels also increase low-density lipoproteins oxidation, and the dominant mechanism by which Hcy adversely affects vascular endothelial function involves oxidative stress and bioactive nitric oxide consumption (Miyazaki et al., 2014;Seo et al., 2010).
Our research also has some restrictions. First, this study is a single-center retrospective study and the sample size is limited. Second, we did not include patients who did not receive thrombolytic therapy, thus may resulting in a selective bias. Third, the Hcy level could be affected by various factors, such as genetic factors and drugs. But we didn't evaluate the reason for Hhcy in this study cohort. Also, almost all ORs for Hcy are close to 1 in our results. The contribution of Hcy to the outcome, although significant, is relatively small and perhaps not clinically significant when considering all other confounders. Hence, the results should be explained with caution, and the results should be further confirmed in a multicenter prospective study with a larger cohort to clearly establish the correlation between Hhcy and unfavorable outcome in ischemic stroke patients accepting IVT.

CONCLUSIONS
In conclusion, the results of this study indicate that increased Hcy level independently predicts unfavorable outcome in AIS patients accepting thrombolytic therapy. However, the contribution of Hcy to the outcome, although significant, is relatively small when all the other confounders are considered. To better guide clinical practice, the further multicenter prospective study still needs to be done to clearly clarify the correlation between Hcy level and clinical outcome of AIS patients treating with intravenous thrombolysis.

ADDITIONAL INFORMATION AND DECLARATIONS Funding
This work was supported by the Shanghai Shenkang Hospital Development Center (No. SHDC12017X17). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Sajan Pandey performed the experiments, authored or reviewed drafts of the paper, and approved the final draft. Ronghao Wan analyzed the data, prepared figures and/or tables, and approved the final draft. Rui Shen conceived and designed the experiments, authored or reviewed drafts of the paper, and approved the final draft. Quanbin Zhang conceived and designed the experiments, authored or reviewed drafts of the paper, and approved the final draft.

Human Ethics
The following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers): The study was approved by the Institutional Ethics Committee of Shanghai Tenth People's Hospital.

Data Availability
The following information was supplied regarding data availability: The raw data are available in Table S1.

Supplemental Information
Supplemental information for this article can be found online at http://dx.doi.org/10.7717/ peerj.9474#supplemental-information.