Lymphocyte-to-Monocyte Ratio Is Independently Associated with Progressive Infarction in Patients with Acute Ischemic Stroke

Methods From April 2017 to December 2020, we retrospectively recruited 477 patients with acute ischemic stroke (within 48 hours after onset). Progressive infarction was defined as an increase of ≥1 point in motor power or ≥2 points on the total National Institutes of Health Stroke Scale (NIHSS) within 7 days after admission and extension of the original infarction were further confirmed by diffusion-weighted imaging. Demographic characteristics, clinical information, and neuroimaging characteristics were evaluated after admission. All blood draws and initial imaging were completed within 24 hours of admission. Results PI occurred in 147 (30.8%) patients. Univariate analysis comparing the two groups revealed that hypertension, initial NIHSS score, discharge NIHSS score, modified Rankin scale score at 90 days, monocyte level, creatinine level, fasting glucose level, LMR, monocyte-to-high-density lipoprotein ratio (MHR), and lesion location were significantly different (P < 0.05). Multivariate logistic regression analysis showed that the odds ratio of PI increased as the quartile of LMR increased, with the lowest quartile as the reference value. Subgroup analyses showed that a high LMR was an independent predictor of PI only in large artery atherosclerosis (LAA) patients. The receiver operating characteristic (ROC) curve was drawn to estimate the predictive value of LMR for PI. For all cases, the area under the curve was 0.583 (95% CI 0.526-0.641), and the best predictive cutoff value was 3.506, with a sensitivity of 53.1% and a specificity of 63.9%. In patients with LAA, the area under the curve was 0.585 (95% CI 0.505-0.665), and the best predictive cutoff value was 3.944, with a sensitivity of 48.7% and a specificity of 72.8%. Conclusions LMR was an independent predictor for progressive infarction in patients with acute ischemic stroke, especially in LAA cerebral infarction patients.


Introduction
Ischemic stroke is a global disease and the leading cause of long-term disability in adults [1], and China faces the greatest challenge from stroke in the world. The Global Burden of Disease Study (GBD) 2019 estimated that deaths caused by stroke in China reached approximately 1.57 million in 2018 [2]. Approximately one-third of patients with acute ischemic stroke (AIS) experience early neurological deterioration (END) with poor outcomes [3]. END, a symptomatic diagnosis, has been defined as an increase of ≥1 point in motor power or ≥2 points in the total National Institute of Health Stroke Scale (NIHSS) score within 7 days after admission. However, various mechanisms are involved in END, including progressive infarction (PI) representing extension of the original infarction, increased intracranial pressure, recurrent cerebral ischemia on behalf of new stroke, and secondary parenchymal hemorrhage [3][4][5]. In recent years, many predictors have been associated with END, such as the initial National Institute of Health Stroke Scale (NIHSS) score [6], the triglyceride-glucose index [7], high-sensitivity C-reactive protein [8], albuminuria [9], and fibrinogen [10].
Recently, the lymphocyte-to-monocyte ratio (LMR), as a surrogate marker of systemic inflammation, and its high or low value, represents the degree of inflammatory response [11,12]. LMR has been found to be associated with poor prognosis in various malignancies [13,14]. Quan et al. found that the LMR value might be associated with higher shortterm and long-term mortality in acute coronary syndrome patients [15]. In studies of AIS, LMR at admission was independently associated with poor 3-month outcome [12] and higher risk for hemorrhagic transformation in patients with AIS [16]. A study by Gong et al. [17] showed that LMR was associated with postthrombolysis END. However, to the best of our knowledge, there are few studies on the relationship between LMR and PI.
PI is the most common cause of END in acute ischemic stroke, accounting for approximately 30% of cases [4]. Since LMR has attracted increasing attention in recent years, the aim of this work is to investigate the relationship between LMR and PI.

Patients.
We retrospectively collected inpatients diagnosed with AIS in the Stroke Unit of First Affiliated Hospital of Nanchang University between April 2017 and December 2020. All procedures were approved by the Ethics Committee of First Affiliated Hospital of Nanchang University. Patients were recruited if they met the following criteria: (1) admission to our hospital within 48 hours after symptom onset and (2) evidence of cerebral infarction on diffusionweighted imaging (DWI) consistent with the clinical deficit. Patients were excluded from this study if they (1) were lacking complete imaging, laboratory, and follow-up data, (2) had fever or infection on admission or had a history of immune system diseases, or (3) had received intravenous thrombolysis or endovascular therapy.

Clinical Characteristics and Laboratory Parameters.
Demographic characteristics and clinical information,  including age, sex, history of hypertension and diabetes, initial NIHSS score, discharge NIHSS score, and modified Rankin scale (mRS) score at 90 days, were recorded. The mRS was evaluated in person or over the telephone. We mainly included laboratory variables based on the following two points. On the one hand, we included most basic laboratory indicators. On the other hand, we collected variables related to END reported in the previous studies. Finally, the levels of white blood cells, red blood cells, hemoglobin, blood platelets, lymphocytes, monocytes, neutrophilic granulocytes, fibrinogen, D-dimer, blood urea nitrogen, creatinine, uric acid, total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, LMR, monocyte-to-high-density lipoprotein ratio (MHR), neutrophil-tolymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) were collected within 24 hours of admission.    images were reviewed and evaluated by two trained neurologists who were blinded to the patients' information. Leukoaraiosis was analyzed by a 4-point score as proposed by Fazekas et al. [18]. To determine the subtype of ischemic stroke, [19] criteria were used [20].

Definition of PI.
Progressive infarction (PI) was defined as an increase of ≥1 point in motor power or ≥2 points on the total NIHSS within 7 days after admission and extension of the original infarction were further confirmed by diffusion-weighted imaging [21]. All patients were grouped according to the definition of PI after careful case review by a stroke physician.

Results
From April 2017 to December 2020, 3373 AIS patients admitted within 48 hours onset were screened according to inclusion criteria and exclusion criteria (Figure 1), and a total of 477 eligible patients were enrolled, including 147 (30.8%) in the PI group and 330 (69.2%) in the non-PI (nonprogressive infarction) group. The baseline data of the two groups are listed in Table 1. We found that hypertension, initial NIHSS score, discharge NIHSS score, mRS score at 90 days, monocyte level, creatinine level, fasting glucose level, LMR, MHR, and lesion location were significantly different between groups (P < 0:05). There were no differences in age, sex, history of diabetes, white blood cell count, red blood cell count, platelets, total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, fibrinogen, D-dimer, TOAST classification, or leukoaraiosis between the two groups (P > 0:05). As shown in Table 2, the highest quartile of LMR (the first quartile as the reference value) was independently associated with PI, as demonstrated by multivariate logistic regression analysis done after grouping LMR by quartile.

BioMed Research International
The multivariable logistic regression model adjusted for age and sex revealed that the highest LMR quartiles were independently associated with PI (OR 2.144, 95% CI 1.211-3.796, P = 0:009). The second adjustment for hypertension, initial NIHSS score, creatinine level, fasting glucose level, and lesion location also suggested that a high LMR was independently associated with PI (OR 2.362, 95% CI 1.326-4.208, P = 0:004). After adjusting for confounding factors such as hypertension, initial NIHSS score, creatinine level, blood glucose level, lesion location, blood urea nitrogen level, uric acid level, TOAST classification, and leukoaraiosis, LMR was still significantly associated with PI (OR 2.598, 95% CI 1.405-4.802, P = 0:002). The results of the subgroup analyses are shown in Table 3. LMR was an independent risk factor for PI in patients with LAA (OR 2.335, 95% CI 1.051-5.191, P = 0:037). It had no correlation in patients with SVO (P > 0:05).
The receiver operating characteristic (ROC) curve was drawn to estimate the predictive value of LMR on PI ( Figure 2). For all cases, we observed that the area under the curve was 0.583 (95% CI 0.526-0.641), and the best predictive cutoff value was 3.506, with a sensitivity of 53.1% and a specificity of 63.9%. In patients with LAA, the area under the curve was 0.585 (95% CI 0.505-0.665), and the best predictive cutoff value was 3.944, with a sensitivity of 48.7% and a specificity of 72.8%. When we divided patients into two groups around the cutoff value (Table 4), the frequency of PI was 24.6% in patients with LMR < 3:506 and 39.6% in patients with LMR ≥ 3:506 (P < 0:001). Of note, after dividing LAA patients into two groups around their optimal cutoff value of 3.944, the analysis demonstrated that the frequency of PI was 26.2% in patients with LMR < 3:944 and 47.4% in patients with LMR ≥ 3:944 (P = 0:001).

Discussion
END occurs in approximately one-third of AIS patients. The definition of END is based on symptomatic changes (an increase of ≥1 point in motor power or ≥2 points in the total NIHSS score within 7 days after admission) [22]. However, causes of END include PI representing extension of the original infarction, increased intracranial pressure, recurrent cerebral ischemia on behalf of new stroke, and secondary cerebral parenchymal hemorrhage.
Among them, PI accounts for approximately 33.6% of END cases [4]. In our study, we purely focused on the relationship between LMR and PI and defined progressive infarction as an increase of ≥1 point in motor power or ≥2 points on the total NIHSS within 7 days after admission but the enlargement of the original infarction was confirmed by diffusion-weighted imaging. In our study, we purely focused on the association between serum inflammatory factors and PI and found that LMR was a predictor of PI in AIS patients, especially in patients with LAA. In the whole cohort, patients with LMR ≥ 3:506 were more likely to develop PI, and LMR ≥ 3:944 increased the risk of PI in patients with LAA.
Among the lymphocytes, T lymphocytes have a damaging role in the process of stroke [23]. In this subgroup of cells, type 1 helper T cells (Th1) are proinflammatory, and type 2 helper T cells (Th2) are anti-inflammatory [24]. Some studies have also reported the damaging effect of doublenegative T lymphocytes in stroke and the protective effect of regulatory T cells on nerves [25,26]. Although the role of lymphocytes in stroke is controversial, decreased lymphocyte counts after stroke have been demonstrated [27,28], which is consistent with the results of this study. Monocytes have a pivotal role in the systemic inflammatory response; when they leave the circulatory system to reach tissues, they can differentiate into macrophages to play an inflammatory role [29]. The monocytes present during cerebral ischemia mainly come from the spleen, and the spleen-derived monocytes are also the first to enter the ischemic focus [30]. This finding was also confirmed in rats with middle cerebral artery occlusion by Seifert et al. [31]. The peripheral blood mononuclear cell count may be reduced due to the rapid recruitment of monocytes to the location of cerebral ischemia. Traditional monocytes fall into three categories: classic monocytes (CD14 hi CD16monocytes), non-classic monocytes (CD14 dim CD16 + monocytes), and intermediate monocytes (CD14 hi CD16 + monocytes) [32][33][34]. Studies have reported that the levels of different types of monocytes increase at different time points after stroke, intermediate monocytes are the main monocytes in PI, and their levels begin to increase 3-16 days after onset [35]. This finding provides evidence for a decrease in peripheral blood mononuclear count within 24 hours after the onset of progressive infarction.
Recent studies have reported that low LMR values have an impact on the outcome and prognosis of stroke patients [12,36]. Cai et al. found that a lower LMR at admission was independently associated with long-term all-cause mortality and major adverse cardiac events after discharge in patients with ST-elevation myocardial infarction [37]. However, the odds of cardiovascular events in patients with early coronary heart disease manifestations were the highest in patients with an elevated LMR [38]. Lombardi et al. also found that a higher LMR is directly associated with any type of clinical diagnosis of dementia [39]. Additionally, our study confirmed that a higher LMR value was an independent risk factor for PI. Of note, some studies on the correlation between LMR and END have used END as an outcome variable, which may reduce the specificity of the predictor. In our study, we focused only on the association between LMR and PI.
There are different mechanisms of ischemic stroke, and they are associated with different incidences of END [40]. According to the TOAST classification, patients with ischemic stroke caused by noncardiac embolism are more likely to worsen than patients with cardioembolic infarction [41]. However, no studies have reported the effect of LMR on AIS classification. Atherosclerosis is an inflammatory disease, and intensified inflammatory activation may lead to local proteolysis, plaque rupture, and thrombus formation, which causes ischemia and infarction [42]. However, the pathogenesis of ischemic stroke in SVO is mostly caused by small vessel disease (lipohyalinosis or fibrinoid necrosis) and has no obvious relationship with the occurrence and development of inflammatory diseases [43]. Therefore, LMR, as a novel biomarker of systemic inflammation, may have a greater impact on LAA patients. Then, we analyzed the relationship between LMR and PI in patients with different mechanisms or different lesion topographies and found that LMR was an independent predictor of PI only in patients with LAA cerebral infarction.
This study has some limitations. First, this was a single-center retrospective study of patients of a single ethnic background. Second, few patients had cardioembolism, undetermined etiology, or other etiology under the TOAST classification, so we only analyzed the relationship between LMR and LAA-type and small-artery-occlusiontype PI patients. Subsequent studies need to further expand the sample size and collect multicenter and multiethnic patients to confirm the findings. Finally, we only recorded the LMR from the first blood draw and did not monitor the LMR for the rest of the hospital stay or at discharge. There is some controversy about the relationship between LMR levels and END, so further research needs to continuously monitor the LMR value to determine whether the LMR is increased or decreased in END patients.
In conclusion, LMR was an independent predictor of progressive infarction in patients with acute ischemic stroke, especially LAA cerebral infarction. What is more, patients with progressive infarction tend to have worse outcome. Therefore, it is important to identify the predictors for progressive infarction.

Data Availability
Data is available on request from the authors.

Ethical Approval
This study was approved by the Ethics Committee of First Affiliated Hospital of Nanchang University (Medical Research Ethic Fast Review No. 3-004).

Consent
Informed consent of patients in this study was not required.

Conflicts of Interest
The authors declares that they have no conflicts of interest.