Introduction

Intracranial atherosclerotic stenosis (ICAS), a progressive pathological process that induces cerebral hypoperfusion, is recognized as the most common cause of the occurrence of ischemic stroke globally and more easily leads to the recurrence of ischemic stroke compared with other stroke subtypes.[1,2] Asian, Hispanic, and African populations have a higher risk of intracranial atherosclerosis.[3] In Chinese populations, ICAS induces approximately 50% of transient ischemic attacks (TIAs) and 33% to 50% of strokes.[4] In consideration of the heavy burden that stroke places on society, it is crucial to identify ICAS to improve preventive strategies.

Previous studies have demonstrated that inflammation plays an essential role in the initiation and progression of ICAS. Inflammatory biomarkers such as C-reactive protein (CRP), interleukin-6, and matrix metalloproteinases have been identified as reliable markers of ICAS.[2] The platelet-to-lymphocyte ratio (PLR), which is calculated by dividing the platelet count by the lymphocyte count, has recently been identified as a potential inflammatory marker. What is more, a higher PLR is regarded as a better reflection of atherosclerosis and platelet activation.[5] Studies have proven that the PLR is significantly correlated with the severity of coronary artery disease (CAD).[6–8] Moreover, it was found that the PLR might be a prognostic marker for acute coronary syndrome[9] and acute cerebral infarction.[10] Many prospective epidemiological studies have demonstrated that an increased PLR value is associated with stenosis in the carotid artery.[11] It was also shown that the occurrence of critical limb ischemia was significantly increased with the increase in PLR among patients with the peripheral arterial occlusive disease.[12] However, rare studies to date have focused on the association between the PLR and ICAS. This study aimed to investigate the association between the PLR and ICAS in a Chinese Han population, as well as whether the PLR could be used as a potential biomarker for ICAS.

Methods Ethical approval

This study was approved by the Institutional Ethics Committee of Qingdao Municipal Hospital (No. QDSSLYY-2014-034) and written informed consent was obtained from all participants or their legal representatives.

Study population

The participants were prospectively recruited from Qingdao Municipal Hospital of the patients for suspected stroke in the department of neurology and individuals who underwent comprehensive health screening at the health screening center from January 2014 to June 2018. The inclusion criteria were as follows: (1) aged over 40 years; (2) of Han Chinese ethnicity; (3) underwent systemic investigations, including magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), and other essential laboratory tests; (4) had available comprehensive clinical information. The exclusion criteria were as follows: (1) had been taking antiplatelet drugs before admission that may affect the PLR; (2) had a history of stroke; (3) had an intracerebral hemorrhage, extracranial atherosclerotic stenosis, atrial fibrillation, cardiac embolism, valvular heart disease, vascular disease and had undergone replacement; (4) had intracranial and external artery dissection, muscular fiber dysplasia, arteritis, moyamoya disease; (5) had an infection, tumor, hematologic disease, autoimmune diseases, chronic liver disease, and renal disease.

Finally, a total of 2134 participants were included for analysis in this study (a flowchart of the screening process for the included participants is listed in [Supplementary Figure 1, https://links.lww.com/CM9/B118)]. The majority of the participants were from the department of neurology (the basic characteristics of participants from the two different clinical departments are presented in [Supplementary Table 1, https://links.lww.com/CM9/B118)].

Demographic information and clinical measurements

Demographic information and clinical data were gathered by experienced and certified neurologists. Blood pressure was measured three times consecutively at admission after 15 min of rest in a supine position and the average was taken for analysis. Acute ischemic stroke included TIA and MRI confirmed acute ischemic stroke (within 7 days of onset). An average systolic pressure of at least 140 mmHg and/or an average diastolic pressure of at least 90 mmHg on ≥3 occasions or taking anti-hypertensive drugs was defined as hypertension.[13] The use of anti-diabetic drugs, a random blood glucose level ≥11.1 mmol/L, or a fasting plasma glucose ≥7.0 mmol/L were defined as diabetes mellitus.[14] Coronary heart disease (CHD) was defined as previous or newly diagnosed CHD. Self-reported behaviors of past or present smoking and drinking were defined as smoking and drinking, respectively. Information on medication usage, including lipid-lowering drugs, was obtained through self-report at admission.

Laboratory measurements

Fasting peripheral venous blood samples were collected in the morning within 24 h after admission and all data were obtained from the same blood sample. The lipid profile and fasting blood glucose (FBG) were assayed using an automated analytical platform (Beckman Coulter AU5800: Beckman Coulter Inc., Brea, CA, USA). The platelet count and lymphocyte count were automatically analyzed from EDTA-anticoagulated whole-blood samples using automated particle counters. Following the measurement, the PLR was calculated by dividing the platelet count by the lymphocyte count.

Assessment of ICAS

All enrolled participants had undergone 3D-Time-of-flight magnetic-resonance-angiography (3D-TOF MRA) at 3.0-T. In this study, the Warfarin-Aspirin Symptomatic Intracranial Disease trial criteria were used to evaluate the degree of intracranial stenosis.[15] More than 50% atherosclerotic stenosis or occlusion in single or multiple large intracranial arteries, including the M1/M2 middle cerebral artery, A1/A2 anterior cerebral artery, P1/P2 posterior cerebral artery, vertebral artery, basilar artery, and intracranial section of the internal carotid artery, was considered as ICAS. Two experienced radiologists read all images of the participants in the absence of clinical information and consulted a third neuroimaging physician if there was a dispute.

Statistical analysis

Eleven participants with data values outside 4 standard deviation limits were removed to eliminate the influence of extreme values. The normality of the distribution of continuous variables was tested using the Kolmogorov–Smirnov test. Non-normally distributed continuous variables were expressed as medians and interquartile ranges, and intergroup differences were compared by the Mann-Whitney U test. Categorical variables were expressed as frequencies and percentages, and Chi-squared tests were used to explore the differences. The relationship between the PLR and ICAS was evaluated by multivariate logistic regression models, adjusted by related risk factors (P < 0.05 by univariate logistic regression) including age, systolic blood pressure (SBP), FBG, high-density lipoprotein (HDL), acute ischemic stroke, a medical history of hypertension, a medical history of diabetes, and lipid-lowering drugs. The results are presented as odds ratios (ORs) and 95% confidence intervals (CIs). The trend test was performed by taking the median values of each quartile into the logistic regression model as a continuous variable. The analyses were further conducted in subgroups stratified by age (<60 vs. ≥60 years) and acute ischemic stroke. Moreover, in participants who completed the CRP test, the same analyses described above were conducted by adding CRP in the multivariate-adjusted models. Statistical analyses were performed using R software (version 3.6.3; R Foundation, Vienna, Austria) and SPSS version 21 (IBM Corp., Armonk, NY, United States). The significance level of all analyses was set at P < 0.05.

Results Participant characteristics and intergroup comparisons

A total of 2134 participants were included in this analysis, among whom there were 518 patients with ICAS and 1616 patients without ICAS. The study population had a median age of 67 years and they were more likely to be male (59.9%). The clinical and demographic characteristics of the included participants are presented in [Table 1]. As the results showed, compared with the non-ICAS group, patients with ICAS were older and had higher levels of SBP and FBG, as well as a higher prevalence of acute ischemic stroke, hypertension, and diabetes mellitus, but a lower level of HDL. The PLR was significantly increased among patients with ICAS compared with patients without ICAS. We also found that patients with 1 to 2 stenotic arteries and ≥3 stenotic arteries had significantly higher PLRs than those without ICAS, but the comparison between the group with 1 to 2 stenotic arteries and the group with ≥3 stenotic arteries revealed no significant differences [Supplementary Figure 2, https://links.lww.com/CM9/B118].

Table 1 - Clinical and demographic characteristics of participants with ICAS or without ICAS. Characteristics Total participants (n = 2134) ICAS (n = 518) Non-ICAS (n = 1616) Statistics P values Age, median (IQR), years 67 (60–77) 70 (62–79) 66 (59–76) −5.325∗ <0.001 Male, n (%) 1278 (59.9) 306 (59.1) 972 (60.1) 0.189† 0.664 SBP, median (IQR), mmHg 145 (130–160) 150 (135–170) 144 (130–160) −4.703∗ <0.001 DBP, median (IQR), mmHg 84 (80–90) 85 (80–90) 82 (80–90) −0.911∗ 0.362 Lipid profile, median (IQR)  TG (mmol/L) 1.30 (0.95–1.78) 1.32 (0.98–1.74) 1.30 (0.94–1.78) −0.619∗ 0.536  TC (mmol/L) 4.95 (4.21–5.77) 4.98 (4.13–5.90) 4.93 (4.23–5.73) −0.232∗ 0.817  HDL (mmol/L) 1.14 (0.96–1.35) 1.12 (0.93–1.32) 1.14 (0.98–1.36) −2.376∗ 0.018  LDL (mmol/L) 3.04 (2.48–3.62) 3.08 (2.44–3.64) 3.03 (2.50–3.61) −0.534∗ 0.593 FBG, median (IQR), mmol/L 5.32 (4.69–6.77) 5.69 (4.80–7.83) 5.23 (4.66–6.50) −5.261∗ <0.001 Platelet, median (IQR), (109/L) 205 (173–239) 207 (172.75–244) 205 (173–238) −0.644∗ 0.520 Lymphocytes, median (IQR), (109/L) 1.95 (1.55–2.42) 1.86 (1.45–2.33) 1.99 (1.57–2.46) −3.720∗ <0.001 PLR, median (IQR) 104.76 (82.20–136.45) 110.17 (87.49–146.17) 102.97 (81.77–131.83) −3.620∗ <0.001 Medical history  Hypertension, n (%) 1646 (77.1) 424 (81.9) 1222 (75.6) 8.644† 0.003  Diabetes mellitus, n (%) 751 (35.2) 224 (43.2) 527 (32.6) 19.441† <0.001  CHD, n (%) 791 (37.1) 208 (40.2) 583 (36.1) 2.796† 0.094  Acute ischemic stroke, n (%) 1293 (60.6) 362 (69.9) 931 (57.6) 24.743† <0.001  Smoking, n (%) 728 (34.1) 174 (33.6) 554 (34.3) 0.083† 0.773  Drinking, n (%) 526 (24.6) 127 (24.5) 399 (24.7) 0.006† 0.937  Lipid-lowering drugs, n (%) 247 (11.6) 107 (20.7) 140 (8.7) 55.126† <0.001

∗Mann-Whitney U test.

†Chi-squared test.CHD: Coronary heart disease; DBP: Diastolic blood pressure; FBG: Fasting blood glucose; HDL: High-density lipoprotein; ICAS: Intracranial arterial stenosis; IQR: Interquartile range; LDL: Low-density lipoprotein; PLR: Platelet-to-lymphocyte ratio; SBP: Systolic blood pressure; TC: Total cholesterol; TG: Triglyceride.

In our subgroup analysis stratified by age, all participants were divided into two subgroups: the <60 years group (mid-life group) and the ≥60 years group (late-life group). We found that PLR levels were higher in patients with ICAS than in those without ICAS in the late-life group (P < 0.001), but no significant difference in PLR level was found while comparing patients with and without ICAS in the mid-life group (P = 0.688) [Figure 1A]. Other characteristics of participants in subgroups stratified by age are also shown in [Table 2].

Figure 1:

The level of PLR and distribution of participants. (A) PLR levels were significantly higher in ICAS groups compared to non-ICAS groups in late-life participants, but no statistical significance in mid-life participants. (B) PLR levels were significantly higher in ICAS groups in both stroke and non-stroke participants. ICAS: Intracranial arterial stenosis; PLR: Platelet-to-lymphocyte ratio.

Table 2 - Characteristics of participants in subgroups stratified by age. Mid-life group (n = 519) Late-life group (n = 1615) Characteristics ICAS (n = 94) Non-ICAS (n = 425) P (sta) ICAS (n = 424) Non-ICAS (n = 1191) P (sta) Age, median (IQR), years 53 (51–57) 54 (49–57) 0.483 (−0.701∗) 74 (66–81) 71 (64–79) <0.001 (−3.631∗) Male, n (%) 73 (77.7) 326 (76.7) 0.843 (0.039†) 233 (55.0) 646 (54.2) 0.800 (0.064†) SBP, median (IQR), mmHg 147 (130–160) 140 (130–160) 0.381 (−0.877∗) 150 (135–170) 145 (130–160) <0.001 (−4.686∗) DBP, median (IQR), mmHg 90 (80–96) 90 (80–100) 0.705 (−0.379∗) 84 (80–90) 80 (80–90) 0.066 (−1.837∗) Lipid profile, median (IQR)  TG (mmol/L) 1.49 (1.11–2.06) 1.51 (1.12–2.05) 0.683 (−0.408∗) 1.27 (0.96–1.70) 1.22 (0.89–1.69) 0.092 (−1.687∗)  TC (mmol/L) 4.87 (4.01–5.72) 5.01 (4.16–5.75) 0.395 (−0.850∗) 5.00 (4.15–5.95) 4.90 (4.24–5.72) 0.497 (−0.679∗)  HDL (mmol/L) 1.04 (0.89–1.27) 1.11 (0.95–1.29) 0.124 (−1.538∗) 1.13 (0.94–1.33) 1.16 (0.99–1.38) 0.026 (−2.222∗)  LDL (mmol/L) 3.01 (2.40–3.48) 3.08 (2.46–3.66) 0.279 (−1.082∗) 3.09 (2.47–3.70) 3.01 (2.51–3.58) 0.247 (−1.157∗) FBG, median (IQR), mmol/L 5.72 (4.82–8.07) 5.31 (4.70–6.83) 0.055 (−1.917∗) 5.69 (4.80–7.68) 5.21 (4.65–6.42) <0.001 (−5.081∗) Platelet, median (IQR), (109/L) 209 (178–244) 210 (180–248) 0.575 (−0.561∗) 207 (172–244) 202 (171–234) 0.210 (−1.254∗) Lymphocytes, median (IQR), (109/L) 2.05 (1.67–2.59) 2.14 (1.72–2.55) 0.502 (−0.671∗) 1.82 (1.39–2.28) 1.93 (1.53–2.40) 0.001 (−3.293∗) PLR, median (IQR) 101.71 (82.26–124.72) 97.89 (79.40–126.08) 0.688 (−0.401∗) 113.65 (87.67–150.33) 104.32 (82.33–133.12) <0.001 (−3.651∗) Medical history  Hypertension, n (%) 73 (77.7) 304 (71.5) 0.228 (1.455†) 351 (82.8) 918 (77.1) 0.014 (6.045†)  Diabetes mellitus, n (%) 41 (43.6) 141 (33.2) 0.055 (3.685†) 183 (43.2) 386 (32.4) <0.001 (15.837†)  CHD, n (%) 23 (24.5) 87 (20.5) 0.391 (0.736†) 185 (43.6) 496 (41.6) 0.477 (0.506†)  Acute ischemic stroke, n (%) 72 (76.6) 257 (60.5) 0.003 (8.625†) 290 (68.4) 674 (56.6) <0.001 (18.111†)  Smoking, n (%) 56 (59.6) 215 (50.6) 0.114 (2.491†) 118 (27.8) 339 (28.5) 0.804 (0.062†)  Drinking, n (%) 39 (41.5) 154 (36.2) 0.340 (0.910†) 88 (20.8) 245 (20.6) 0.936 (0.006†)  Lipid-lowering drugs, n (%) 8 (8.5) 19 (4.5) 0.110 (2.548†) 99 (23.3) 121 (10.2) <0.001 (46.229†)

∗Mann-Whitney U test.

†Chi-squared test.CHD: Coronary heart disease; DBP: Diastolic blood pressure; FBG: Fasting blood glucose; HDL: High-density lipoprotein; ICAS: Intracranial arterial stenosis; IQR: Interquartile range; LDL: Low-density lipoprotein; PLR: Platelet-to-lymphocyte ratio; SBP: Systolic blood pressure; sta: Statistics; TC: Total cholesterol; TG: Triglyceride.

We also conducted a subgroup analysis stratified by acute ischemic stroke (stroke group and non-stroke group). The characteristics of participants in subgroups stratified by stroke status are summarized in [Supplementary Table 2, https://links.lww.com/CM9/B118]. In the non-stroke group, PLR levels were higher in patients with ICAS than in those without ICAS (P = 0.017). Similar results were also found in the stroke group (P = 0.005) [Figure 1B].

Association of the PLR and ICAS in all recruited participants

Findings from multivariable logistic regression for the association between the PLR and ICAS are provided in [Table 3]. After adjustment for confounding factors, the PLR was demonstrated to be associated with ICAS in all participants (OR 1.005, 95% CI 1.003–1.007, P < 0.001). Further analyses were performed using quartiles of the PLR. The regression analyses indicated that patients in higher PLR quartiles were more inclined to suffer ICAS than those in the first quartile. Specifically, with the first quartile as the reference category, the ORs and 95% CIs were 1.000 (0.739–1.352) in the second quartile, 1.123 (0.833–1.513) in the third quartile, and 1.705 (1.278–2.275) in the fourth quartile [Figure 2A]. The fourth (P < 0.001) PLR quartile was significantly associated with ICAS. We also conducted the trend test and finally obtained the P value for trend (P < 0.001) which was significant, and the test also indicated a dose–response relationship between the PLR and the risk of suffering ICAS. In addition, repeated analyses were performed separately in participants from the two different clinical departments, and similar results were obtained [Supplementary Table 3, https://links.lww.com/CM9/B118].

Table 3 - Multivariate-adjusted OR and 95% CI for ICAS in mid-life and late-life groups. Total (n = 2134) Mid-life group (n = 519) Late-life group (n = 1615) Characteristic Univariate, OR (95% CI) P value Multi-OR (95% CI) P value Multi-OR (95% CI) P value Multi-OR (95% CI) P value Age 1.024 (1.015–1.033) <0.001 1.020 (1.011–1.030) <0.001 1.026 (0.979–1.077) 0.288 1.017 (1.003–1.032) 0.015 Male 1.046 (0.854–1.278) 0.664 – – – – – – SBP 1.010 (1.006–1.015) <0.001 1.006 (1.001–1.011) 0.021 0.998 (0.987–1.008) 0.666 1.008 (1.003–1.013) 0.004 DBP 0.998 (0.992–1.001) 0.300 – – – – – – FBG, mmol/L 1.094 (1.054–1.134) <0.001 1.070 (1.020–1.123) 0.006 1.015 (0.991–1.126) 0.787 1.093 (1.034–1.156) 0.002 TG, mmol/L 0.985 (0.895–1.069) 0.737 – – – – – – TC, mmol/L 1.023 (0.947–1.104) 0.562 – – – – – – HDL, mmol/L 0.617 (0.445–0.834) 0.003 0.629 (0.452–0.846) 0.004 0.646 (0.273–1.321) 0.288 0.640 (0.447–0.881) 0.010 LDL, mmol/L 0.994 (0.926–1.017) 0.710 – – – – – – Hypertension 1.454 (1.136–1.877) 0.003 1.201 (0.914–1.588) 0.193 1.395 (0.776–2.579) 0.276 1.154 (0.847–1.585) 0.369 Diabetes 1.574 (1.285–1.927) <0.001 1.150 (0.878–1.502) 0.307 1.292 (0.675–2.434) 0.432 1.107 (0.820–1.489) 0.502 CHD 1.189 (0.970–1.455) 0.095 – – – – – – Acute ischemic stroke 1.707 (1.383–2.115) <0.001 1.601 (1.282–2.006) <0.001 2.119 (1.261–3.678) 0.006 1.503 (1.174–1.930) 0.001 Smoking 0.970 (0.785–1.194) 0.773 – – – – – – Drinking 0.991 (0.785–1.244) 0.937 – – – – – – Lipid-lowering drugs 2.745 (2.083–3.609) <0.001 2.219 (1.652–2.975) <0.001 1.592 (0.618–3.785) 0.309 2.333 (1.702–3.196) <0.001 PLR 1.005 (1.003–1.007) <0.001 1.005 (1.003–1.007) <0.001 1.001 (0.995–1.007) 0.650 1.006 (1.003–1.008) <0.001

CHD: Coronary heart disease; CI: Confidence interval; DBP: Diastolic blood pressure; FBG: Fasting blood glucose; HDL: High-density lipoprotein; LDL: Low-density lipoprotein; Multi-OR: Multivariate OR; OR: Odds ratio; PLR: Platelet-to-lymphocyte ratio; SBP: Systolic blood pressure; TC: Total cholesterol; TG: Triglyceride.

Figure 2:

Multivariate-adjusted OR and 95% CI for ICAS according to PLR levels. (A) All participants, (B) late-life group, (C) mid-life group, (D) stroke group, and (E) non-stroke group. CI: Confidence interval; ICAS: Intracranial arterial stenosis; OR: Odds ratio; PLR: Platelet-to-lymphocyte ratio.

To remove the potential impact of CRP on the relationship between the PLR and ICAS, sensitivity analysis was further conducted on participants who completed the CRP test. In this analysis, we added CRP to the multivariate-adjusted models and found that the results remained stable [Supplementary Table 4, https://links.lww.com/CM9/B118].

Association of the PLR and ICAS in the mid-life and late-life groups

In the late-life group, the PLR was a significant risk factor for ICAS in regression analysis (OR 1.006, 95% CI 1.003–1.008, P < 0.001). However, no association between the PLR and ICAS was observed in the mid-life group (OR 1.001, 95% CI 0.995–1.007, P = 0.650). When referenced to the first quartile, higher PLR levels had a significant association with the risk of ICAS in the late-life group. The OR was 0.893 (95% CI 0.635–1.257; P = 0.518) for the second quartile, 1.106 (95% CI 0.791–1.546; P = 0.556) for the third quartile, and 1.831 (95% CI 1.327–2.527; P < 0.001) for the fourth quartile (P for trend < 0.001) [Figure 2B]. However, no significant association mentioned above was found in the mid-life group [Figure 2C]. The results indicated that the relationship between the PLR and ICAS did differ significantly in different age subgroups.

Association of the PLR and ICAS in the stroke and non-stroke groups

In the multivariate logistic regression model, an elevated PLR was associated with ICAS in the stroke (OR 1.005, 95% CI 1.002–1.008, P < 0.001) and non-stroke groups (OR 1.005, 95% CI 1.001–1.009, P = 0.027) [Supplementary Table 5, https://links.lww.com/CM9/B118]. In addition, taking the first quartile as a reference, the fourth PLR quartile had the most significant association with ICAS in both the stroke (OR 1.698, 95% CI 1.198–2.407, P = 0.003) [Figure 2D] and non-stroke groups (OR 1.809, 95% CI 1.075–3.045, P = 0.025) [