Relationships among Neutrophil-to-Lymphocyte Ratio, Serum Albumin Level, and Platelet-to-Lymphocyte Ratio with In-stent Restenosis in Cerebrovascular Bare-Metal Stenting

Background: In-stent restenosis (ISR) remains a serious clinical problem that affects the ecacy of endovascular treatment. Inammation and platelet activation play an important role in the development of ISR after cerebrovascular bare-metal stenting. The neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR), which are the indicators of inammation, have been associated with stent restenosis, while low serum albumin (ALB) level has been associated with the adverse outcomes of cardiovascular diseases. We therefore explored the associations among preprocedural NLR, ALB level, and PLR with ISR after cerebrovascular bare-metal stenting and then investigated the value of preprocedural NLR, ALB level, and PLR for the development of ISR. Methods: In this study, we enrolled 223 patients who underwent cerebrovascular bare-metal stenting (both intracranial and extracranial) and retrospectively analyzed their clinical, hematological, and angiographic data. Radiographic evaluations were performed at 6 and 12 months after the procedure. Results: ISR was dened as ≥ 50% stenosis of the treated lesion. Patients were divided into two groups based on the occurrence of ISR. ISR occurred in 89 (39.91%) of the 223 included patients. Increased preprocedural NLR (p < 0.001) and PLR (p < 0.001) and decreased pre-procedural ALB level (p < 0.001) were signicantly associated with ISR. Conclusions: High pre-procedural NLR and PLR, and low pre-procedural ALB level may be used as predictors of ISR after cerebrovascular bare-metal stenting. Trial Registration: retrospective registration.


Introduction
Stroke is the leading cause of death and disability among Chinese adults 1 . About 3 million people suffer a stroke every year in China. Notably cerebrovascular atherosclerotic stenosis is the main cause of ischemic stroke, and it is associated with a high risk of stroke recurrence 2 . Presently, the clinical treatment methods available for cerebral atherosclerotic stenosis include drugs (e.g. non-steroidal antiin ammatory drugs [NSAIDs]), endovascular treatment, and surgical treatment. Endovascular treatment has become a common treatment option for cerebrovascular stenosis owing to its effectiveness and the minimal invasive nature 3 . Endovascular treatment can be classi ed into 3 subtypes: simple balloon dilatation angioplasty (BDA), bulb-expanding stent or self-expanding stent(whose e cacy is comparable to that of BDA), and stent implantation(which has the highest effectiveness among the 3 subtypes) 4 .
However, the occurrence of in-stent restenosis (ISR) after stent placement can render the stent ineffective, which is associated with a high risk of recurrent cerebrovascular accidents 5 . Therefore, the identi cation of simple and effective indicators that can predict the occurrence of ISR is critical to facilitate the timely initiations of preventive treatment in order to reduce the incidence rate of ISR.
The pathophysiological mechanism of ISR has not been fully elucidated until date, and in ammatory response and platelet activation are generally considered to be the chief causes of ISR. The conventional indicators, such as hypersensitive C-reactive protein (h-CRP) 6 ,procalcitonin (PCT) levels 7 , and the mean platelet volume (MPV) 8 , have been proven to be correlated with ISR. However, the conventional indicators have low speci city and present with limitations in clinical prediction of ISR. The neutrophil-tolymphocyte ratio (NLR) is a sensitive indicator of in ammation, and has been con rmed to be associated with ISR after implantations of carotid artery stent 9 , although its relationship with the other types of cerebrovascular ISR remains to be clari ed. Serum albumin (ALB) is a negative acute-phase in ammatory protein. Its concentration decreases during the in ammatory reaction phase. The decrease in the ALB level is related to an increase in the blood viscosity 10 . The platelet-to-lymphocyte ratio (PLR) is a sensitive indicator that re ects platelet activation and in ammation. Past studies have con rmed that ALB and PLR are associated with coronary artery stent restenosis 11,12 , however, their relationship with restenosis in cerebrovascular bare-metal stents remains unclear. The present study aimed to explore the relationships between the preoperative peripheral blood ALB level and PLR with cerebrovascular bare-metal restenosis as well as to compare the predictive e cacies of NLR, ALB, and PLR for intracranial restenosis of cerebrovascular bare-metal stents in order to establish a strong predictor of ISR for clinical reference in the future.

Study Population
We retrospectively evaluated the medical records of 223 patients who underwent elective implantation of cerebrovascular stents (both intracranial + extracranial) at the Department of Neurology between January 2018 and December 2020. The information on patients' sex, age, body mass index (BMI), smoking status, the presence of hypertension/diabetes/coronary heart disease, lipid pro le, stent placement location, stent type, and other characteristics were retrieved from the medical records. The patients were followedup for 6-12 months using computed tomography angiography (CTA) or digital subtraction angiography (DSA) and, based on the results into the ISR and no ISR(NO-ISR) groups. The patient inclusion criteria included the following: the implantation of intracerebral vascular stent for the dilation of original stenosis of > 70% at our hospital and a follow-up study at our hospital within 6-12 months involving examinations such as hematology, CTA, or DSA. The patient exclusion criteria included the following: (1) the lack of relevant baseline data; (2) receipt of irregular anti-platelet therapy after surgery; and the presence of (3) acute and chronic infection; (4) severe liver and kidney diseases; (5) blood diseases or benign and malignant tumors; (6) iodine allergy; (7) a recent history of major trauma and surgery; (8) heart failure; (9) systemic immune diseases and the use of in ammation inhibitors; and (10) pregnancy. The smoking habit referred to daily smoking since > 6 months. BMI was de ned as weight (kg) divided by height squared (m 2 ). The presence of hypertension was de ned according to the 2017ACC/AHA diagnostic criteria 13 or based on prior treatment for hypertension. The presence of diabetes was de ned according to the 2019ESC/EASD guidelines for diabetes management diagnostic criteria 14 or prior diabetes treatment. Presence of dyslipidemia was de ned according to the 2019ESC/EAS guidelines for the management of dyslipidemia 15 or based on prior treatment for dyslipidemia.

Laboratory Assays
The demographic characteristics and the laboratory and imaging results were retrieved from the medical records. The fasting peripheral blood samples were collected before surgery. EDTA tubes were used for hematological testing and non-anticoagulant tubes were used for biochemical analysis. An automated blood analyzer (Sysmex XE-2100 Kobe, Japan) was used for sample analysis, and the standard laboratory techniques were employed for testing. NLR was de ned as the ratio of neutrophil count to the lymphocyte count, and PLR was de ned as the ratio of platelet count to the lymphocyte count.

Procedures and Evaluation
All cerebrovascular stent implantations were performed in compliance with the relevant guidelines and recommendations of the hospital. All patients received the following drug treatment in accordance with the guidelines before surgery, aspirin (100-mg/day) and clopidogrel (75-mg/day) for 3-5 days, statins for lowering the lipid levels, and speci c drugs for the management of hypertension and diabetes. If stenting was planned within the subsequent 72 h, an additional 300 mg of clopidogrel was administered 4 h prior to the surgery. All surgical procedures were performed by quali ed doctors. The operation was performed under local or general anesthesia, and 8 F arterial sheathing was inserted into the femoral artery by the modi ed Seldinger method; in addition, heparin (70-IU/kg) was administered intraoperatively, and the catheter was irrigated with continuous high-pressure infusion of heparin saline to reduce the risk of thromboembolic events. After the arrival of 8 F guiding sheath to the stenosis segment, the stenosis degree was calculated in accordance with the standard North American symptomatic carotid endarterectomy test (NASCET). The selection of appropriate protective device, requirement for preexpansion, and the determination of stent location and size were performed according to the condition of the blood vessel. One of the following bare-metal stents was selected: Wallstent (Boston Scienti c, US), Neuroform EZ (US), Enterprise (Codman, US), Express SD/LD(Boston Scienti c, US), Apollo (Shanghai Minimally Invasive Medical Device Co., Ltd. ), Wingspan (Boston Scienti c). A residual stenosis rate of ≤ 30% indicated that the angiography was successful. If the residual stenosis rate was > 50%, it was usually expanded after angiography, and, if thromboembolism was not detected, the protective device was recovered. After 3-6 months of postoperative administration of aspirin and clopidogrel, patients with clopidogrel resistance were prescribed a single anti-platelet agent for life (usually aspirin). Patients with clopidogrel resistance should be prescribed ticagrelor or other anti-platelet agents, and statins for longterm lipid regulation. The patients were followed up for 6-12 months, and CTA or DSA were rechecked to assess the occurrence of ISR. ISR was de ned as > 50% stenosis within the stent or within 3-mm of the edge of the stent, as con rmed by angiography 5 .

Statistical Analyses
All data analyses were performed using the SPSS version 18.0. Categorical variables are expressed in numbers and percentages, and continuous variables were expressed as mean ± standard deviation or as the median and quartile range. Quantitative data were expressed as (`X ± S), and t-test was performed to compare the mean values between the two groups; the count data was expressed by ratio (n, %), and the χ 2 test was used for comparison. Multivariate logistic regression analysis was performed to determine the independent predictor of ISR. Receiver operator characteristic curves (ROCs) of NLR, ALB, and PLR for the prediction of ISR were used to calculate the cut-off value, area under the curve (AUC), sensitivity, and speci city of each indicator, while AUC was used as the indicator of performance. p < 0.05 was considered to be statistically signi cant.

Results
The baseline demographic, clinical, and procedural characteristics of the patients are depicted in Table 1. The mean age of the patients was 62 ± 9 years, and the subjects included 172 (77.13%) male patients.
Patients were categorized into two groups, namely ISR and non-ISR (NO-ISR) groups. ISR occurred in 39.91% (n = 89) of the patients.
Patients in the ISR group were older than those in the NO-ISR group. The ISR group showed a signi cantly higher prevalence rate of diabetes and signi cantly lower BMI than did the NO-ISR group. No signi cant differences were noted between the two groups with respect to age, sex, smoking status, incidence rates of hypertension and hyperlipidemia, and stent location, length, and diameter. Table 2 lists the laboratory ndings of the study population at the baseline. Patients with ISR had a higher baseline value of NLR (p < 0.001), and PLR (p < 0.001) and lower values of ALB (p < 0.001) and lymphocyte count (p = 0.003) than did those without ISR. As shown in Table 2 and Fig. 1, patients with ISR showed signi cantly lower ALB levels than did patients without ISR (39.43 ± 2.96 vs 42.13 ± 2.77 g/L, p < 0.001). In contrast, ISR patients showed higher NLR and PLR values than those without ISR (3.19 ± 1.49 vs 2.46 ± 1.12; p < 0.001, 148.60 ± 33.73 vs 125.36 ± 34.88, p < 0.001).
The ROC curve analysis was employed to calculate the truncation values AUCs, sensitivities, speci cities, and 95% con dence intervals (CIs) of NLR, ALB, and PLR (Table 3), and the ROC curves of NLR, ALB, and PLR were are depicted in Fig. 2 (Table 4).

Discussion
In this study, we analyzed the predictive value of preoperative peripheral blood NLR and PLR counts and the ALB level for ISR in patients who underwent cerebrovascular bare-metal stent implantation. We noted that higher preoperative NLR (> 2.28), PLR (> 140. 25), and lower ALB level (≤ 42.32) were associated with the development of ISR, and hence these parameters could help predict the occurrence of ISR. The incidence rate of ISR after cerebrovascular stenting was recorded to be 20-30% 16 , and the ISR rate was found to vary according to the clinical centers and locations. The occurrence of ISR increases the risk of cerebrovascular accidents, increase costs, and reduces the quality of life 17 . Therefore, the etiology of stent restenosis has become a research hotspot. At present, the pathophysiological mechanism of ISR is considered to mainly involve: immune in ammatory response, vascular remodeling, new atherosclerosis, and thrombosis, among others. [18][19][20] The relationship between NLR, as a sensitive indicator of in ammation, and ISR was con rmed in previous studies. NLR has been suggested as a predictor of cerebrovascular ISR 9 and coronary ISR 21 . Mechanical damage and blood ow changes during stent placement leads to the deposition of platelets and brin 22 . The interaction between white blood cells and platelets leads to the recruitment of white blood cells, including neutrophils, into the scaffold segment, which in turn triggers an in ammatory response, in which the involved in ammatory factors lead to the formation of neointima. The neointima is dysfunctional and does not inhibit platelet aggregation. Some subtypes of lymphocytes, such as regulatory T-cells (Tregs), suppress autoimmune in ammation and maintain homeostasis 23 . Therefore, increased NLR counts indicate an active in ammatory response and promotes the occurrence of ISR.
PLR is a new indicator that re ects platelet activation and in ammation status, and it is widely used to predict adverse outcomes of cardiovascular and cerebrovascular disease, such as stroke 24 , myocardial infarction 25 , and heart failure 26 . S Y et al. con rmed that high PLR is associated with the incidence of coronary ISR 12 . The possible mechanism may be as follows: in ammatory mediators stimulate megakaryocytes to proliferate and produce relatively greater higher number of platelets, which play an important role in the pathogenesis of atherosclerosis 27 . Platelets participate in both the formation of plaques and the formation of brin plugs as a result of the rupture of unstable plaques. They can also release thrombin and other substances, thus exacerbating the in ammatory response 28 . Thus, increased platelet count increases the risk of thrombosis, while lymphocytes achieve an in ammation-antiin ammatory balance by suppressing their own in ammatory response 23 . Therefore, in ammation and high platelet activation increases with an increase in the PLR count.
The two new types of in ammatory indicators PLR and NLR are closely related to the progression of atherosclerosis 29 . Shah et al 30 found that low lymphocyte count is an independent risk factor for coronary heart disease, which will not be affected by the absolute counts of blood cells under different physiological conditions, such as dehydration or exercise. Thus, they may have good predictive value for ISR.
ALB exerts physiological effects such as the regulation of in ammation, anti-oxidation, anti-coagulation, and inhibition of platelet aggregation, which has been con rmed to be related to stroke, cardiovascular disease, and coronary ISR 11,31 . It can be reversibly combined with nitric oxide (NO) and prostaglandin through oxidation and glycosylation in order to achieve anti-oxidation and regulation of in ammation 32 . ALB also possesses heparin-like activity, probably due to the similarity in their structures and electrostatic charges 33 . M P et al. compared the low ALB group, physiological albumin group, and high ALB group using platelet function analyzer and found that the low ALB group was more likely to demonstrate thrombosis 34 . Therefore, hypoalbuminemia can aggravate in ammation, promote platelet aggregation, and aggravate the occurrence of ISR.
Based on our analysis, the most important method that helps reduce the incidence of ISR is monitoring and prevention, which makes standardization of long-term medication and regular review essential. Whole blood cell test is a non-invasive, economical, and simple test method. The identi cation of blood markers can help predict ISR to facilitate effective patient follow-up study. Once ISR is detected, active intervention is needed to reduce the occurrence of cerebrovascular events. Balloon dilatation or repeated stenting should be considered when anti-platelet drugs or statins are ineffective 35 . There continues to be a risk of ISR after repeated stent placement. Because of the high mutation rate of anti-platelet drug resistance genes in the Chinese population, conducting genetic screening for drug resistance, if the conditions permits, is recommended for guide drug selection 36 .

Limitation
This study has some limitations. First of all, this is a single-center retrospective study that can explain the relationship between clinical factors and ISR. Second, although we included data on laboratory examination and imaging, we could not use intravascular ultrasound or optical coherence tomography or stent restenosis plaque biopsy for further examination. These methods are more quantitative and informative than visual estimation. Finally, we did not conduct a long-term follow-up. Thus, large-scale, prospective, and randomized clinical trials are warranted in the future to assess the clinical applicability of our ndings.

Conclusion
In conclusion, this study demonstrates that diabetes, high BMI, NLR and PLR counts, and low ALB level and lymphocyte count are associated with the high risk of ISR after implantation of cerebrovascular baremetal stents. Thus, the count of NLR and PLR and the value of BMI increase before surgery, while ALB value decline. These factors can act as independent risk factor for ISR in patients implanted with cerebrovascular bare-metal stents. In summary, NLR > 2.28, ALB level ≤ 42.32, and PLR > 140.25 are powerful indicators that independently predict ISR in patients implanted with cerebrovascular bare-metal stents.