Postoperative blood glucose increase can predict all-cause mortality within 3 months after successful interventional recanalization in patients with acute large vessel occlusion cerebral infarction

Abstract

Aims

To evaluate whether postoperative blood glucose increase (BGI) can serve as a predictive indicator for all-cause mortality within 3 months in patients with acute cerebral infarction with large vessel occlusion.

Methods

BGI was defined as fasting blood glucose levels higher than preoperative random blood glucose levels on the first morning after surgery. A retrospective analysis was conducted on the clinical data of patients with acute cerebral infarction due to anterior circulation large vessel occlusion in two centers from January 2019 to November 2023, who successfully underwent interventional recanalization. A multivariable Cox regression was performed to determine independent factors of all-cause mortality after successfully interventional recanalization.

Results

A total of 321 patients were included, of which 73 patients (22.74%) had BGI. The all-cause mortality rate within 3 months after surgery in BGI patients was higher than that in non BGI patients (24.66% vs. 3.23%, P < 0.001). The Kaplan-Meier analysis showed that patients with BGI had a higher postoperative mortality rate at 3 months (P < 0.001). After adjusting for potential covariates, multivariable analysis showed a significant correlation between BGI and 3-month postoperative mortality rate (adjusted HR: 5.694, 95%CI: 2.379–13.626, P<0.001). The area under the ROC curve predicted by BGI, preoperative ASPECT score, and a combination model including BGI and ASPECT score for all-cause mortality within 3 months after surgery were 0.753, 0.762 and 0.853 respectively.

Conclusion

BGI may be a relatively good indicator for predicting all-cause mortality within 3 months after successful interventional recanalization in patients with acute cerebral infarction due to large vessel occlusion, and the predictive efficacy of the combination model including BGI and ASPECT score is higher.

Introduction

Acute cerebral infarction with large vessel occlusion has a high incidence rate, mortality and disability rate [1]. Interventional recanalization is one of the most important treatment methods for acute cerebral infarction with large vessel occlusion [2,3,4,5]. Although the successful rate of vascular recanalization can reach up to 90%, 10 -30% of patients still died within 3 months after surgery [6,7,8].

Few studies have focused on the factors related to all-cause mortality within 3 months after successful interventional recanalization surgery in patients with acute large vessel occlusion cerebral infarction, and the construction of corresponding predictive models. The mechanism of blood glucose increase in patients with acute cerebral infarction may be the release of norepinephrine and cortisol induced by stress response [9]. Studies had found that postoperative blood glucose increase in patients with acute cerebral infarction is associated with futile recanalization [10], and hospitalization blood glucose elevation is associated with adverse functional outcomes after interventional recanalization [11,12,13]. However, the relationship between postoperative blood glucose increase and all-cause mortality within 3 months after surgery is not yet clear. Recently, perioperative glucose dynamics have been shown to be a valuable biomarker for poor prognosis in patients with acute large vessel occlusion cerebral infarction undergoing interventional recanalization [14]. Patients with poor prognosis theoretically experienced longer periods of cerebral ischemia and stress response, which may lead to postoperative fasting blood glucose levels higher than preoperative random blood glucose [15].

Considering that many studies had suggested that stress hyperglycemia was associated with futile recanalization after interventional recanalization of acute cerebral infarction with large vessel occlusion, and compared with other indicators that may be independently related to all-cause mortality, postoperative fasting blood glucose and preoperative random blood glucose were relatively easily obtained indicators, and the results were relatively objective and reliable, we tried to explore whether BGI can be used as the main predictor of all-cause mortality. The aim of this study is to evaluate whether postoperative BGI can serve as a predictor of all-cause mortality within 3 months after successful interventional recanalization in patients with acute cerebral infarction due to anterior circulation large vessel occlusion. The innovation of this study was to explore whether the new index of whether postoperative fasting blood glucose is higher than preoperative random blood glucose can predict all-cause mortality within 3 months after successful interventional recanalization. This study was of great help to further understand the correlation between postoperative blood glucose and all-cause mortality.

Materials and methods

Research object

We retrospectively analyzed the clinical data of 321 patients with acute cerebral infarction due to anterior circulation large vessel occlusion who were admitted to Shenzhen Hospital of Southern Medical University and Heyuan People’s Hospital from January 2019 to November 2023. The Ethics Committee of Shenzhen Hospital of Southern Medical University and Heyuan People’s Hospital approved the acquisition of retrospective study patient data from the hospital’s clinical database for this study, and waived written informed consent. Inclusion criteria: (1) Baseline National Institutes of Health Stroke Scale (NIHSS) score at admission ≥ 6 points; (2) At admission, the Alberta Stroke Early CT (ASPECT) score was ≥ 6 points; (3) The time from onset to femoral artery puncture was ≤ 24 h; (4) Cerebral angiography confirmed occlusion of large blood vessel in the anterior circulation; (5) The patient or family members signed and agreed to undergo interventional recanalization treatment; (6) Interventional recanalization was successful, and the eTICI grading of forward blood flow after recanalization was 2b-3. Exclusion criteria: (1) Previous modified Rankin scale (mRS) score > 2 points; (2) Prior to the onset of cerebral infarction, there was a pulmonary infection or severe heart, lung and kidney dysfunction; (3) On the first morning after surgery, fasting blood glucose was not tested. The flowchart of patient inclusion was shown in Fig. 1.

Fig. 1

The flowchart of patient inclusion. mRS: modified Rankin Scale; BGI: Blood glucose increase; eTICI: expanded Thrombolysis in Cerebral Infarction

Treatment methods

After discovering occlusion of large blood vessel in the anterior circulation during cerebral angiography, one or more treatment methods such as stent thrombectomy, thrombus aspiration, balloon dilation, and stent implantation were used to intervene and recanalize the occluded vessels in patients. We managed patients according to existing guidelines after surgery.

Patient data collection

In terms of baseline data, we collected patient’s age, gender, medical history, NIHSS score at onset, mRS score before onset, ASPECT score at onset, treatment of intravenous thrombolysis, location of occluded vessel, collateral circulation score (ASITN/SIR), extended thrombosis in cerebral infarction (eTICI) grading, time from onset to occluded vessel recanalization, time from puncture to occluded vessel recanalization, TOAST (Trial of Org10172 in Acute Stroke Treat-ment) classification, preoperative random blood glucose, fasting blood glucose on the first morning after surgery and other information. In terms of complications, we collected information on whether patients have symptomatic cerebral hemorrhage, asymptomatic cerebral hemorrhage, subarachnoid hemorrhage, cerebral hernia, stroke associated pneumonia, deep vein thrombosis and gastrointestinal bleeding. In terms of prognosis, we collected the mRS scores of patients 3 months after onset through telephone follow-up or outpatient follow-up.

Definition and classification of variables in data

Postoperative BGI was defined as a blood glucose value greater than preoperative random blood glucose on the first morning after surgery. The fasting blood glucose value in the first morning after surgery was obtained by venous blood test, and fasting for at least 8 h was required before venous blood sampling.The site of vascular occlusion can be divided into 5 categories: M1 segment of the middle cerebral artery, M2 segment of the middle cerebral artery, internal carotid artery, anterior cerebral artery and anterior circulation tandem lesions. The ASITN/SIR scoring system using Digital subtraction angiography (DSA) imaging was used to evaluate collateral circulation status, with a score of 0–2 indicating poor collateral circulation and 3–4 indicating good collateral circulation. The degree of vascular recanalization was evaluated using the extended thrombolysis in cerebral infarction (eTICI). The eTICI grade of 2b-3 was defined as successful recanalization, while the eTICI grade of 0-2a was defined as failed recanalization. The TOAST classification can be divided into three categories: large artery atherosclerosis, cardiogenic embolism and other causes.

Prognosis assessment

The main endpoint of this study was whether the patient experienced all-cause mortality within 3 months after surgery, with an mRS score of 6. The secondary endpoint was the mRS score at 3 months after surgery, with a score of 0–2 indicating effective recanalization and 3–6 indicating poor prognosis.

Statistical processing

This study used SPSS 26.0 statistical software to conduct statistical analysis on the data. When we chose the statistical analysis methods, we referred to some previous relevant studies [10, 16]. If the quantitative data conformed to a normal distribution, it was expressed as mean ± standard deviation (x ± s), and independent sample t-test was used for inter group comparison; data that did not follow a normal distribution were described with median and interquartile intervals, and Mann-Whitney U test was used for inter group comparison. Count data was expressed in percentage, and intergroup comparisons were conducted using chi square or Fisher’s exact test. A p-value less than 0.05 was considered statistically significant. Because all-cause mortality within 3 months after successful interventional recanalization involved survival time, so we performed univariate and multivariable Cox regression analysis on the indicators related to all-cause mortality within 3 months after surgery. We conducted a proportional hazards hypothesis test on indicators with statistical differences in univariate Cox regression analysis. The method used for continuous variables was Schoenfeld residual method, which involved drawing Schoenfeld residual plots of continuous variables and rank of time, and conducting Pearson correlation analysis on continuous variables. The method used for categorical variables was to plot survival curves and quadratic logarithmic survival curves. Indicators that met the proportional hazards assumption were included in the multivariate Cox regression analysis. We draw Kaplan-Meier curve, forest plot and Receiver Operating Characteristic (ROC) curve of all-cause mortality within 3 months after surgery.

Results

Clinical baseline data of patients with and without BGI

This study included a total of 321 patients with an average age of 66.41 ± 11.25 years, including 215 males (66.98%). The median NIHSS score at the onset of the disease was 12 (10–16) points, and the average preoperative mRS score was 0.12 ± 0.49. A total of 73 patients (22.74%) developed BGI. According to the presence or absence of BGI, patients were divided into two groups: those with BGI and those without BGI. There were statistical differences between the two groups in terms of drinking history, NIHSS score at the time of onset, ASPECT score of preoperative CT, good collateral circulation, TOAST classification, preoperative random blood glucose, and fasting blood glucose in the first morning after surgery. The proportion of patients in BGI group with drinking history was higher, NIHSS score at the time of onset was higher, preoperative ASPECT score was lower, the proportion of good collateral circulation was lower, the proportion of large atherosclerosis type of pathogenesis was lower, the proportion of cardiogenic embolism pathogenesis was higher, preoperative random blood glucose was lower, and the fasting blood glucose in the first morning after surgery was higher. There was no statistical difference between the two groups in terms of age, gender, hypertension history, diabetes history, cerebral infarction history, coronary heart disease history, atrial fibrillation history, smoking history, pre onset average mRS score, receiving intravenous thrombolysis treatment, location of occluded vessel, time from onset to recanalization of occluded vessel, and time from puncture to recanalization of occluded vessel. The research flowchart of this study was shown in Fig. 1. The clinical baseline data of patients with and without BGI were shown in Table 1.

Table 1 Clinical baseline data of patients with and without postoperative BGI after successful interventional recanalization in acute cerebral infarction with anterior circulation occlusion of large blood vessel

Comparison of complications between patients with and without BGI

There was a statistically significant difference (p < 0.05) in the incidence of symptomatic cerebral hemorrhage, cerebral hernia, and stroke related pneumonia between the two groups of patients. Patients with BGI had a higher incidence of symptomatic cerebral hemorrhage, cerebral hernia and stroke related pneumonia than those without BGI. The complications of two groups of patients were shown in Table 2.

Table 2 Comparison of complications and prognosis in patients with and without postoperative BGI after interventional recanalization in acute cerebral infarction with large blood vessel occlusion of anterior circulation

Comparison of prognosis between patients with and without BGI

There were statistically significant differences (p < 0.05) in the all-cause mortality, mRS score, and incidence of futile recanalization within 3 months after interventional recanalization between the two groups of patients. Patients with BGI had higher all-cause mortality within 3 months after surgery, higher mRS score and higher incidence of futile recanalization. The prognosis condition of two groups of patients was shown in Table 2. The distribution of mRS scores in two groups of patients after 3 months of onset was shown in Fig. 2.

Fig. 2

Distribution of mRS scores at 3 months after interventional recanalization in patients with and without BGI. mRS: modified Rankin Scale; BGI: Blood glucose increase

Clinical baseline data of patients who died or did not die within 3 months after surgery

According to whether the patient died within 3 months after interventional recanalization surgery, the patients were divided into two groups: all-cause mortality group and no mortality group. There were statistical differences between the two groups in smoking history, NIHSS score at onset, ASPECT score before surgery, proportion of good collateral circulation, time from onset to recanalization of occluded vessels, TOAST classification and incidence of BGI (p < 0.05). Patients in the all-cause mortality group had a higher proportion of smoking history, higher NIHSS score at onset, lower ASPECT score before surgery, lower proportion of good collateral circulation, shorter time from onset to recanalization of occluded vessel, lower proportion of large artery atherosclerosis type of pathogenesis, higher proportion of cardiogenic embolism pathogenesis and higher incidence of BGI. There was no significant difference between the two groups in age, sex, hypertension history, diabetes history, cerebral infarction history, coronary heart disease history, atrial fibrillation history, drinking history, pre onset mRS score, receiving intravenous thrombolysis, occluded vessel position, eTICI recanalization blood flow grade and time from puncture to recanalization of occluded vessel (p > 0.05). The baseline data of two groups of patients who died and did not die were shown in Table 3.

Table 3 Univariate analysis of all-cause mortality and no mortality in patients with acute cerebral infarction of anterior circulation large artery occlusion within 3 months after interventional recanalization surgery

Kaplan-Meier curve of all-cause mortality within 3 months after successful interventional recanalization in patients with and without BGI

The Kaplan-Meier curve of all-cause mortality within 3 months after successful interventional recanalization in patients with and without BGI with large vessel occlusion in acute cerebral infarction are shown in Fig. 3. When only BGI was included as a factor, patients with BGI had an 8.709 times higher risk of all-cause mortality within 3 months compared to patients without BGI (95% Confidence interval (CI): 3.352–22.43).

Fig. 3

The Kaplan-Meier curve of all-cause mortality within 3 months after successful interventional recanalization in patients with and without BGI

BGI: Blood glucose increase; HR: hazard ratio

Univariate and Multivariable Cox regression analysis and forest plot of factors related to all-cause mortality within 3 months after surgery

The indicators with statistical differences with all-cause mortality after 3 months in chi square test, t-test and non parametric test were included in Cox regression analysis. Indicators with statistical differences in chi square test, t-test and non parametric test were included in Univariate Cox regression analysis. Although the residual plots of NIHSS score at onset, preoperative ASPECT score, and rank of time could not clearly evaluate whether these two variables met the proportional hazards hypothesis (as shown in Supplementary Fig. 1), the P-values of these two variables in Pearson correlation analysis were both greater than 0.05 (as shown in Supplementary Table 1), indicating that NIHSS score at onset and preoperative ASPECT score met the proportional hazards hypothesis. Survival curve and quadratic logarithmic survival curve of the categorical variables including good collateral circulation (Yes vs. No), pathogenesis (Cardiogenic embolism vs. Large artery atherosclerotic, Other reasons vs. Large artery atherosclerotic) and BGI (With vs. Without) did not show any evident curve crossover (as shown in Supplementary Fig. 2, Supplementary Fig. 3, and Supplementary Fig. 4), indicating that good collateral circulation, pathogenesis, and BGI all conformed to the proportional hazards hypothesis. We included the indicators above that met the proportional hazards assumption in the multivariate Cox regression analysis. In Multivariable Cox regression analysis, it was found that within 3 months after interventional recanalization there was a strong correlation between BGI (Hazard ration (HR): 5.693, 95%CI: 2.379–13.626, p < 0.001), preoperative ASPECT score (HR:0.467, 95%CI: 0.252–0.864, p < 0.015) and all-cause mortality. There was a statistically significant difference in the comparison of the pathogenesis of cardiogenic embolism and large artery atherosclerosis (HR: 3.046, 95%CI: 1.090–8.511, p = 0.043) and the comparison of other reasons and large artery atherosclerosis (HR: 3.469, 95%CI:1.042–11.543, p = 0.043). Univariate and multivariable Cox regression analysis are shown in Table 4. The forest plot of multivariable Cox regression analysis of all-cause mortality 3 months after successful interventional recanalization is shown in Fig. 4. Our multivariable logistic regression analysis and forest plot showed that when the postoperative blood glucose increased, the relative risk of all-cause mortality within 3 months after successful interventional recanalization was 5.694 times that of patients without postoperative blood glucose increase.

Fig. 4

The forest plot of multivariable Cox regression analysis of all-cause mortality 3 months after successful interventional recanalization. NIHSS: National Institute of Health stroke scale; ASPECT: Alberta Stroke Program Early CT Score; TOAST: Trial of Org10172 in Acute Stroke Treat-ment; BGI: Blood glucose increase; HR: hazard ratio; CI: confidence interval

Table 4 Univariate and multivariate Cox regression analysis of all-cause mortality within 3 months after interventional recanalization surgery for acute cerebral infarction with anterior circulation large vessel occlusion

ROC curve of predicting all-cause mortality within 3 months after surgery

We plotted ROC curves for predicting all-cause mortality within 3 months after interventional recanalization using BGI and preoperative ASPECT scores, with areas under the curves of 0.753 and 0.762 respectively. We then plotted the ROC curve using the combination model including BGI and ASPECT score for predicting all-cause mortality within 3 months after interventional recanalization and the area under the curve was 0.853. The combination model of BGI and preoperative ASPECT score had a higher predictive power for all-cause mortality within 3 months after surgery. The ROC curve was shown in Fig. 5A.

Fig. 5

A ROC curve for predicting mortality within 3 months after interventional recanalization using BGI, preoperative ASPECT score, combination of BGI and preoperative ASPECT score for all patients. B ROC curve for predicting mortality within 3 months after interventional recanalization for patients with diabetes history using BGI, preoperative ASPECT score, combination model including BGI and ASPECT score. C ROC curve for predicting mortality within 3 months after interventional recanalization for patients without diabetes history using BGI, preoperative ASPECT score, ccombination model including BGI and ASPECT score. ROC: Receiver Operating Characteristic; BGI: Blood glucose increase; ASPECT: Alberta Stroke Program Early CT Score

ROC curve of BGI predicting all-cause mortality within 3 months after interventional recanalization for patients with diabetes and those without diabetes

According to the past history of diabetes, patients were divided into two groups: with diabetes history group and without diabetes history group. BGI, preoperative ASPECT score and the combination model including BGI and ASPECT score were used respectively to drawn the ROC curve of all-cause mortality within 3 months after interventional recanalization. It was found that the area under the curve of BGI was 0.785 and 0.747, the area under the curve of ASPECT score was 0.672 and 0.781, the area under the curve of combination model including BGI and ASPECT score was 0.832 and 0.856 of the two groups of patients respectively. The ROC curves of two groups of patients were shown in Fig. 5B and C. No matter whether the patient had a history of diabetes or not, BGI can predict all-cause mortality within 3 months after interventional recanalization well, and combination of BGI and preoperative ASPECT score was more effective in predicting all-cause mortality within 3 months after interventional recanalization.

ROC curve for predicting all-cause mortality within 3 months after surgery using BGI in patients with different eTICI reperfusion blood flow grades

According to the different grades of eTICI reperfusion blood flow, patients were divided into three groups: 2b, 2c and 3. ROC curves were plotted using BGI to predict all-cause mortality within 3 months after intervention recanalization, with areas under the curves of 0.808, 0.691, and 0.728 respectively, as shown in Fig. 6A and B, and 6C. BGI can predict all-cause mortality within 3 months after interventional recanalization in different eTICI reperfusion blood flow grades.

Fig. 6

ROC curves of BGI predicting all-cause mortality within 3 months in patients with eTICI grade 2b, 2c and 3 after intervention recanalization were shown in A, B and C respectively. ROC: Receiver Operating Characteristic; BGI: Blood glucose increase

Discussion

The existing clinical research data showed that the mortality rate within 3 months after interventional recanalization in patients with acute cerebral infarction due to anterior circulation occlusion was 10 -30% [6,7,8]. At present, studies on the prognosis of patients after interventional recanalization mainly focuses on futile recanalization, and few studies focus on all-cause mortality within 3 months after interventional recanalization. As the treatment window extended to 24 h after onset [17,18,19], more patients with acute cerebral infarction with anterior circulation occlusion will receive interventional recanalization treatment. It is very necessary to establish a predictive model that can early predict all-cause mortality within 3 months after interventional recanalization, so as to provide targeted intervention for patients with high risk of mortality in the early stage and reduce all-cause mortality within 3 months after surgery.

A retrospective study by Cabral A et al. [20] found that higher blood glucose variability after interventional recanalization was an independent risk factor for mortality 3 months after surgery. A retrospective study by Zhang J et al. [21] found that stress hyperglycemia was associated with lower rate of good prognosis, higher mortality, and higher bleeding risk in patients after mechanical thrombectomy, regardless of whether the patient had a history of diabetes. Peng Z et al. [22] measured the ratio of stress hyperglycemia through the ratio of blood glucose to glycated hemoglobin, and found that stress hyperglycemia would reduce the chance of good prognosis of patients who received interventional recanalization. However, the mechanism of all-cause mortality within 3 months after interventional recanalization caused by increased fasting blood glucose after operation is not completely clear. Tao Tang et al. found that postoperative blood glucose increase can serve as a predictive factor for futile recanalization in patients received successful interventional recanalization after acute cerebral infarction of anterior circulation occlusion. The study by Bing Yang et al. [16] found that a higher ratio of glucose to glycated hemoglobin can predict postoperative mortality in patients with acute cerebral infarction with large vessel occlusion undergoing interventional recanalization. Our study found that the blood glucose increase in the first morning after interventional recanalization compared with the random blood glucose before surgery was a powerful predictor of all-cause mortality within 3 months after surgery. The predictive efficacy was equivalent in patients with and without diabetes history. Our findings were consistent with previous studies. The preoperative random blood glucose and fasting blood glucose on the first morning after surgery that we had selected can serve as important supplements to previous studies indicators of mortality after interventional recanalization.

A study had found that there was a higher correlation between BGI and futile recanalization in patients with mTICI grade 3 forward blood flow during interventional recanalization, while in patients with mTICI grade 2b forward blood flow, BGI was not strongly correlated with futile recanalization, suggesting that blood glucose increase after incomplete recanalization may be a compensatory protective measure. Lowering blood glucose in patients with incomplete recanalization may not necessarily be beneficial for patients [23]. Our study found that the area under the curve of the ROC of BGI for predicting all-cause mortality within 3 months after surgery in eTICI 2b, 2c and 3 grade reperfusion blood flow patients was 0.808, 0.690, and 0.727 respectively. Our study suggested that BGI may be a good predictor of all-cause mortality within 3 months after surgery in patients with incomplete recanalization. Blood glucose increase after incomplete recanalization may not be a compensatory protective measure, which differed from previous studies.

In our study, compared to patients without BGI, patients with BGI had higher NIHSS scores at onset, higher preoperative ASPECT scores, and poorer collateral circulation conditions. This suggested that patients with more severe neurological deficits, larger infarct areas, and poorer collateral compensation were more likely to develop BGI. The pathogenesis of patients with BGI was that the proportion of large artery atherosclerotic type in TOAST classification was lower, and the proportion of cardiogenic embolism type was higher, suggesting that patients with cardiogenic embolism acute cerebral infarction are more likely to have a relatively longer stress state compared to that of large artery atherosclerotic type. At the same time, it was found that patients with acute cerebral infarction due to cardiogenic embolism had a higher risk of all-cause death within 3 months after successful interventional recanalization than those with large artery atherosclerosis in the multivariable Cox regression analysis.

Our study also found that patients with BGI had a higher incidence of symptomatic cerebral hemorrhage, herniation and stroke associated pneumonia. Patients with BGI would experience sustained elevated blood glucose levels within 24 h of onset. Elevated blood glucose levels may have harmful effects on successful interventional recanalization by exacerbating inflammatory reactions, damaging cerebrovascular reactivity, exacerbating intracellular acidosis and increasing reperfusion injury, such as bleeding transformation, increased infarct size and worsening pulmonary infections [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24]. From the perspective of complications, we speculated that the mechanism of all-cause mortality caused by postoperative hyperglycemia may be related to the inflammatory reaction after cerebral infarction, damaging cerebral vascular activity, and finally leading to hemorrhagic transformation, brain hernia and other complications. For patients with successful interventional recanalization with BGI, we should pay more attention to the degree of inflammation and reperfusion injury in the early stage, take effective measures to reduce the incidence of pulmonary infection, alleviate the severity of pulmonary infection, alleviate the severity of cerebral edema and avoid the occurrence of hyperperfusion syndrome, so as to reduce the risk of all-cause mortality within 3 months after interventional recanalization in patients. Asaithambi G et al.‘s study [25] found that intensive glycemic control was not associated with lower mortality or higher long-term favorable outcome rates in patients with a history of diabetes after interventional recanalization compared with standard treatment. Chang JY et al.‘s study [26] found that whether patients were bridged with intravenous thrombolysis, occlusion site, recanalization degree and treatment time, controlling glycated hemoglobin less than or equal to 7.0% before cerebral infarction may be beneficial to the recovery of neurological function in patients treated with interventional recanalization. It can be said that whether the strict control of blood glucose after acute cerebral infarction can improve the prognosis of patients and reduce the mortality of patients has not been confirmed. At present, there is little research exploring the prognostic role of blood glucose control after interventional recanalization surgery for acute cerebral infarction with large vessel occlusion. Our study provided some ideas for the strict control of blood glucose in patients after interventional recanalization surgery.

Considering that hypertension, smoking, pathogenesis, postoperative fasting blood glucose and other indicators were the basic situation of patients, not complications or outcomes after cerebral infarction, these indicators belonged to the baseline situation. Symptomatic intracerebral hemorrhage, cerebral hernia, stroke associated pneumonia and other complications after cerebral infarction were not the basic situation of patients before and after the onset of cerebral infarction. Our study was not to explore the predictive effect of complications on all-cause mortality of patients, but to explore the predictive effect of patients’ baseline conditions, especially the change of fasting blood glucose after surgery, on all-cause mortality of patients, so we did not include complications in the univariate and multivariable regression analysis of all-cause mortality.

In our study, only 8.1% of patients suffered from all-cause mortality within 3 months after successful interventional recanalization, which was lower than previous studies of 10 -30%. A very important reason why the all-cause mortality of patients in our study was lower than that of previous studies may be that we excluded patients with failed interventional recanalization and only included patients with successful interventional recanalization.

This study had some limitations. Firstly, this study was a two center, retrospective study, making it difficult to avoid selection bias and information bias. Secondly, the impact of missing data on the study’s outcomes can not be eliminated due to some data missing. Thirdly, the sample size included in this study was relatively limited, and patients who received failed interventional recanalization were not included. Fourthly, this study only included two indicators, preoperative random blood glucose and fasting blood glucose on the first morning after surgery, to evaluate the dynamic changes in perioperative blood glucose. Although these two indicators were easy to obtain and had high predictive power for all-cause mortality within 3 months after interventional recanalization, more indicators such as postoperative fasting blood glucose and glycated hemoglobin were not included, which may not fully reflect the dynamic blood glucose changes during the perioperative period. And a single-point measurement of postoperative blood glucose may cause misclassification bias. Fifthly, the exclusion of patients with failed recanalization may introduce selection bias, potentially inflating the success rates of the predictive model. In the future, prospective clinical studies with larger sample sizes are needed to further validate the efficacy of BGI in predicting all-cause mortality within 3 months after successful interventional recanalization in acute cerebral infarction with large vessel occlusion.

Conclusion

BGI may be a relatively good indicator for predicting all-cause mortality within 3 months after successful interventional recanalization in patients with acute cerebral infarction due to large vessel occlusion, and the predictive efficacy of the combination model including BGI and ASPECT score is higher.