FormalPara Key Summary Points

Why carry out this study?

No evidence has established a direct causal relationship between early microcirculation disturbance after aneurysmal subarachnoid hemorrhage and neurological function prognosis.

We sought to evaluate the relationship between regional cerebral oxygen saturation measured by near-infrared spectroscopy and other indicators in the early phase after subarachnoid hemorrhage (SAH) and neurological changes during hospitalization.

What was learned from the study?

Our data suggest that regional cerebral oxygen saturation (rSO2) levels are generally lower in patients with neurological deterioration than in those without neurological deterioration and that near-infrared spectroscopy can be used to monitor at least one time point in the population with neurological deterioration.

However, further studies are needed to determine the role of microcirculation and other predictive factors in the neurocritical management of early brain injury (EBI) after aneurysmal subarachnoid hemorrhage, as these factors can reduce the incidence of adverse outcomes and mortality during hospitalization.

Introduction

Subarachnoid hemorrhage (SAH) is one of the most common and severe injuries in the central nervous system and is mostly caused by ruptured intracranial aneurysms. In recent years, with the extensive application of surgical techniques and intensive neurological care, most ruptured aneurysms can be effectively treated by craniotomy clipping or endovascular embolization. However, according to a new survey in the USA, the in-hospital mortality rate of patients with SAH in the last 15 years has remained at approximately 13–14% [1], while the rate of in-hospital death/unordered discharge is as high as 20.6% [2]. Delayed cerebral ischemia (DCI) caused by intracranial artery spasm after SAH has long been considered a key pathophysiological mechanism leading to poor patient prognosis [3]. With the landmark clinical trials CONSCIOUS-2 (registration number NCT00558311) and CONSCIOUS-3 (registration number NCT00940095), we gradually realized the importance of early brain injury (EBI) after SAH, but there is still a lack of in-depth understanding of the pathophysiological changes in EBI and effective clinical monitoring and intervention [3,4,5].

Since the EBI concept of SAH was proposed, there have been many studies on its pathophysiological mechanism. Among these abnormalities, sudden aneurysm rupture leads to blood entering the subarachnoid space, induces autonomic cerebral blood flow regulation disorders and intracranial microcirculation disorders, and causes brain tissue ischemia and hypoxia, which are the most important pathophysiological changes in the early stage of SAH [6]. The vast majority of patients with SAH lack brain tissue ischemic focus on imaging; even among patients with high-grade SAH, secondary cerebral infarction accounts for only 21% of the brain tissue [7]. However, cerebrospinal fluid (CSF) metabolomics analysis suggests that, with or without intracranial vasospasm, the tissue hypoxia marker 2-hydroxyglutarate is significantly associated with long-term neurological function prognosis in patients with SAH [8]. Helbok et al. adopted brain tissue microdialysis and other methods and reported that more than 60% of patients with SAH had brain tissue hypoxia within 24 h after the onset of the disease (PbtO2 < 20 mmHg) [9]. In 2003, Uhl et al. observed segmental and bead-like intracranial microvascular spasm for the first time during the process of craniotomy clipping of ruptured aneurysms, causing 75% cortical microcirculation dysfunction [10]. Our previous studies also confirmed in a mouse model that oxygenated hemoglobin can induce the contraction of pericapillary cells and can cause beaded microvascular spasm and microcirculation disturbance, similar to what occurs in the human body [11, 12]. It has been preliminarily confirmed that intracranial microcirculation disturbance in the very early and subsequent periods after SAH is the direct cause of delayed ischemia and poor prognosis of neurological function [13], and the results of the CONSCIOUS-2 clinical trial have been further interpreted, suggesting that intracranial microvascular spasm is at least partially independent of endothelin receptors. The mechanism of delayed ischemia is different from that of symptomatic intracranial great vascular spasm [14]. However, rather than merely considering the predictive correlation mentioned in previous studies, no evidence has established a direct causal relationship between early microcirculation disturbance after SAH and neurological function prognosis, which has become the key pathophysiological mechanism of EBI in patients with SAH.

In the present study, we sought to (1) evaluate the relationship between regional cerebral oxygen saturation (rSO2) and other indicators in the early phase after SAH and neurological changes during hospitalization and (2) identify the early predictors associated with neurological deterioration during the hospitalization period in patients with SAH. The results will help medical care personnel better understand this critical period, improve their ability to perceive EBI indicators, identify those patients at risk early, and provide more positive strategies.

Methods

Study Design

We retrospectively enrolled aneurysmal patients with SAH admitted to the Neurosurgical Intensive Care Unit of Southwest Hospital between January 2020 and December 2022. All patient data were extracted from medical records, and all patients had angiographically documented aneurysms with SAH, confirmed by either computed tomography (CT) or lumbar puncture. The study protocol was reviewed and approved by the Ethics Committee of Southwest Hospital of Army Medical University (No. (B) KY2023040). Written informed consent to participate was waived by the ethical committee because of the observational nature of the study. This study was conducted in accordance with the ethical guidelines of the Helsinki Declaration and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist.

The inclusion criteria for patients were as follows: (1) aged between 18 and 80 years, (2) less than 72 h from the time of rupture to admission for treatment, (3) had ruptured bleeding from a single aneurysm, and (4) underwent either surgical clipping (SC) or endovascular coiling (EC) surgery. The exclusion criteria were as follows: (1) a previous history of ruptured aneurysm or SAH; (2) severe cardiac, brain, or lung decompensation; (3) nonaneurysmal SAH; (4) discharge against medical advice; (5) long-term use of antiplatelet drugs; (6) rebleeding; and (7) missing medical, radiological, or laboratory information.

A total of 252 consecutive patients with SAH were eventually enrolled (Fig. 1) and then divided into three groups based on the National Institute of Health Stroke Scale (NIHSS) score. Neurosurgical nurses who had been systematically trained used the NIHSS scale to independently assess daily changes in nervous system function during patient hospitalization. Neurological function deterioration was defined as an NIHSS score ≥ 4 according to previous studies [15, 16]. The enrolled patients were divided into the no neurological deterioration (NND) group (NIHSS score < 4 points during hospitalization; n = 158). For the early neurological deterioration group (END), the NIHSS score change within 72 h after surgery was ≥ 4 points (n = 70); for delayed neurological deterioration (DND), the NIHSS score change after 72 h of surgery was ≥ 4 points (n = 24). The flow diagram of the enrolled cohort is illustrated in Fig. 1.

Fig. 1
figure 1

Flow diagram of patient inclusion. NIHSS National Institute of Health Stroke Scale, PCR polymerase chain reaction

Full size image

Parameter Retraction

All patient data were collected from patient medical records. The selection of indicators associated with the prognosis of EBI in patients with aSAH was guided by the underlying concept and pathophysiological mechanism [17]. Several of these indicators have been previously reported to be correlated with EBI prognosis. We collected demographic information, medical history, history of brain surgery, size and location of the aneurysm, in-hospital complications, and indicators theoretically correlated with EBI. The indicators of microcirculation disorder in EBI included regional cerebral oxygen saturation (rSO2) measured by near-infrared spectroscopy (NIRS), brain oxygen monitoring [17, 18], laboratory examination, clinical signs and symptoms; the consciousness scores included loss of consciousness and postictus at admission; and the World Federation of Neurosurgical Societies (WFNS) grade. The imaging characteristics and scores included acute hydrocephalus, the Barrow Neurological Institute (BNI) grading scale before surgical treatment [19], after neurological recovery World Federation of Neurosurgical Societies (rWFNS) grade [20], and the SAFIRE grade [21], which was previously reported to be associated with the prognosis of SAH.

Laboratory examinations included routine blood examination (white blood cell count, red blood cell count, hemoglobin, neutrophil count, lymphocyte count, platelet count, monocyte count, and neutrophil-to-lymphocyte ratio (NLR) [22]); blood biochemical examination (albumin, potassium ion, sodium ion, glucose, total cholesterol, triglyceride, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and anion gap); partial clotting indices (fibrinogen mean, D-dimer, FDP); and procalcitonin. In addition, the operation time, operation modality, and dynamic NIHSS score during hospitalization were also collected.

Statistical Analysis

Prior to conducting the data analyses, all the variables were thoroughly examined for missing values. Among the predictors, the percentage of missing data varied from 0 to 23%. To incorporate these data into the analysis, we employed multiple imputations using mean imputations to address the missing values. Any variables with a missing data rate exceeding 50% were removed from the analysis.

The sum test was used to compare the quantitative variables. We compared differences in demographic information, clinical symptoms, grading of scales, laboratory findings, and treatments between patients in the three groups. The descriptive variables are presented as the means (standard deviations) or medians with interquartile ranges (IQRs) for continuous variables and as frequencies (percentages) for categorical variables. Chi-squared tests or Fisher’s exact tests were used to compare categorical variables, one-way analysis of variance (ANOVA) or the Kruskal–Wallis rank test with p < 0.05 in the univariate analysis were used for univariate logistic regression, and only variables with p < 0.05 in the univariate logistic regression were included in the multivariate unordered multiple classification logistic regression analysis to identify the independent EBI risk factors associated with early and delayed neurological deterioration during the perioperative period. Associations are expressed as odds ratios (ORs) and 95% confidence intervals (CIs). To further investigate the changes in regional cerebral oxygen saturation with the location and duration of major diffuse intracranial hemorrhage, we used repeated measures ANOVA to determine the differences and trends. Moreover, a receiver operating characteristic (ROC) curve was generated to determine the role of potential predictors of neurological deterioration in patients with aSAH, and p < 0.05 (two-tailed) was considered to indicate statistical significance. All the statistical analyses were performed using SPSS Statistics 25.0 (IBM, Armonk, New York, USA), and the ROC curve and line chart of the changes in regional cerebral oxygen concentrations were drawn with GraphPad Prism 8.0.2.

Results

Patient Characteristics

Of all 252 patients included in the study, 158 (62.7%) patients were included in the no neurological deterioration group, 70 (27.8%) patients were included in the early neurological deterioration (END) group, and 24 (9.5%) patients were included in the delayed neurological deterioration (DND) group (Fig. 1). The demographic and clinical characteristics are shown in Table 1. The mean age was 56 years (SD 10.91), and 38.5% of the patients were male. There were 119 (47.2%) patients with preoperative hypertension, 18 (7.1%) with diabetes mellitus, 6 (2.4%) with coronary heart disease, 58 (23.0%) with a smoking history, 60 (23.8%) with a drinking history, and 12 (4.8%) with a history of brain surgery. In this study, 19 (7.5%) patients had acute hydrocephalus at admission, 197 (78.17%) patients had WFNS grades of 4–5, 131 (51.98%) had Fisher grade of 3–4, and 129 (51.20%) had BNI score of 4–5. In addition, 128 (50.8%) patients had SAFIRE grades 3–5 within 24 h after surgery. Aneurysms were more commonly found in the anterior circulation in 235 (93.3%) patients in the present study. Among all the patients, 184 (73.02%) underwent interventional radiological procedures, whereas 68 (26.98%) underwent surgical clipping. Twenty-two (8.7%) patients underwent a second surgery within a short period, 5 (22.73%) of whom underwent decompressive craniectomy (DC) via implantation of an intracranial pressure probe for cerebral edema and 17 (77.27%) of whom underwent extraventricular drainage for acute hydrocephalus. The remaining related variables were collected, and the characteristics of the three groups are also summarized in Supplementary Table 1.

Table 1 Demographic, baseline characteristics and laboratory findings of patients with aneurysmal subarachnoid hemorrhage
Full size table

Compared with those in the non deterioration group, the median age was greater in the END and DND groups (58 versus 54 years, 59 versus 54 years, p = 0.029). All three groups were predominantly female; however, the sex difference was not significant. The END and DND groups had a greater proportion of patients with a history of hypertension, brain surgery, acute hydrocephalus, reoperation, surgical clipping, BNI score 4–5, Fisher grade 3–4, or SAFIRE grade 3–5. On the other hand, these groups exhibited a lower proportion of patients with coronary heart disease, a history of smoking and drinking, those who received endovascular coiling, and those with BNI scores 1–3, Fisher grade 1–2, or SAFIRE grade 1–2. Additionally, a greater proportion of patients in the early deterioration group had diabetes mellitus and a WFNS grade of 1–3 than did those in the no deterioration group, whereas a lower proportion of patients in the delayed deterioration group had these characteristics than did those in the no deterioration group. However, there was no significant difference in the location of the responsible aneurysm among the three groups. No COVID-19 cases were observed in our study cohort, despite the ongoing COVID-19 pandemic.

Laboratory Findings in Patients with aSAH in the EBI Period

The laboratory data collected during the perioperative period are summarized in Table 1. Routine blood examinations, biochemical blood examinations, and blood clotting index examinations were conducted at four time points (i.e., at admission, within 24 h of surgery, within 24–48 h of surgery, and within 48–72 h of surgery). The results showed that the mean white blood cell (WBC) counts were 12.62 (SD 4.32), 25.44 (SD 3.42), 10.98 (SD 3.57), and 9.75 (SD 3.59), respectively, with only the WBC count at admission being significantly different (p = 0.008) between the three groups. The average red blood cell (RBC) counts were 4.32 (SD 0.56), 5.58 (SD 0.56), 3.80 (SD 0.57), and 3.80 (SD 0.69), respectively, with only the RBC count within 24 h of surgery being significantly different (p = 0.001) among the three groups. The median neutrophil counts were 10.04 (IQR 7.69–13.06), 8.27 (IQR 6.52–10.32), 9.15 (IQR 7.31–11.67), and 8.05 (IQR 6.06–10.15), respectively, with the neutrophil count at admission (p = 0.007), within 24–48 h (p = 0.031), and within 48–72 h of surgery (p = 0.014) being significantly different between the three groups. The median lymphocyte counts were 1.06 (IQR 0.75–1.48), 1.09 (IQR 0.72–1.66), 0.60 (IQR 0.42–0.86), and 0.65 (IQR 0.43–0.94), respectively, with only the lymphocyte count within 48–72 h of surgery being significantly different (p = 0.002) between the three groups.

The median neutrophil-to-lymphocyte ratio (NLR) values were 10.42 (IQR 5.77–14.55), 7.20 (IQR 4.60–13.14), 15.71 (IQR 9.92–22.85), and 12.42 (IQR 7.82–19.72), respectively, there were significant difference in NLR between 24–48 h and 48–72 h (p = 0.018, p = 0.001, respectively). The mean albumin levels were 43.04 (SD 3.77), 46.50 (SD 5.32), 35.62 (SD 3.52), and 35.39 (SD 3.91), respectively, with albumin levels within 24 h (p = 0.021) and 24–48 h of surgery (p = 0.014) being significantly different. The mean serum potassium ion levels were 3.74 (SD 0.40), 5.94 (SD 0.51), 3.80 (SD 0.39), and 3.79 (SD 0.38), respectively, with only the serum potassium ion level within 24–48 h of surgery (p = 0.021) being significantly different among the three groups. The mean anion gap were 12.04 (SD 3.73), 10.53 (SD 3.42), 12.08 (SD 2.45), and 11.81 (SD 2.37), respectively, with the anion gap occurring within 24–48 h of surgery (p = 0.021) and within 48–72 h of surgery (p = 0.037) being significantly different. The median procalcitonin (PCT) levels were 0.05 (IQR 0.04–0.11), 0.09 (IQR 0.04–0.23), 0.13 (IQR 0.05–0.32), and 0.21 (IQR 0.06–0.80), respectively, with only the PCT within 24 h of surgery being significantly different (p = 0.022) among the three groups. Regarding cerebral oxygen saturation (rSO2), within 24 h (p = 0.016), 24–48 h (p = 0.030), and 48–72 h of surgery (p = 0.007) of surgery, the left brain rSO2 concentration significantly differed among the three groups. However, there were no significant differences in the remaining variables.

Compared with those in the no neurological deterioration group, patients in the early deterioration group had higher neutrophil counts; serum sodium ion levels at admission; procalcitonin levels within 24 h of surgery; neutrophil count; NIR and anion gap within 24–48 and 48–72 h of surgery; and serum sodium ion within 48–72 h of surgery. Additionally, these patients had lower WBC counts; RBC counts; albumin levels; serum potassium ion counts; and lymphocyte counts. The delayed neurological deterioration group also had the same result, excluding the WBC count at admission and the left rSO2 within 24 h of surgery.

Factors Predictive of Neurological Deterioration

The univariate and multivariate analyses for predictive indicators of neurological deterioration among patients with aSAH are summarized in Table 2. A logistic regression model was constructed to reveal the potential indicators for END and DND in patients with aSAH. According to the univariate analysis, the following risk factors were associated with END: age > 55 years, reoperation, SAFIRE grade 3–5, reduced WBC count and increased neutrophil count at admission, NIR, and serum sodium ion within 48–72 h of surgery; decreased albumin and rSO2 within 24 h of surgery; decreased albumin and serum potassium ion within 24–48 h of surgery; and decreased lymphocyte count and rSO2 within 48–72 h of surgery.

Table 2 Predictors of neurological deterioration in patients
Full size table

In the DND group, the following risk factors were identified: reoperation, with history of brain surgery, WFNS grade 4–5, Fisher grade 3–4, BNI score 4–5, SAFIRE grade 3–5, increased WBC and neutrophil count at admission; procalcitonin within 24 h of surgery; neutrophil count; NIR and anion gap within 24–48 h of surgery; NIR, serum sodium ion, and anion gap within 48–72 h of surgery; and decreased RBC count within 24 h of surgery and rSO2 in the left frontal area within 24–48 h.

Multivariate logistic regression revealed early predictors of END in patients with aSAH. The results showed that reoperation, with history surgery of brain, WFNS grade 4–5, Fisher grade 3–4, serum sodium ion concentration at admission, SAFIRE grade 3–5, rSO2 (left frontal area) within 24 h of surgery, serum potassium ion concentration within 24–48 h of surgery, and rSO2 (left and right frontal areas) within 48–72 h of surgery were independent predictive indicators of early neurological deterioration. However, only with history of surgery in the brain and a decreased of RBC count within 24 h of surgery were potential independent predictive indicators for DND.

By CT scanning, we found that according to the degree of primary and secondary intracranial distribution of ruptured aneurysm hemorrhage can be divided into the cerebral hemispheres located on the left side, the right hemisphere, and bilateral cerebral hemispheres of the left and right sides of the brain equally. To further determine the changes and relationship between the primary diffuse cerebral hemisphere and regional cerebral oxygen saturation (rSO2) of aneurysmal hemorrhage in patients with neurological deterioration, repeated ANOVA was conducted (Table 3, Fig. 2). The changes in rSO2 in the left and right frontal areas of the brain of patients with aSAH with neurological deterioration during the perioperative period (i.e., at admission, within 24 h of surgery, within 24–48 h of surgery, and within 48–72 h of surgery) are shown in Table 3. The results showed that the oxygen saturation in the left and right frontal areas of the brain of the three groups of patients exhibited significant differences (p < 0.001, p = 0.001, and p = 0.013, respectively) during the perioperative period, and changes in the rSO2 values of the three groups of patients over time are graphically presented in Fig. 2.

Table 3 Cerebral oxygen saturation characteristics of the subjects
Full size table
Fig. 2
figure 2

a–c Changes in regional oxygen saturation on the left and right sides of the brain, respectively. *Significant difference (P < 0.05) in the change in the left and right scores at a fixed time point

Full size image

To further confirm the role of the aforementioned (in Table 2) covariates collectively for the prediction of aSAH among patients with neurological deterioration, a ROC curve analysis was conducted (Table 4, Fig. 3). The area under the curve (AUC) for early neurological deterioration was 0.928 (95% CI 0.896–0.920, p < 0.001), which provided a sensitivity and specificity of 88.57% and 84.18%, respectively. The AUC of delayed neurological deterioration was 0.652 (95% CI 0.537–0.772, p = 0.008), which provided a sensitivity and specificity of 60% and 72.78%, respectively.

Table 4 Receiver operating characteristic (ROC) curve data
Full size table
Fig. 3
figure 3

Receiver operating characteristic (ROC) curve. AUC area under the curve

Full size image

Discussion

In this study, the rSO2 was typically lower in patients who experienced neurological deterioration than in those who did not and at least one point in time, and near-infrared spectroscopy could be used to significantly monitor cerebral oxygen differences between the left and right cerebral hemispheres in patients with neurological deterioration. An unordered multiple-classification logistic regression model was constructed, and the results revealed that multiple factors were effective predictors of early neurological deterioration: reoperation, history of brain surgery, WFNS grade 4–5, Fisher grade 3–4, SAFIRE grade 3–5, abnormal serum sodium and potassium levels, and reduced regional cerebral oxygen saturation during the perioperative period. However, for delayed neurological deterioration in patients with aSAH, only a history of brain surgery and perioperative decreased RBC count were predictive indicators. Nevertheless, in the preceding step of the statistical analysis, local cerebral oxygen saturation and biochemical markers were shown to be associated with early and delayed neurological deterioration in patients with EBI, revealing an important role and potential clinical significance of microcirculatory function in neurological deterioration.

In our previous review, we proposed that post-SAH microcirculation disorders can be divided into three stages: compensatory, decompensated, and irreversible [17]. The compensation period for microcirculation disorders mainly depends on the autonomic regulation of cerebral vessels, the effective perfusion of brain tissue that has decreased perfusion, the beginning of the accumulation of harmful substances other than blood disintegration products, and the beginning of brain tissue ischemia and hypoxia. During the decompensation period, as a result of intracranial inflammatory reactions, the blood–brain barrier is damaged, vasogenic edema and other pathophysiological changes occur, and microcirculation blood stasis and effective perfusion are further reduced, resulting in more severe cerebral ischemia and hypoxia. In the irreversible stage, as a result of the extensive formation of microthrombi, a neutrophil trapping network, neuroinflammation, and other factors [23], regardless of systemic circulation and intracranial perfusion pressure, the intracranial microcirculation completely fails, and the blood flow is in a state of “no entry and no exit”, presenting a classic “no-reflow” phenomenon [24]. Therefore, identifying the early microcirculation status of patients with SAH to facilitate timely intervention and treatment in the compensatory and decompensated stages of microcirculation disorders is urgently needed for disease monitoring in the diagnosis and treatment of patients with SAH and EBI.

The changes in rSO2 during the perioperative period also differed between patients with aSAH with neurological deterioration (END and DND) and patients without neurological deterioration. However, the rSO2, an independent predictor of the DND model, was removed from the multivariate logistic regression analysis. The rSO2 measured by NIRS appears to be a reliable marker of regional cerebral tissue oxygenation, indirectly reflecting cerebral perfusion [25]. A review analysis concluded that higher rSO2 monitored during cardiopulmonary resuscitation (CPR) was consistently associated with an increase in the return of spontaneous circulation (ROSC) rate. However, the differential ability of the rSO2 to predict neurological outcomes is unclear [26]. One study involving 163 patients with aSAH showed that the rSO2 was related to DCI and poor 3-month functional outcomes [27]. In contrast, we collected data on rSO2 during the perioperative period in patients with aSAH, which can also be understood as the EBI period (i.e., 1–3 or 4 days after bleeding); this study collected data 5–10 days after bleeding. This difference could help us explain why the rSO2 was not significant in the independent predictive DND model but was an independent predictor of the END model. Notably, we discovered that not all rSO2 values, both bilaterally and at all stages, had a significant impact. When the hemorrhage mainly spread in both hemispheres, there was a significant difference between the left and right rSO2, and the difference was also found on the fourth day after hemorrhage. This pathological change during hypoperfusion showed a similar process in the mouse model; that is, the phenomenon of red cell block or red cell-free capillaries occurred on the fourth day after bleeding [28], which may help explain why at least one time point can be significantly monitored. Regarding the significant differences in bilateral rSO2, in our cohort, the degree of diffusion of ruptured aneurysm hemorrhage in the right and left cerebral hemispheres was characterized by a primary–secondary distribution difference. The combination of a hemorrhage-induced neuroinflammatory response, neuronal cell death, and microvascular spasm due to microcirculatory ischemia results in tissue damage and altered local cerebral oxygenation. Hemodynamic differences before and after rupture of cerebral aneurysms, assessed by quantitative methods, have been explored and found [29]. We cannot deny that penetrating factors such as scalp hair and light scattering affect the infrared light emitted by NIRS during monitoring, but despite these potential confounders, NIRS is still a reliable technique for measuring temporal changes in brain oxygenation over time [30]. Differences in outcomes due to potential influencing factors and the retrospective nature of the study need to be further explored in larger prospective studies.

In the present study, the demographic and clinical manifestations were similar to those in previous studies [31]. Age was significantly associated with neurological deterioration among patients with aSAH. The older the patient is, the worse the basic condition is. In elderly patients, cerebral parenchyma atrophy is common, and the subarachnoid space is enlarged to contain additional blood from ruptured aneurysms [32]. The degree of meningeal fibrosis increases with age, which further leads to impaired CSF circulation on the basis of hemorrhage and reduced CSF absorption [33]; these patients are more prone to hydrocephalus and even neuron damage. In one study involving elderly patients (≥ 60 years) with poor-grade aSAH (WFNS IV and V), survival analysis revealed that increasing age was associated with an increased risk of death after aSAH [34]. Notably, patient chronological age and lower functional status were both independent predictors of worsening clinical and radiological status at admission, as defined by a WFNS grade 4–5 and Fisher grade 3–4, respectively [35]. In the present study, patients who experienced neurological deterioration (END and DND) were not only older but also had a significantly greater proportion of worsening clinical and radiological states than patients without neurological deterioration. Similarly, the higher the Hunt–Hess grade, WFNS grade, and Fisher grade were in patients with aSAH, the greater the risk of poor prognosis was [36,37,38]. The BNI, which predicts symptomatic vasospasm based on the maximum clot thickness of any pool or fissure, was significantly superior to the Fisher grade in predicting the occurrence of DCI according to multicenter external validation analyses [19]. Both showed similar predictive efficacy in our cohort, but only the Fisher grade was included in the final model of END. Heterogeneity in the selection and definition of various primary endpoints (including symptomatic vasospasm, neurodegeneration, DCI) may contribute to the observed disparity. In addition, the selection bias inherent in retrospective studies contributes to the uneven distribution of sample size characteristics to some extent.

Furthermore, we noted that undergoing a second surgery within a short period is a significant contributing factor to neurological complications. We analyzed the reasons for reoperation in our dataset and found that 77.27% of the patients had acute hydrocephalus. This could lead to poor neurological outcomes and severe cognitive deficits [39,40,41,42]. However, complete statistics on the occurrence of hydrocephalus after surgery were not available in our study. The reasons for this were as follows: acute hydrocephalus requires intraventricular drainage (EVD) to reduce harmful secondary reactions and CSF flow obstruction after bleeding [36]. However, the prolonged use of an EVD may complicate the treatment of aSAH and increase the risk of meningitis and/or ventriculitis triggering neurological deterioration. To date, two-sided characterizations of the effectiveness of DC in reducing mortality and the development of neurological deficits due to the loss of brain protection from DC cranial defects and interference with CSF dynamics [37, 38, 43, 44] have been performed.

A large-scale study revealed that patients with responsible aneurysms (< 5 mm in diameter) located in the anterior communicating region had a lower incidence of focal neurological deficits during hospitalization than did those with nonanterior communicating aneurysms. Additionally, aneurysms within the diameter range of 5–25 mm were identified as independent risk factors for new focal neurological deficits and poor mRS scores during hospitalization [45]. However, our statistical analysis demonstrated no significant association between location or size and neurological deterioration. In this study, among patients who did not experience neurological deterioration, the responsible aneurysm was found to have a smaller diameter than was found in the other two groups of patients with neurological deterioration, and a greater proportion of these aneurysms were located in the anterior circulation. Although these differences did not reach statistical significance, they prompted us to consider certain details, including (1) variations in sample size included in the study, hierarchical classification of aneurysm location and size, and definition and measurement of the study endpoint and (2) inherent selection bias that cannot be completely avoided in retrospective real-world studies.

According to the laboratory findings, previous research has shown that inflammatory/immunologic reactions markedly influence outcomes and predict the clinical course of stroke [46, 47]; in particular, the neutrophil-to-lymphocyte ratio [NIR] has received increased amounts of attention [22, 48, 49]. This study was similar to multiple other studies and showed that NIR radiation is significantly associated with adverse functional outcomes in patients with aSAH, but this association was no longer significant after stepwise regression analysis [31, 50]. As the most common type of leukocyte, neutrophils play a major role in inflammation. Elevated neutrophil counts have been associated with poor outcomes [51]. In this study, higher neutrophil counts were significantly associated with neurological deterioration and were an independent predictor of neurological deterioration. A recent multicenter observational study of 6041 patients with aSAH in China revealed that, compared to lower neutrophil counts, higher neutrophil counts were associated with an increased risk of in-hospital mortality, hospital-acquired infections, and delayed nerve ischemia defects [52]. Interestingly, the NIR was also reported to be an independent predictor of poor neurological outcomes. Possible explanations include the following: (1) Neutrophil counts representing acquired infections shortly before or after the aSAH ictus. (2) Neutrophil-derived free oxygen radicals, proteolytic enzymes, and other products are widely considered involved in the pathogenesis of brain–blood barrier dysfunction [53, 54]. For example, our previous study indicated that neutrophil-derived neutrophil extracellular traps lead to microthrombosis and microcirculation dysfunction [55]. Therefore, aSAH severity and mortality could be tightly associated with neutrophil counts [52], and neurological physicians should pay more attention to patients with aSAH and higher neutrophil counts during the perioperative period.

As previously discussed, we propose that clinical medical care personnel should prioritize the EBI clinical stage (high stage of microcirculation disorders) and pay closer attention to biological indicators; additionally, correcting electrolyte imbalances and improving systemic circulation are crucial for optimizing cerebral perfusion and oxygenation. It has been reported that there is a positive correlation between rSO2 and cardiac output [56], independent of the mean arterial blood pressure (MAP). However, it is worth noting that rSO2 may improve as a result of adequate blood flow perfusion and normal blood pressure levels. Furthermore, rSO2 also relies on a sufficient oxygen supply, with the concentration of hemoglobin in the blood potentially contributing to increased rSO2. The integration of neurointensive care techniques can enable monitoring of cerebral blood flow and electroencephalography (EEG) signals. The latter is extremely sensitive to ischemia and hypoxia, can reflect rSO2 to a certain extent by recording changes in brain electrical activity, and can be used to evaluate the metabolic state of brain tissue oxygen demand and oxygen consumption; moreover, the latter has been proven to have excellent ability to predict ischemia in the ICU environment [57]. Sedative drugs are often used to maintain sedation during neurological intensive care. It may affect cerebral blood flow and oxygenation, and professional scientific evaluation methods can help surgeons avoid confusion in clinical judgment.

Limitations

Several limitations of the current study cannot be ignored. First, as a result of its retrospective nature, the bias introduced by the nature of the study type cannot be completely avoided. Second, the sample sizes of patients with early neurological deterioration and delayed neurological deterioration were relatively small; thus, the validity of the predictors derived from our cohort requires further verification in a future study with a larger sample size. Third, as the data were collected from electronic medical records, several severity score data, such as the SAPS, SOFA, Graeb, and SEBES scores, were not analyzed or collected to better supplement and support the conclusions. In addition, subsequent multicenter prospective clinical trials are needed to further verify the effects of these risk factors, and subsequent basic experiments are needed to explore their pathophysiological mechanism and understand their role in disease progression to guide clinicians in the treatment and judgment of disease.

Conclusion

We retrospectively and dynamically collected clinical information on EBI after aSAH. After reoperation, a history of brain surgery, WFNS grades 4–5, Fisher grades 3–4, SAFIRE grades 3–5, abnormal serum sodium and potassium levels, and reduced rSO2 during the perioperative period were predictive indicators of early neurological deterioration, whereas a history of brain surgery and a decreased perioperative RBC count were predictive indicators of delayed neurological deterioration. However, further studies are needed to determine the role of microcirculation and other predictive factors in the neurocritical management of EBI after SAH, as these factors can reduce the incidence of adverse outcomes and mortality during hospitalization.