The association between hemoglobin concentration and clinical outcomes after aneurysmal subarachnoid hemorrhage: Insights from the LongTEAM registry

Abstract Objective The aim of this study is to explore whether preoperative hemoglobin levels could serve as a prospective biomarker for early brain injury in patients with aneurysmal subarachnoid hemorrhage (aSAH). This investigation seeks to discern its association with postoperative complications and unfavorable clinical outcomes. Methods We conducted a comprehensive analysis of data derived from the LongTeam registry, including patients with aSAH diagnosed between January 2015 and September 2021. These patients were stratified into three distinct groups based on their hemoglobin levels: anemic, standard, and elevated HGB. We employed logistic models featuring spline transformations to assess the relationship between HGB levels and in‐hospital complications. Furthermore, a multivariate Cox proportional hazard model was employed to estimate the impact of elevated hemoglobin levels on the hazard function, which was elucidated through Kaplan–Meier curves. Results Our study comprised a total of 988 patients, among whom 115 (11.6%) presented preoperative anemia, and 63 (6.4%) exhibited elevated preoperative HGB levels. Following adjustments for potential confounding factors, no significant disparity in risk was evident between anemic patients and those with standard HGB levels. However, individuals with elevated HGB levels displayed a heightened incidence and an increased risk of developing deep vein thrombosis (DVT, odds ratio [OR] = 2.39, 95% confidence interval [CI] = 1.16–4.91, p = 0.018; hazard ratio [HR] = 2.05, 95% CI 1.08–3.92, p = 0.015). Aberrant HGB concentrations did not demonstrate an association with other clinical outcomes. Conclusion Our findings emphasize that abnormal HGB levels show no association with adverse outcomes at the 90 days mark after accounting for clinical confounding factors in patients with aSAH. Simultaneously, the study illuminates the potential of HGB as an early indicator for identifying patients at a heightened risk of developing DVT.


| INTRODUC TI ON
Aneurysmal subarachnoid hemorrhage (aSAH) is a life-threatening condition characterized by high morbidity and mortality. 1,2rrently, researchers acknowledge two pivotal phases profoundly affecting the prognosis of aSAH patients: the early brain injury (EBI) phase and the subsequent phase of delayed complications. 3 These two periods are intricately intertwined, with EBI capable of exacerbating the onset of delayed complications.EBI encompasses the pathophysiological mechanisms unfolding within 72 h post-aneurysm rupture, excluding iatrogenic injuries.The pathophysiological alterations associated with EBI not only impact the brain but also give rise to systemic complications, precipitating adverse patient outcomes. 4Hence, identifying significant indicators of EBI and its downstream complications is of paramount importance for predicting patient prognosis.
Numerous investigations have underscored the significance of preoperative laboratory assessments in aSAH patients, positing that specific preoperative laboratory parameters can elucidate the pathophysiological trajectory entailed in EBI.6][7][8][9] Hemoglobin (HGB), an intricate protein responsible for oxygen conveyance within erythrocytes, stands as an efficacious indicator of anemia severity. 102][13] Nevertheless, the scrutiny of preoperative HGB concentrations in aSAH patients and their relevance in predicting clinical outcomes post-aSAH remains insufficient. 14,15is study endeavors to appraise the association between preoperative HGB concentration and clinical outcomes subsequent to aSAH, encompassing 10 in-hospital complications and 90 days functional outcomes.These findings will furnish healthcare practitioners with the means to identify patients at high risk of adverse outcomes, enabling the implementation of more efficacious preventive strategies.

| Study design
Patient data were collected from the LongTEAM registry study (Registration No. NCT 04785976), a comprehensive, single-center, observational cohort conducted at Beijing Tiantan Hospital in China. (16,17)The data collection period extended from January 2015 to  S1.

| Outcome assessment
Neurosurgeons conducted follow-up with patients either through telephone consultations or outpatient appointments.The primary outcome was the modified Rankin Scale (mRS) at 90 days post-discharge.
The unfavorable outcome was defined as an mRS score ≥3.The secondary outcomes were the occurrence of in-hospital complications.

| Statistical analysis
We categorized the patients into three groups based on World Health Organization (WHO) criteria and established research: anemic (hemoglobin <120 g/L for women, hemoglobin <130 g/L for men), standard hemoglobin (hemoglobin 120-155 g/L for women, hemoglobin 130-170 g/L for men), and elevated hemoglobin (hemoglobin >155 g/L for women, hemoglobin >170 g/L for men). 19,20ntinuous variables were presented as mean ± standard deviation (SD) for normally distributed data and as median with interquartile range (IQR) for skewed distributions.Categorical variables were expressed as numerical values (percentages).Baseline characteristics among the three groups were compared using chi-square test for categorical variables, ANOVA test for normal distributed continuous variables, and the nonparametric Kruskal-Wallis test for skewed distributed continuous variables.
To evaluate the relationship between HGB levels and clinical outcomes, we employed a logistic regression model.HGB level were discretized into a categorical variable with three tiers: anemic, standard, and elevated.Clinical outcomes were evaluated using binary indicators to determine the occurrence of in-hospital complications and whether the 90 days mRS score exceeded 3. We

| RE SULTS
A total of 988 patients diagnosed with aSAH were included in this study.
Refer to Table 1 for a comprehensive overview of the patients' baseline characteristics according to HGB concentration groups.

| Hemoglobin concentration and clinical outcomes
We identified 181 (18.3%) patients with unfavorable outcomes at 90 days.There were 376 (38.1%) patients with cardiac events, 268 and unfavorable outcomes at 90 days (OR 2.13, 95%CI 1.17-3.89,p = 0.013), compared to patients with standard HGB.In the fully adjusted model (model 3), patients with elevated HGB were only associated with a higher incidence of DVT (OR 2.39, 95%CI 1.16-4.91,p = 0.018), compared to patients with standard HGB.However, the primary endpoint did not demonstrate statistical significance.
Figure 1 illustrates the predicted OR for the onset of DVT as per the adjusted restricted cubic regression model, using HGB levels as the independent variable.The reference level (OR 1) was set at the median HGB level of 140 g/L.Initially, the curve exhibits a gentle incline but escalates notably for HGB levels surpassing the median, signifying that the divergence in risk between the anemic and standard group is inconsequential.Nevertheless, within the elevated HGB group, an elevation in HGB level translates into a substantial upswing in the adjusted risk.

| DISCUSS ION
However, such evidence was not observed among patients with anemia.Our findings emphasize the significance of this phenomenon, even after adjusting for the heterogeneity in age, gender, treatment modalities, medical history, and initial neurological assessments.
2][23] These studies have identified the relationship between lower HGB levels post-aSAH and an elevated risk of adverse patient outcomes, establishing a well-founded hypothesis that links reduced HGB levels to and heightened risk of cerebral hypoxia. 15wever, limited research has focused on the influence of preoperative HGB concentration on the prognosis of aSAH patients, and the relationship between HGB concentration and clinical outcomes remains contentious.A study in 2006, examining HGB levels from day 0 (the day of SAH occurrence) to day 14 demonstrated that higher day 0 HGB levels were associated with a reduced risk of cerebral infarction, independent of vasospasm. 24Conversely, a 2019 study revealed that elevated preoperative blood HGB levels were linked to an increased risk of cerebral ischemia, pulmonary embolism, DVT, and unfavorable outcome among poor-grade high-altitude aSAH patients following clipping. 25Another study in 2022 indicated that patients with higher HGB levels upon admission may experience a greater incidence of DCI and DVT during hospitalization, leading to unfavorable outcomes at 90 days. 26It is worth noting that both studies did not classify HGB levels according to the WHO criteria (one study employed HGB = 160 g/L as the cutoff value to differentiate between standard and elevated HGB groups, while the other treated HGB as a continuous variable without grouping patients).
Furthermore, both studies did not specifically delineate and analyze patients with anemia, which may limit the generalizability of their findings.Our study has contributed new evidence indicating that when we incorporate confounding factors into our analysis, abnormal HGB levels exhibit no discernible association with adverse outcomes.However, higher HGB levels are associated with an increased propensity for DVT.
Individuals with aSAH may confront an elevated susceptibility to DVT due to immobility.The bleeding subsequent to aneurysm rupture induces a prothrombotic state, thereby amplifying the risk of DVT. 279][30] In our study, the   A 2015 study indicated that DVT formation predominantly occurs within the initial 2 weeks following aSAH, with the peak period ranging from days 5 to 9-consistent with our findings. 31 regarding the timely administration of anticoagulant agents to prevent DVT occurrence while mitigating the associated risk of hemorrhagic events. 33e underlying mechanism governing the formation of DVT subsequent to aSAH remains enigmatic.A study conducted in 2020 suggested that an elevated d-dimer level upon hospitalization might augment the likelihood of DVT, implying a robust initial coagulation activation resulting in a hypercoagulable state. 34rthermore, factors such as advanced age, African American ethnicity, male gender, and prolonged stays in the intensive care unit have been linked to an increased risk of DVT. 30 Our investigation introduces novel insights into the identification of individuals predisposed to DVT, with a specific emphasis on patients presenting with elevated HGB levels upon admission, who are 2.4-fold increased susceptibility to developing DVT during hospitalization compared to those with standard HGB levels.To date, the precise mechanism through which heightened HGB concentrations contribute to DVT formation remains elusive.One plausible hypothesis posits that elevated HGB function as a risk factor for heightened blood viscosity, thereby inducing a hypercoagulable state and facilitating thrombosis within the deep venous system of aSAH patients with elevated HGB levels, particularly during prolonged periods of immobility and reduced circulatory flow.Nevertheless, further research is warranted to validate this hypothesis and ascertain whether patients with elevated HGB can derive benefits from early anticoagulant therapy. 35

| LI M ITATI O N S
This study harbors several limitations.Firstly, despite adjusting for numerous confounding factors, the presence of pre-existing medical conditions that may influence hemoglobin concentrations remains unknown.Secondly, the investigation did not delve into the specific subtypes of anemia.Thirdly, the number of patients with elevated HGB was relatively small, potentially impeding the generalizability of the results.Fourthly, data concerning daily HGB values during the initial stages of hospitalization were not collected.
Fifthly, the analysis solely encompassed patients with a solitary aneurysm.

| CON CLUS ION
In this expansive, single-center, observational cohort study conducted in China, 6.4% of patients exhibited elevated HGB levels upon admission, in accordance with the criteria set forth by the World Health Organization.Following adjustment for clinical confounding factors, abnormal HGB levels demonstrate no association with adverse outcomes at 90 days mark.However, it is worth noting that patients who presenting with elevated HGB upon admission exhibited an increased risk of DVT.Our findings furnish robust clinical evidence for identifying individuals at an increased risk of DVT and offer fresh insights into early prevention strategies for DVT in patients suffering from aSAH.

September 2021 . 18 2. 2 |
Diagnosis of aSAH was established through various imaging techniques, including computed tomography, computed tomography angiography, digital subtraction angiography, or lumbar puncture.Inclusion criteria encompassed the following: (1) age ≥ 18 years; (2) emergent admission; (3) manifestation of a single aneurysm; (4) receipt of surgical clipping or endovascular coiling; (5) a time interval between rupture and admission, and from admission to treatment, both within a 72 h window.On the contrary, exclusion criteria encompassed patients lacking baseline HGB levels, those with sickle cell disease, cancer, previous SAH, or physical disability resulting from antecedent illnesses.Furthermore, patients lost to follow-up within 90 days post-discharge were also excluded from the study.This study duly obtained approval from the ethics committee of Beijing Tiantan Hospital (KY 2021-008-01).Thoroughly informed consent for clinical analysis was acquired from all individual participants or their authorized representatives.The analysis strictly adhered to the principles of the Declaration of Helsinki and local ethics policies.Patient management adhered to well-established guidelines.Data collection Data collection involved trained neurosurgeons who gathered demographic information, medical histories, lifestyle-related risk factors, details regarding the location and size of ruptured aneurysms, symptoms experienced after onset, clinical status (including World Federation of Neurological Societies [WFNS] grade, modified Fisher scale [mFS], Graeb score, Subarachnoid Hemorrhage Early Brain Edema Score [SEBES], and acute hydrocephalus), and the employed treatment modalities.Concurrently, laboratory examinations, including HGB concentration measurement, were routinely performed within 24 h of admission.Furthermore, comprehensive data were amassed concerning 10 in-hospital complications: namely cardiac events, delayed cerebral ischemia (DCI), intracranial infection, stress ulcer bleeding, urinary tract infection, hypoproteinemia, pneumonia, muscular calf vein thrombosis (MCVT), deep vein thrombosis (DVT), and lipid metabolism disorder.Detailed diagnostic criteria are accessible in Table
constructed three nested models for each binary outcome, incorporating various levels of adjustment:(1) Model 1 adjusted for age, gender, and treatment; (2) Model 2 further adjusted for medical history, including additional covariates such as hypertension, heart disease, current smoking, and current drinking; (3) Model 3 additionally considered WFNS grade, Graeb score, and SEBES score to address heterogeneity in patients' initial clinical status.With regard to the primary outcome, we incorporate DCI into Model 3. Adjusted odds ratio (OR) with 95% confidence intervals (CIs) were calculated, using the standard group as the reference.To visualize the nonlinear correlation between HGB and outcome variables, we developed a logistic model that integrated HGB as a restricted cubic spline with three knots.The Kaplan-Meier method was utilized to estimate the cumulative incidence of DVT over the follow-up period, with the log-rank test employed to assess risk differences among the three HGB groups.A multivariate Cox proportional hazard (PH) regression model was fitted to estimate the impact of HGB on the hazard function of developing in-hospital DVT.Hazard ratios (HR) with 95%CIs were calculated, with the standard group serving as the reference.Furthermore, we conducted subgroup analysis, stratifying by age (>55 and ≤ 55 years), gender (male and female), hypertension (yes and no), WFNS grade (1-3 and 4-5), and treatment modality (surgical clipping and endovascular coiling).This analysis aimed to identify any significant interactions between these factors and HGB concentrations when assessing in-hospital DVT.The analysis was performed using the following software: SPSS Statistics 26.0 (IBM), R software (Version 4.1.2),and GraphPad PRISM 8.3.0.(GraphPad Software Inc,).The significance level was set at p < 0.05.

F I G U R E 1
Adjusted odds ratios of DVT according to baseline HGB concentrations.The reference is the median of the baseline HGB concentration (140 g/L).Data were fitted using a logistic regression model of the restricted cubic spline with five knots (the 5th, 25th, 50th, 75th, and 95th percentiles) for baseline HGB concentration, adjusting for potential covariates (model 3).F I G U R E 2 Kaplan-Meier curves for in-hospital DVT.

F
I G U R E 3 (A) The Timing of DVT formation.(B) The relationship between length of stays and DVT.(C) The relationship between the number of complications and DVT.(D) The relationship between total hospital charges and DVT.(1 US dollar = 7.05 renminbi [updated on September 21, 2022].incidence of DVT stood at 8.6%, revealing that patients experiencing DVT during hospitalization endured prolonged stays, augmented hospital expenses, and increased in-hospital complications.These factors compounded the medical burden on both patients' families and society at large.Presently, there exists limited knowledge regarding the temporal patterns of DVT formation following aSAH.Familiarity with the timing aids clinicians in implementing targeted preventive measures.

Table 3
presents the ORs pertaining to DVT based on age, gender, hypertension, WFNS grade, and treatment modality.Following the adjustment for all conceivable confounding factors as per model 3, no modifications in the likelihood of DVT were observed concerning age, gender, hypertension, WFNS grade, and treatment modality (all p-values for interaction >0.05).
In this expansive cohort study conducted in China, featuring a comprehensive analysis of in-hospital complications and 90 days functional outcomes among aSAH patients, we have unveiled a notable association.It appears that individuals with elevated TA B L E 1 Baseline information.Adjusted odds ratios for clinical outcomes according to the hemoglobin groups.
Subgroup analysis of association between hemoglobin groups and in-hospital deep vein thrombosis.