The impact of perioperative anemia on postoperative delirium after hip fracture surgeries in older adults: a multicenter retrospective cohort study

Background: Postoperative delirium (POD) is a frequent yet underestimated neurocognitive complication following hip fracture surgery, particularly among older adults. While anemia is a prevalent and potentially modifiable nutritional disorder in this population, but its impact on POD remains uncertain. We aimed to examine the association between perioperative anemia and POD risk in older adults undergoing surgical treatment for intertrochanteric fractures (ITF).

Methods: We conducted a multicenter retrospective cohort study of elderly patients with ITF who underwent surgical fixation in China. Patients were categorized by their nadir perioperative hemoglobin (Hb): ≥103 g/L, 75–<103 g/L, and <75 g/L. Propensity score matching (PSM) with a 1:1:1 optimal algorithm was used to balance baseline variables. Participants were followed for up to 24 months. POD incidence was assessed and compared across groups. We used multivariable logistic regression to identify independent predictors and performed exploratory mediation analyses.

Results: Among 1,694 patients, the overall POD incidence was 9.7%. After PSM, POD incidence differed across Hb groups: 12.4% (<75 g/L), 2.9% (75–<103 g/L), and 7.6% (≥103 g/L). Multivariate analyses identified severe anemia (OR: 3.13; 95% CI: 1.59–6.18; p < 0.001), preoperative transfusions (OR: 1.67, 95% CI: 1.02–2.74, p = 0.042), and transfusion volume (OR: 1.39, 95% CI: 1.02–1.89, p = 0.037) as independent predictors of POD. Exploratory mediation analyses showed the lowest perioperative Hb levels accounted for 55.9% of the effect of anemia on POD, with preoperative transfusions and transfusion volume accounting for 35.4 and 8.7%, respectively.

Conclusion: Severe perioperative anemia is strongly associated with increased POD risk in older adults with ITF, independent of other clinical factors. Optimizing perioperative anemia management should be a central component of surgical care and POD prevention strategies in geriatric orthopedic populations.

1 Introduction

Hip fractures are increasingly common in aging populations and demand substantial surgical and perioperative resources (1, 2). Intertrochanteric fractures (ITF) account for a large proportion of orthopedic admissions and impose a considerable public health burden due to multimorbidity and associated socioeconomic costs (3, 4). Despite improvements in caring for these vulnerable individuals, the one-year mortality rates are alarmingly high, with percentages ranging from 7 to 10% within a month and 12 to 35% within the initial year (5, 6). Notably, in-hospital mortality may approach 10% even with optimal management (7).

As a prevalent and significant complication following hip fracture surgery, postoperative delirium (POD) is an acute brain dysfunction characterized by impaired consciousness and altered cognitive functioning (8–11). Approximately 23% of hospitalized patients experience POD, which is associated with prolonged length of stay, higher healthcare expenditures, and increased readmissions and mortality (12, 13). Neuman et al. (14) reported an alarming 36% mortality rate at six-month follow-up in patients who developed POD, with related medicare expenditures exceeding 6.5 billion dollars annually (15). Numerous studies have identified various risk factors for delirium, including advanced age, body mass index (BMI), American Society of Anesthesiologists (ASA) classification, malnutrition, general anesthesia, parkinsonism, polypharmacy, anemia, depression, dementia, cognitive decline, frailty, alcohol use, visual or hearing impairment, and institutionalization (16–20).

However, some of these risk factors remain controversial, and controlling symptoms through medication alone often does not yield satisfactory outcomes (21, 22). Against this backdrop, anemia merits focused attention as a common and potentially modifiable preoperative condition. Pathophysiologically, anemia may exacerbate cerebral hypoxia and organ vulnerability, contributing to adverse neurological outcomes (23–25). Inadequately managed anemia in hip fracture patients has been linked to functional decline, longer hospitalization, and worse morbidity and mortality (26–29). Observational data suggest that moderate-to-severe anemia is an independent risk factor for POD (30), whereas erythrocyte transfusion, though frequently used, may not prevent POD and could itself be associated with delirium risk (31–33). These findings are particularly relevant in older adults, in whom multimorbidity and malnutrition are common (34). Nevertheless, despite emerging evidence, the link between anemia and delirium remains inadequately explored and subject to debate (35, 36).

To address this gap, we conducted a multicenter retrospective cohort study across four specialist trauma centers at large tertiary referral university hospitals to examine the association between perioperative anemia and POD in older adults undergoing surgical treatment for ITF. Our objective was to generate clinically actionable evidence to inform risk stratification and guide interventions for POD in geriatric orthopedic populations.

2 Methods

2.1 Study design, setting, and population

This multicenter retrospective study included all ITF patients undergoing intramedullary fixation with the proximal femoral nail anti-rotation (PFNA) and was conducted in four large Level I trauma centers located in tertiary referral university hospitals in China between July 2022 and June 2023. All centers, located in major urban areas, adhered to standardized treatment protocols (32). Inclusion criteria included individuals who were 65 years or older, with an admission delay of less than 48 h from initial injury, and were followed up for at least 2 years. Patients who had open or pathological fractures, additional fractures of the ITF, multiple injuries, inability to communicate, mental illness, refusal of surgery, and those treated conservatively due to severe comorbidities were excluded based on specific criteria. Patients were categorized into three groups based on the lowest perioperative hemoglobin (Hb) level: Hb ≥ 103 g/L, 103 > Hb ≥ 75 g/L, and Hb < 75 g/L. The reference points of 75 and 103 g/L were determined using the restricted cubic spline (RCS) curve, which evaluates linear and nonlinear associations in the dose–response relationship between the lowest perioperative Hb level and POD (Figure 1).

Figure 1

Figure 1. Dose-effect relationship between the lowest perioperative Hb level and POD. Evaluated by using a restricted cubic spline model with four knots and adjusted by all perioperative covariables with p-value <0.1. p-nonlinearity showed no statistical significance, which was estimated using the likelihood ratio test comparing the restricted cubic spline model with the linear model. Relative risks were indicated by blue solid line and 95% CIs by blue area, in which the reference points indicated by pink solid lines were 75 g/L and 103 g/L for the Hb_lowest. POD, postoperative delirium; Hb_lowest, the lowest perioperative hemoglobin level.

As an observational study, clinical decision-making remained uninfluenced by investigators. Clinical information during the perioperative period was gathered through predetermined forms, with regular follow-ups at 1, 3, 6, 12, and 24 months after surgery through phone conversations or outpatient appointments. The study protocol was approved by Ethics Committees of the four participating institutions of Honghui Hospital, Xi’an Jiaotong University, Hebei Medical University Third Hospital, Beijing Jishuitan Hospital, Guizhou Hospital, and Baoding First Central Hospital in accordance with the principles of the Declaration of Helsinki (as revised in 2013). All participants or their immediate family members provided written informed consent. To protect privacy, patient information was de-identified. This study adhered to the guidelines outlined in the Helsinki Declaration and the findings were presented in accordance with the Strengthening the Reporting of cohort, cross-sectional, and case–control studies in Surgery (STROCSS) criteria (37).

2.2 Assessment of postoperative delirium

POD was diagnosed according to DSM-5 criteria, based on daily bedside evaluations performed by trained clinicians from postoperative day 1 until discharge. Evaluations systematically assessed acute onset and fluctuation, inattention, and disturbances in awareness and cognition, and excluded alternative explanations (e.g., coma, intoxication). A single evaluation meeting full DSM-5 criteria was considered diagnostic of POD. Clinicians received standardized training before study start, and uncertain cases were resolved by senior adjudication.

2.3 Covariates and definitions

In order to reduce potential confounding bias, a wide variety of possible confounding variables were taken into account using both current research and expert knowledge. Data were extracted from electronic patient records at both institutions, including sex, age, BMI, residence (rural or urban), smoking or drinking, the commonly used visual analog scores (VAS), Geriatric Depression Scale (GDS), functional independence measure (FIM), presence of anxiety, the ASA grade (classified as I to VI), modified Elixhauser comorbidity method (mECM) (38), fracture type, time from injury to hospital admission/surgery, anesthesia type, intraoperative blood loss, surgical duration, and transfusion volume.

Analyses were conducted on the outcomes related to complications such as POD, sudden death, acute heart failure, acute respiratory failure, myocardial infarction, cerebral peduncles, stress ulcer, arrhythmia, pulmonary infection, electrolyte disbalance, hypoproteinemia, stress hyperglycemia, and deep vein thrombosis (DVT), as well as the average LOS and survival time. During the follow-up period, information on survival status and date of death was collected. Follow-up started when the cohort was enrolled, with the endpoint being either death from any cause or the latest follow-up visit, whichever occurred first. Subsequently, mortality rates were documented at 30-day, 90-day, 180-day, 12-month, and 24-month intervals.

BMI classifications were defined as follows: normal (<24 kg/m²), overweight (24 ≤ BMI < 28 kg/m²), and obesity (≥28 kg/m²). Upon admission, the mECM evaluated the comorbidities of patients, categorizing them into groups of less than 0, 0, 1 to 5, 6 to 13, and 14 or more. The 15-question GDS and FIM assessed symptoms of depression and the capacity to carry out ADL, respectively (2). Hb levels at admission were categorized based on thresholds 8 g/dL, 10 g/dL, and 12 g/dL. We followed the standard clinical guidelines (39–41) for perioperative anemia management in elderly hip fracture patients. As for blood transfusions, we adhered to the current restrictive guidelines when the Hb value was less than 80 g/L in general patients or between 80 and 100 g/L with a serious cardiac condition or when anemia becomes symptomatics, which is general practice in hospitals (42, 43). In addition, transfusions were performed preoperatively, intraoperatively, and postoperatively, depending on the patient’s condition and intraoperative blood loss. RCS curves set reference points for patient age, time from injury to hospital admission, time from injury to surgery, intraoperative blood loss, duration of operation, and transfusion volume at specific values of 81 years, 9 h, 6 days, 205 mL, 90 min, and 4 units, respectively (Figure 2).

Figure 2

Figure 2. Dose-effect relationship between patient age (A), time from injury to hospital admission (B), time from injury to surgery (C), intraoperative blood loss (D), duration of operation (E), transfusion volume (F), and POD. Evaluated by using a restricted cubic spline model with four knots and adjusted by all perioperative co-variables with p-value <0.1. p-nonlinearity showed no statistical significance, which was estimated using the likelihood ratio test comparing the restricted cubic spline model with the linear model. Relative risks were indicated by blue solid line and 95% CIs by blue area, in which the reference points indicated by pink solid lines were 81 years, 9 h, 6 days, 205 mL, 90 min, and 4 units for the patient age (A), time from injury to hospital admission (B), time from injury to surgery (C), intraoperative blood loss (D), duration of operation (E), and transfusion volume (F). POD, postoperative delirium.

2.4 Statistical analysis

To evaluate the normality of continuous variables, the Shapiro–Wilk test was utilized. Categorical data was presented in terms of frequency and percentage, whereas continuous data was presented as the mean and standard deviation (SD). Continuous and categorical variables were compared using Student’s t-tests and chi-square tests, respectively. Missing data for continuous BMI (2.3%) were imputed using linear regression. To account for potential linear and nonlinear associations in the dose–response relationships, RCS with four nodes were used for controlling for all covariates with a significance level of less than 0.1, which were placed at the 5th, 35th, 65th, and 90th percentiles of the distribution for continuous variables based on best practices and recommendations from statistical literature, suggesting that positioning the knots at these percentiles provides an optimal balance between flexibility and stability of the model (44). The choice of four knots, as opposed to a higher or lower number, was guided by both the size of our dataset and the need for sufficient flexibility without introducing excessive complexity into the model. Empirical research demonstrates that models with four to five knots are generally appropriate for datasets of moderate size, such as ours, allowing for the detection of nonlinear patterns without reducing statistical power (45–47). The analyses visually examined the dose-effect relationship between the lowest perioperative Hb level, age, time from injury to hospital admission, time from injury to surgery, intraoperative blood loss, duration of operation, transfusion volume, and POD, setting reference points based on RCS shapes and stratifying related data into subgroups accordingly.

To enhance the reliability of our findings and guarantee their applicability to actual clinical settings, we employed propensity score matching (PSM) with a 1:1:1 optimal matching algorithm via the caliper matching of 0.05 to address differences in baseline characteristics across the three groups, ultimately minimizing the effects of selection bias and potential confounding variables. Besides, our study primarily assessed the lowest perioperative hemoglobin levels as a determinant of POD and we acknowledge blood transfusions are commonly administered to correct significant perioperative anemia and the subsequent changes in hemoglobin levels could impact perioperative transfusions and then impact postoperative outcomes, including the risk of POD, particularly for patients who received substantial transfusions. Therefore, we additionally conducted univariate logistical and multiple regression analyses and then mediation analysis was carried out by using multivariate regression analyses, to determine relationships and analyze the mediating effect. The mediation effects were assessed using the product of coefficients method, which directly assessed whether the mediation or indirect effect was significant. In the multivariate analysis, we grouped perioperative transfusion into six categories: whether received preoperative transfusion, intraoperative transfusion, and postoperative transfusion or not, and the total preoperative, intraoperative, and postoperative transfusion volume. Dichotomous outcomes were reported using the odds ratio (OR) and 95% CI. All data analyses were performed using IBM SPSS Statistics for Windows, version 26.0 (IBM, Armonk, NY, USA). A two-tailed p value <0.05 was considered significant.

3 Results

3.1 Patient characteristics and group comparisons

Between July 2022 and June 2023, 2,899 consecutive patients with ITF were screened for eligibility, and 1,694 were included in the final analysis after applying exclusion criteria. In total, 1,205 patients were eliminated based on the exclusion criteria. In particular, there were 303 patients younger than 65, 162 who were treated conservatively, 514 with a delay in admission of 48 h or longer, 87 with open hip fractures, pathological fractures, or multiple injuries, 62 who were unable to communicate or had mental illness, and 77 who were lost to follow-up. The average age of participants was 80.2 years (range 65–106), with an overall POD incidence of 9.7% (145/1694). Baseline characteristics across the three Hb groups (Hb ≥ 103 g/L, 103 > Hb ≥ 75 g/L, and Hb < 75 g/L) differed significantly before PSM in variables such as age, BMI, and transfusion volume. After PSM, all groups showed comparable demographic and clinical characteristics, with the exception of transfusion volume (all p > 0.05, Table 1).

Table 1

Table 1. Comparisons of baseline characteristics before and after matching in three groups of participants.

3.2 Incidence of postoperative delirium (POD) and other outcomes

The incidence of POD varied significantly across Hb groups after PSM: Hb < 75 g/L (12.4%), 103 > Hb ≥ 75 g/L (2.9%), and Hb ≥ 103 g/L (7.6%). Statistical analyses revealed a significant association between severe anemia (Hb < 75 g/L) and POD (p = 0.001). Table 2 provides a detailed overview of pre- and post-matching results, which encompass mortality rates, complications, mean LOS, and mean survival time.

Table 2

Table 2. Comparisons of mortalities, complications, LOS, and survival time before and after matching in three groups of patients.

Statistical analysis showed a significant difference in mean LOS among the three groups before PSM, but this discrepancy became insignificant after PSM. Significantly, the disparities in POD occurrence and mean survival time became notable following the matching process (p = 0.001 and p = 0.003, respectively), in contrast to their lack of significance before matching. Significant differences in 6-month, 1-year, and 2-year mortality rates and stress ulcer incidence were also observed among groups post-PSM, (all p < 0.05). There was no significant difference in complication or mortality rates at 30- and 90-day intervals before or after PSM.

3.3 Multivariate analysis and mediation analysis

Multivariate logistic regression analysis identified severe anemia (OR: 3.13, 95% CI: 1.59–6.18, p < 0.001), preoperative transfusions (OR: 1.67, 95% CI: 1.02–2.74, p = 0.042), and total preoperative transfusion volume (OR: 1.39, 95% CI: 1.02–1.89, p = 0.037) as significant predictors of POD. Mediation analysis showed that 55.9% (OR: 3.13, 95% CI: 1.59–6.18) of the total effect of anemia on POD was mediated by the lowest perioperative Hb levels, while preoperative transfusions and transfusion volume accounted for 35.4% (OR: 1.98, 95% CI: 1.09–3.93) and 8.7% (OR: 1.12, 95% CI: 1.02–1.24), respectively (Table 3; Figure 3).

Table 3

Table 3. The univariate and multivariate analyses of all involved possible factors associated with POD.

Figure 3

Figure 3. Mediation analysis results of anemia and delirium influenced by whether preoperative transfusion and total preoperative transfusion volume. POD, postoperative delirium; Hb_lowest: the lowest perioperative hemoglobin level.

4 Discussion

To the best of our knowledge, POD is not only a frequent occurrence following hip surgery but a significant cause of postoperative mortality. Alarmingly, clinicians fail to recognize and address POD in up to 80% of cases, highlighting a critical gap in postoperative care (48). Previous studies have reported that the incidence of POD in older individuals varies from 2 to 7% in Asian countries, with rates exceeding 50% following hip fractures and cardiac surgeries (49, 50). Therefore, assessing and mitigating risk factors for POD in hospitalized patients is of utmost importance. Choi et al. (49) demonstrated that risk factors such as cognitive decline, parkinsonism, and higher ASA scores are linked to a higher likelihood of developing POD after hemiarthroplasty. Additionally, a comprehensive meta-analysis (19) found a strong association between cognitive dysfunction, BMI, albumin concentrations, and various underlying health conditions with POD following hip fracture surgery. Despite numerous studies attempting to identify key risk factors influencing POD, findings remain inconsistent and controversial. Recent research indicates that hematological indices, including the administration of red blood cell (RBC) transfusions, may also play a role in the progression or mitigation of POD (17, 32, 51). This suggests a potential, yet unexplored, relationship between anemia and POD.

Previous studies examining the link between anemia and POD have yielded conflicting results. For instance, a study of surgical elderly patients found no significant association between anemia and delirium, attributing this to low statistical power (35). Conversely, a cohort study with a larger sample size identified a significant independent association between moderate-to-severe anemia and delirium (30). More recently, a prospective cohort analysis concluded that postoperative anemia is highly prevalent and increases the risk of POD as well as prolongs the LOS (29). These inconsistencies highlight the need for further investigation into the relationship between anemia and POD, considering the limitations of previous studies such as small sample sizes and uncontrolled confounding factors.

In this study, we utilized a multicenter retrospective cohort design to explore the potential link between perioperative anemia and POD. Our cohort study findings demonstrated that ITF patients with severe anemia (lowest perioperative Hb level <75 g/L) are more susceptible to POD. The biological mechanisms underlying the relationship between anemia and POD may involve the body’s dependence on hemoglobin for oxygen uptake and transport. Anemic states, characterized by low Hb levels and leads to a reduction in oxygen-carrying capacity, which can result in chronic or acute hypoxia, ultimately leading to brain dysfunction, cognitive impairment, and mood abnormalities (52–54). Importantly, the brain is highly sensitive to oxygen deprivation, and hypoxia has been shown to disrupt normal neuronal function, contributing to cognitive impairment and neuroinflammation—both of which are known precursors to delirium. Besides, hypoxia can also lead to increased oxidative stress and mitochondrial dysfunction, triggering neuroinflammatory responses which can impair normal brain metabolism and further exacerbate cognitive decline and contribute to delirium (55–57). Brombo et al. (30) found no overall association was discovered between anemia and delirium, however, a significant correlation was noted when anemia was combined with cognitive decline, suggesting that anemia-induced delirium may be mediated by cognitive functions. Specifically, anemia in middle-aged and elderly populations is often accompanied by malnutrition and chronic diseases, particularly in elderly patients with pre-existing vulnerabilities such as frailty or comorbidities (34, 58).

In addition, anemia-related hypoxia can impair cerebral blood flow and the delivery of essential nutrients to the brain. The brain may adapt to chronic anemia through mechanisms such as modulating cervical oxygen partial pressure receptors and cerebral blood flow (59). Hoffman et al. (60) found that the brain response to hypoxia was diminished in older rats compared to younger rats, suggesting that elderly patients’ ability to adapt to anemia may mask a potential link between anemia and POD. Conversely, another study indicated that elderly patients with reduced cerebral oxygenation are more prone to developing delirium postoperatively, especially when combined with the physiological stress of surgery (29). We speculate that substantial blood loss during hip fracture surgeries may expose elderly patients to acute anemia, which their brains have not yet adapted to, resulting in POD and negatively impacting clinical outcomes and mortality. Moreover, Weiskopf et al. found that acute anemia, characterized by a sharp drop in Hb concentration, was associated with decreased responsiveness, impairing cognitive function and memory in humans (61). Furthermore, the co-occurrence of anemia and delirium is often seen in elderly populations where malnutrition and chronic disease further compromise brain function, increasing the susceptibility to POD.

In summary, the risk of POD may be related to the severity of anemia and the patient’s ability to tolerate it. This causal inference may also be interpreted by our retrospective cohort study findings, which suggest that the Hb < 75 g/L group has the highest frequency of POD (12.4%), while the 103 > Hb ≥ 75 g/L group has the lowest frequency of POD (2.9%). Interestingly, the Hb ≥ 103 g/L group had the second-highest rate of POD (7.6%), approximately 2.6 times higher than the 103 > Hb ≥ 75 g/L group. The high frequency of extensive blood transfusions (≥4 units) in the Hb ≥ 103 g/L group could explain our previous research results (17, 32). These complex results may explain the contradictions among previous observational studies. Importantly, anemia, being an easily assessed and modifiable indicator, is simple and feasible for predicting and managing POD. However, it is crucial to consider whether the patient has adapted to chronic anemia. Future studies with larger sample sizes are necessary to investigate the correlation between various types of symptomatic anemia and POD.

This study has several strengths. First, in a multicenter retrospective cohort we adjusted for numerous previously identified risk factors as covariates, enhancing the robustness of the findings. Second, we complemented the multicenter observational cohort study with mediation analyses, thereby bolstering the credibility and reliability of our conclusions. Nonetheless, several limitations warrant consideration. First, we did not have complete data on anesthetic and perioperative medications and therefore could not adjust for their use. Certain agents are known to influence delirium risk (e.g., ketamine, opioids, and benzodiazepines), whereas others are not (e.g., dexmedetomidine and lidocaine). Lack of control for these exposures may have introduced residual confounding. Second, the number of POD events was modest, which limited the statistical power for additional stratified or interaction analyses. Third, our cohort was intentionally restricted to a homogeneous surgical technique with the same internal fixation. Although this design choice reduces procedural heterogeneity and the attendant risk of bias, it may limit generalizability because different operative approaches can differentially affect delirium risk. Fourth, despite the strength of use of PSM to control for confounding variables, certain variables remained unbalanced after matching. This residual imbalance is a known limitation of PSM, particularly in large, heterogeneous cohorts where subtle differences between subgroups may persist. While PSM improves comparability between groups, it cannot entirely eliminate all discrepancies, especially in observational studies with complex datasets. It is important to acknowledge that unaccounted-for variables such as unmeasured perioperative variables, other protocols for treating anemia including iron supplementation both oral or intravenous, erythropoiesis-stimulating agents and oxygen therapy may still influence the results, and further studies utilizing alternative methods may be needed to validate these findings. Another notable limitation of our study is the exclusion of patients with pre-existing mental illness and those unable to communicate may introduce selection bias, which may reduce the generalizability of our findings to real-world clinical settings. Specifically, older adults with cognitive impairments or psychiatric comorbidities may present unique risk factors for both anemia and POD that were not captured in our study. Consequently, caution is warranted when applying our conclusions to broader clinical populations, particularly those more vulnerable to POD due to mental health or neurological conditions. This would ensure a more comprehensive understanding of how anemia and categorize patients based on the etiology of anemia (e.g., iron deficiency anemia, anemia of chronic disease, vitamin B12 deficiency) and examine the distinct pathways through which each type of anemia contributes to POD across diverse patient populations, thereby enhancing the clinical applicability of the findings in future research. Lastly, considering the total effect of exposure can pass both through the mediators (indirect effect) and through a direct pathway (direct effect), the role of the mediating effect cannot be overstated. However, the inclusion of important covariates as potential confounders in the mediation analyses ensures that estimates of the mediation effects are less likely to be exaggerated. In conclusion, while PSM was employed to reduce confounding, it cannot completely eliminate all residual bias inherent in observational studies. To mitigate this, we complemented the observational study with mediation analyses, which strengthens the causal inference and minimizes the risk of confounding, thereby providing a more reliable estimate of the impact of perioperative anemia on POD. Consequently, caution is warranted when applying our conclusions to broader clinical populations, particularly those more vulnerable to POD due to mental health or neurological conditions. This would ensure a more comprehensive understanding of how anemia and categorize patients based on the etiology of anemia (e.g., iron deficiency anemia, anemia of chronic disease, vitamin B12 deficiency) and examine the distinct pathways through which each type of anemia contributes to POD across diverse patient populations, thereby enhancing the clinical applicability of the findings in future research.

5 Conclusion

In this multicenter retrospective cohort of older adults undergoing surgery for ITF, severe perioperative anemia was strongly associated with an increased risk of postoperative delirium, independent of other clinical factors. This research provides a significant step forward in understanding the importance of optimizing perioperative anemia management as a key component of surgical nutritional care and POD prevention strategies in geriatric orthopedic populations. Given the observational design, prospective validation is warranted. Future research should include multicenter prospective cohorts and adequately powered randomized trials to evaluate structured anemia-management strategies using standardized delirium assessments and comprehensive measurement of perioperative exposures.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Ethics statement

The studies involving humans were approved by Ethics Committee of the four participating institutions including Honghui Hospital, Xi’an Jiaotong University (IRB number: 202312020), Hebei Medical University Third Hospital (IRB number: 2022-052-1), Beijing Jishuitan Hospital, Guizhou Hospital (IRB number: LW2025072301), and Baoding First Central Hospital (IRB number: 2022-064). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

XX: Writing – review & editing, Investigation, Conceptualization, Validation, Funding acquisition, Writing – original draft, Methodology, Formal analysis, Data curation. PX: Validation, Project administration, Conceptualization, Funding acquisition, Supervision, Writing – review & editing, Resources. JuW: Writing – review & editing, Conceptualization, Investigation, Software, Formal analysis, Data curation, Methodology, Visualization. HY: Methodology, Formal analysis, Software, Visualization, Conceptualization, Data curation, Writing – review & editing. YLi: Validation, Investigation, Writing – review & editing, Data curation, Methodology. QZ: Investigation, Resources, Writing – review & editing, Data curation, Methodology, Formal analysis. TW: Validation, Data curation, Writing – review & editing, Investigation, Resources. YLo: Investigation, Validation, Data curation, Writing – review & editing, Resources. EL: Writing – review & editing, Methodology, Investigation, Visualization, Validation, Resources. JiW: Writing – review & editing, Formal analysis, Resources, Methodology, Validation. JX: Validation, Resources, Data curation, Investigation, Writing – review & editing. JG: Supervision, Visualization, Resources, Writing – review & editing, Funding acquisition, Conceptualization, Writing – original draft, Formal analysis, Project administration, Data curation.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was funded by the National Natural Science Foundation of China (No. 82502974), Scientific Research and Innovation Platform for Intelligent and Precise Treatment of Bone and Joint Diseases in Shaanxi Province (No. 2024PT-13), the Natural Science Basic Research Program of Shaanxi Province (Nos. 2025JC-YBQN-1113 and 2025JC-YBQN-1181), and Postdoctoral Fund of Shaanxi Province (No. 2023BSHGZZHQYXMZZ02). The funders did not have any role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Acknowledgments

The authors thank Home for Researchers editorial team (www.home-for-researchers.com) for improving the English language in this manuscript. The authors thank Chongqing Nanpeng Artificial Intelligence Technology Research Institute Co., Ltd. for supporting part of the data extraction and processing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Abbreviations

ITF, Intertrochanteric fractures; POD, Postoperative delirium; LOS, Length of hospital stay; BMI, Body mass index; ASA, American Society of Anesthesiologists; PFNA, proximal femoral nail anti-rotation; Hb, hemoglobin; RCS, restricted cubic spline; VAS, Visual analog scores; GDS, Geriatric Depression Scale; FIM, Functional independence measure; mECM, modified Elixhauser comorbidity method; DVT, Deep vein thrombosis; SD, Standard deviation; PSM, propensity score matching; RBC, Red blood cell.

References

1. Zhang, Q, Wu, Y, Han, T, and Liu, E. Changes in cognitive function and risk factors for cognitive impairment of the elderly in China: 2005-2014. Int J Environ Res Public Health. (2019) 16:62847. doi: 10.3390/ijerph16162847,

PubMed Abstract | Crossref Full Text | Google Scholar

2. Guo, J, Wang, T, Zheng, X, Long, Y, Wang, X, Zhang, Q, et al. Use of intravenous paracetamol preoperatively favors lower risk of delirium and functional recovery in elderly patients with hip fracture: a propensity score-matched analysis. Pain Res Manag. (2022) 2022:1–9. doi: 10.1155/2022/1582727,

PubMed Abstract | Crossref Full Text | Google Scholar

3. Cui, Y, Lu, Y, Huang, Q, Zhang, C, Sun, L, Ren, C, et al. Clinical application effects of different preoperative blood management schemes in older patients with delayed intertrochanteric fracture surgery. Clin Interv Aging. (2022) 17:825–35. doi: 10.2147/CIA.S362020,

PubMed Abstract | Crossref Full Text | Google Scholar

4. Tang, Y, Kang, L, Guo, M, and Fan, L. Intertrochanteric fracture surgery patients with diabetes mellitus are prone to suffer perioperative neurological and endocrine/metabolic complications: a propensity-score matched analysis. Ther Clin Risk Manag. (2022) 18:775–83. doi: 10.2147/TCRM.S366846,

PubMed Abstract | Crossref Full Text | Google Scholar

5. Mundi, S, Pindiprolu, B, Simunovic, N, and Bhandari, M. Similar mortality rates in hip fracture patients over the past 31 years. Acta Orthop. (2014) 85:54–9. doi: 10.3109/17453674.2013.878831,

PubMed Abstract | Crossref Full Text | Google Scholar

6. Roth, T, Kammerlander, C, Gosch, M, Luger, TJ, and Blauth, M. Outcome in geriatric fracture patients and how it can be improved. Osteoporos Int. (2010) 21:S615–9. doi: 10.1007/s00198-010-1401-4,

PubMed Abstract | Crossref Full Text | Google Scholar

7. Brauer, CA, Coca-Perraillon, M, Cutler, DM, and Rosen, AB. Incidence and mortality of hip fractures in the United States. JAMA. (2009) 302:1573–9. doi: 10.1001/jama.2009.1462,

PubMed Abstract | Crossref Full Text | Google Scholar

8. Chen, Y, Liang, S, Wu, H, Deng, S, Wang, F, Lunzhu, C, et al. Postoperative delirium in geriatric patients with hip fractures. Front Aging Neurosci. (2022) 14:1068278. doi: 10.3389/fnagi.2022.1068278,

PubMed Abstract | Crossref Full Text | Google Scholar

9. Keenan, CR, and Jain, S. Delirium. Med Clin North Am. (2022) 106:459–69. doi: 10.1016/j.mcna.2021.12.003,

PubMed Abstract | Crossref Full Text | Google Scholar

10. Mattison, MLP. Delirium. Ann Intern Med. (2020) 173:ITC49–64. doi: 10.7326/AITC202010060

Crossref Full Text | Google Scholar

11. Oosterhoff, JHF, Karhade, AV, Oberai, T, Franco-Garcia, E, Doornberg, JN, and Schwab, JH. Prediction of postoperative delirium in geriatric hip fracture patients: a clinical prediction model using machine learning algorithms. Geriatr Orthop Surg Rehabil. (2021) 12:21514593211062277. doi: 10.1177/21514593211062277,

PubMed Abstract | Crossref Full Text | Google Scholar

12. Gibb, K, Seeley, A, Quinn, T, Siddiqi, N, Shenkin, S, Rockwood, K, et al. The consistent burden in published estimates of delirium occurrence in medical inpatients over four decades: a systematic review and meta-analysis study. Age Ageing. (2020) 49:352–60. doi: 10.1093/ageing/afaa040,

PubMed Abstract | Crossref Full Text | Google Scholar

13. Setters, B, and Solberg, LM. Delirium. Prim Care. (2017) 44:541–59. doi: 10.1016/j.pop.2017.04.010,

PubMed Abstract | Crossref Full Text | Google Scholar

14. Neuman, MD, Silber, JH, Magaziner, JS, Passarella, MA, Mehta, S, and Werner, RM. Survival and functional outcomes after hip fracture among nursing home residents. JAMA Intern Med. (2014) 174:1273–80. doi: 10.1001/jamainternmed.2014.2362,

PubMed Abstract | Crossref Full Text | Google Scholar

15. Rubin, FH, Neal, K, Fenlon, K, Hassan, S, and Inouye, SK. Sustainability and scalability of the hospital elder life program at a community hospital. J Am Geriatr Soc. (2011) 59:359–65. doi: 10.1111/j.1532-5415.2010.03243.x,

PubMed Abstract | Crossref Full Text | Google Scholar

16. Li, T, Li, J, Yuan, L, Wu, J, Jiang, C, Daniels, J, et al. Effect of regional vs general anesthesia on incidence of postoperative delirium in older patients undergoing hip fracture surgery: the RAGA randomized trial. JAMA. (2022) 327:50–8. doi: 10.1001/jama.2021.22647,

PubMed Abstract | Crossref Full Text | Google Scholar

17. Wang, T, Guo, J, Hou, Z, and Zhang, Y. Risk factors of postoperative delirium in elderly patients with intertrochanteric fracture: an age-stratified retrospective analysis of 2307 patients. Geriatr Orthop Surg Rehabil. (2022) 13:21514593221081779. doi: 10.1177/21514593221081779,

PubMed Abstract | Crossref Full Text | Google Scholar

18. Kalisvaart, KJ, Vreeswijk, R, de Jonghe, JF, van der Ploeg, T, van Gool, WA, and Eikelenboom, P. Risk factors and prediction of postoperative delirium in elderly hip-surgery patients: implementation and validation of a medical risk factor model. J Am Geriatr Soc. (2006) 54:817–22. doi: 10.1111/j.1532-5415.2006.00704.x,

PubMed Abstract | Crossref Full Text | Google Scholar

19. Oh, ES, Li, M, Fafowora, TM, Inouye, SK, Chen, CH, Rosman, LM, et al. Preoperative risk factors for postoperative delirium following hip fracture repair: a systematic review. Int J Geriatr Psychiatry. (2015) 30:900–10. doi: 10.1002/gps.4233,

PubMed Abstract | Crossref Full Text | Google Scholar

20. Qi, YM, Li, YJ, Zou, JH, Qiu, XD, Sun, J, and Rui, YF. Risk factors for postoperative delirium in geriatric patients with hip fracture: a systematic review and meta-analysis. Front Aging Neurosci. (2022) 14:960364. doi: 10.3389/fnagi.2022.960364,

PubMed Abstract | Crossref Full Text | Google Scholar

21. Palakshappa, JA, and Hough, CL. How we prevent and treat delirium in the ICU. Chest. (2021) 160:1326–34. doi: 10.1016/j.chest.2021.06.002,

PubMed Abstract | Crossref Full Text | Google Scholar

22. Davis, D, Searle, SD, and Tsui, A. The Scottish intercollegiate guidelines network: risk reduction and management of delirium. Age Ageing. (2019) 48:485–8. doi: 10.1093/ageing/afz036,

PubMed Abstract | Crossref Full Text | Google Scholar

23. Adanho, CSA, Yahouédéhou, S, Santana, SS, Vieira, C, Santiago, RP, de Santana, JM, et al. Association of laboratory markers and cerebral blood flow among sickle cell anemia children. Front Pediatr. (2022) 10:914466. doi: 10.3389/fped.2022.914466

Crossref Full Text | Google Scholar

24. Heo, J, Youk, TM, and Seo, KD. Anemia is a risk factor for the development of ischemic stroke and post-stroke mortality. J Clin Med. (2021) 10:122556. doi: 10.3390/jcm10122556,

PubMed Abstract | Crossref Full Text | Google Scholar

25. LaVallee, C, Bansal, I, Kamdar, S, Kwong, WJ, and Boccia, RV. Relationship between initial parenteral Iron therapy dosing and treatment effectiveness: a real-world retrospective analysis. J Blood Med. (2022) 13:133–42. doi: 10.2147/JBM.S349070,

PubMed Abstract | Crossref Full Text | Google Scholar

26. Zhang, BF, Wang, J, Wen, PF, Wu, YJ, Guo, JB, Wang, YK, et al. The association between hemoglobin at admission and mortality of older patients with hip fracture: a mean 3-year follow-up cohort study. Eur Geriatr Med. (2023) 14:275–84. doi: 10.1007/s41999-023-00759-0,

PubMed Abstract | Crossref Full Text | Google Scholar

27. Joosten, E, Lemiengre, J, Nelis, T, Verbeke, G, and Milisen, K. Is anaemia a risk factor for delirium in an acute geriatric population? Gerontology. (2006) 52:382–5. doi: 10.1159/000095126,

PubMed Abstract | Crossref Full Text | Google Scholar

28. Tahir, M, Malik, SS, Ahmed, U, Kozdryk, J, Naqvi, SH, and Malik, A. Risk factors for onset of delirium after neck of femur fracture surgery: a prospective observational study. SICOT J. (2018) 4:27. doi: 10.1051/sicotj/2018018,

PubMed Abstract | Crossref Full Text | Google Scholar

29. Kunz, JV, Spies, CD, Bichmann, A, Sieg, M, and Mueller, A. Postoperative anaemia might be a risk factor for postoperative delirium and prolonged hospital stay: a secondary analysis of a prospective cohort study. PLoS One. (2020) 15:e0229325. doi: 10.1371/journal.pone.0229325,

PubMed Abstract | Crossref Full Text | Google Scholar

30. Brombo, G, Bianchi, L, Savino, E, Magon, S, Cherubini, A, Corsonello, A, et al. The relationship of hemoglobin levels, delirium and cognitive status in hospitalized geriatric patients: results from the CRIME study. J Prev Alzheimers Dis. (2015) 2:178–83. doi: 10.14283/jpad.2015.49,

PubMed Abstract | Crossref Full Text | Google Scholar

31. Janssen, TL, Steyerberg, EW, Faes, MC, Wijsman, JH, Gobardhan, PD, Ho, GH, et al. Risk factors for postoperative delirium after elective major abdominal surgery in elderly patients: a cohort study. Int J Surg. (2019) 71:29–35. doi: 10.1016/j.ijsu.2019.09.011,

PubMed Abstract | Crossref Full Text | Google Scholar

32. Guo, J, Geng, Q, Xu, K, Jing, W, Li, Z, He, J, et al. Development and validation of models for predicting mortality in intertrochanteric fracture surgery patients with perioperative blood transfusion: a prospective multicenter cohort study. Int J Surg. (2024) 110:4754–66. doi: 10.1097/JS9.0000000000001472,

PubMed Abstract | Crossref Full Text | Google Scholar

33. van der Zanden, V, Beishuizen, SJ, Swart, LM, de Rooij, SE, and van Munster, BC. The effect of treatment of Anemia with blood transfusion on delirium: a systematic review. J Am Geriatr Soc. (2017) 65:728–37. doi: 10.1111/jgs.14564,

PubMed Abstract | Crossref Full Text | Google Scholar

34. Stauder, R, Valent, P, and Theurl, I. Anemia at older age: etiologies, clinical implications, and management. Blood. (2018) 131:505–14. doi: 10.1182/blood-2017-07-746446,

PubMed Abstract | Crossref Full Text | Google Scholar

35. Myint, PK, Owen, S, McCarthy, K, Pearce, L, Moug, SJ, Stechman, MJ, et al. Is anemia associated with cognitive impairment and delirium among older acute surgical patients? Geriatr Gerontol Int. (2018) 18:1025–30. doi: 10.1111/ggi.13293,

PubMed Abstract | Crossref Full Text | Google Scholar

36. Ormseth, CH, LaHue, SC, Oldham, MA, Josephson, SA, Whitaker, E, and Douglas, VC. Predisposing and precipitating factors associated with delirium: a systematic review. JAMA Netw Open. (2023) 6:e2249950. doi: 10.1001/jamanetworkopen.2022.49950

Crossref Full Text | Google Scholar

37. Mathew, G, Agha, R, Albrecht, J, Goel, P, Mukherjee, I, Pai, P, et al. STROCSS 2021: strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. Int J Surg. (2021) 96:106165. doi: 10.1016/j.ijsu.2021.106165,

PubMed Abstract | Crossref Full Text | Google Scholar

38. Guo, J, Di, J, Gao, X, Zha, J, Wang, X, Wang, Z, et al. Discriminative ability for adverse outcomes after hip fracture surgery: a comparison of three commonly used comorbidity-based indices. Gerontology. (2021) 68:62–74. doi: 10.1159/000515526,

PubMed Abstract | Crossref Full Text | Google Scholar

39. Griffiths, R, Babu, S, Dixon, P, Freeman, N, Hurford, D, Kelleher, E, et al. Guideline for the management of hip fractures 2020: guideline by the Association of Anaesthetists. Anaesthesia. (2021) 76:225–37. doi: 10.1111/anae.15291,

PubMed Abstract | Crossref Full Text | Google Scholar

40. Carson, JL, Guyatt, G, Heddle, NM, Grossman, BJ, Cohn, CS, Fung, MK, et al. Clinical practice guidelines from the AABB: red blood cell transfusion thresholds and storage. JAMA. (2016) 316:2025–35. doi: 10.1001/jama.2016.9185,

PubMed Abstract | Crossref Full Text | Google Scholar

41. Carson, JL, Stanworth, SJ, Roubinian, N, Fergusson, DA, Triulzi, D, Doree, C, et al. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev. (2016) 2016:CD002042. doi: 10.1002/14651858.CD002042.pub4,

PubMed Abstract | Crossref Full Text | Google Scholar

42. Carson, JL, Terrin, ML, Noveck, H, Sanders, DW, Chaitman, BR, Rhoads, GG, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. (2011) 365:2453–62. doi: 10.1056/NEJMoa1012452,

PubMed Abstract | Crossref Full Text | Google Scholar

43. Foss, NB, Kristensen, MT, Jensen, PS, Palm, H, Krasheninnikoff, M, and Kehlet, H. The effects of liberal versus restrictive transfusion thresholds on ambulation after hip fracture surgery. Transfusion. (2009) 49:227–34. doi: 10.1111/j.1537-2995.2008.01967.x,

PubMed Abstract | Crossref Full Text | Google Scholar

44. Rawlings, AM, Sharrett, AR, Albert, MS, Coresh, J, Windham, BG, Power, MC, et al. The Association of Late-Life Diabetes Status and Hyperglycemia with Incident Mild Cognitive Impairment and dementia: the ARIC study. Diabetes Care. (2019) 42:1248–54. doi: 10.2337/dc19-0120,

PubMed Abstract | Crossref Full Text | Google Scholar

45. Desquilbet, L, and Mariotti, F. Dose-response analyses using restricted cubic spline functions in public health research. Stat Med. (2010) 29:1037–57. doi: 10.1002/sim.3841,

PubMed Abstract | Crossref Full Text | Google Scholar

46. Jeong, SW, Kim, SH, Kang, SH, Kim, HJ, Yoon, CH, Youn, TJ, et al. Mortality reduction with physical activity in patients with and without cardiovascular disease. Eur Heart J. (2019) 40:3547–55. doi: 10.1093/eurheartj/ehz564,

PubMed Abstract | Crossref Full Text | Google Scholar

47. Yap, WK, Shih, MC, Kuo, C, Pai, PC, Chou, WC, Chang, KP, et al. Development and validation of a nomogram for assessing survival in patients with metastatic lung Cancer referred for radiotherapy for bone metastases. JAMA Netw Open. (2018) 1:e183242. doi: 10.1001/jamanetworkopen.2018.3242,

PubMed Abstract | Crossref Full Text | Google Scholar

48. Marcantonio, ER. Postoperative delirium: a 76-year-old woman with delirium following surgery. JAMA. (2012) 308:73–81. doi: 10.1001/jama.2012.6857,

PubMed Abstract | Crossref Full Text | Google Scholar

49. Choi, YH, Kim, DH, Kim, TY, Lim, TW, Kim, SW, and Yoo, JH. Early postoperative delirium after hemiarthroplasty in elderly patients aged over 70 years with displaced femoral neck fracture. Clin Interv Aging. (2017) 12:1835–42. doi: 10.2147/CIA.S147585,

PubMed Abstract | Crossref Full Text | Google Scholar

50. Sahadevan, S, Saw, SM, Gao, W, Tan, LC, Chin, JJ, Hong, CY, et al. Ethnic differences in Singapore's dementia prevalence: the stroke, Parkinson's disease, epilepsy, and dementia in Singapore study. J Am Geriatr Soc. (2008) 56:2061–8. doi: 10.1111/j.1532-5415.2008.01992.x,

PubMed Abstract | Crossref Full Text | Google Scholar

51. Melamud, MM, Bobrik, DV, Brit, PI, Efremov, IS, Buneva, VN, Nevinsky, GA, et al. Biochemical, hematological, inflammatory, and gut permeability biomarkers in patients with alcohol withdrawal syndrome with and without delirium tremens. J Clin Med. (2024) 13:102776. doi: 10.3390/jcm13102776,

PubMed Abstract | Crossref Full Text | Google Scholar

52. Mallat, J, Rahman, N, Hamed, F, Hernandez, G, and Fischer, MO. Pathophysiology, mechanisms, and managements of tissue hypoxia. Anaesth Crit Care Pain Med. (2022) 41:101087. doi: 10.1016/j.accpm.2022.101087,

PubMed Abstract | Crossref Full Text | Google Scholar

53. Snyder, B, Simone, SM, Giovannetti, T, and Floyd, TF. Cerebral hypoxia: its role in age-related chronic and acute cognitive dysfunction. Anesth Analg. (2021) 132:1502–13. doi: 10.1213/ANE.0000000000005525,

PubMed Abstract | Crossref Full Text | Google Scholar

54. Onder, G, Penninx, BWJH, Cesari, M, Bandinelli, S, Lauretani, F, Bartali, B, et al. Anemia is associated with depression in older adults: results from the InCHIANTI study. J Gerontol A Biol Sci Med Sci. (2005) 60:1168–72. doi: 10.1093/gerona/60.9.1168,

PubMed Abstract | Crossref Full Text | Google Scholar

55. Dong, B, Wang, J, Li, P, Li, J, Liu, M, and Zhang, H. The impact of preoperative malnutrition on postoperative delirium: a systematic review and meta-analysis. Perioper Med. (2023) 12:55. doi: 10.1186/s13741-023-00345-9,

PubMed Abstract | Crossref Full Text | Google Scholar

56. Han, TS, Lisk, R, Osmani, A, Sharmin, R, El Gammel, S, Yeong, K, et al. Increased association with malnutrition and malnourishment in older adults admitted with hip fractures who have cognitive impairment and delirium, as assessed by 4AT. Nutr Clin Pract. (2021) 36:1053–8. doi: 10.1002/ncp.10614,

PubMed Abstract | Crossref Full Text | Google Scholar

57. Chiavarini, M, Ricciotti, GM, Genga, A, Faggi, MI, Rinaldi, A, Toscano, OD, et al. Malnutrition-related health outcomes in older adults with hip fractures: a systematic review and Meta-analysis. Nutrients. (2024) 16:1069. doi: 10.3390/nu16071069,

PubMed Abstract | Crossref Full Text | Google Scholar

58. Inouye, SK, Westendorp, RG, and Saczynski, JS. Delirium in elderly people. Lancet. (2014) 383:911–22. doi: 10.1016/S0140-6736(13)60688-1,

PubMed Abstract | Crossref Full Text | Google Scholar

59. Burtscher, J, Mallet, RT, Burtscher, M, and Millet, GP. Hypoxia and brain aging: neurodegeneration or neuroprotection? Ageing Res Rev. (2021) 68:101343. doi: 10.1016/j.arr.2021.101343,

PubMed Abstract | Crossref Full Text | Google Scholar

60. Hoffman, WE, Albrecht, RF, and Miletich, DJ. Cerebrovascular response to hypoxia in young vs aged rats. Stroke. (1984) 15:129–33. doi: 10.1161/01.STR.15.1.129,

PubMed Abstract | Crossref Full Text | Google Scholar

61. Weiskopf, RB, Kramer, JH, Viele, M, Neumann, M, Feiner, JR, Watson, JJ, et al. Acute severe isovolemic anemia impairs cognitive function and memory in humans. Anesthesiology. (2000) 92:1646–52. doi: 10.1097/00000542-200006000-00023,

PubMed Abstract | Crossref Full Text | Google Scholar