Blood Transfusion for Elderly Patients with Hip Fracture: a Nationwide Cohort Study

Development and validation of machine learning models to predict perioperative transfusion risk for hip fractures in the elderly

BACKGROUND

Patients with hip fractures frequently need to receive perioperative transfusions of concentrated red blood cells due to preoperative anemia or surgical blood loss. However, the use of perioperative blood products increases the risk of adverse events, and the shortage of blood products is prompting us to minimize blood transfusion. Our study aimed to construct a machine learning algorithm predictive model to identify patients at high risk for perioperative transfusion early in hospital admission and to manage their patient blood to reduce transfusion requirements.

METHODS

This study collected patients hospitalized for hip fractures at a university hospital from May 2016 to November 2022. All patients included in the analysis were randomly divided into a training set and validation set according to 70:30. Eight machine learning algorithms, CART, GBM, KNN, LR, NNet, RF, SVM, and XGBoost, were used to construct the prediction models. The models were evaluated for discrimination, calibration, and clinical utility, and the best prediction model was selected.

RESULTS

A total of 805 patients were included in the study, of whom 306 received transfusions during the perioperative period. We screened eight features used to construct the prediction model: age, fracture time, fracture type, hemoglobin, albumin, creatinine, calcium ion, and activated partial thromboplastin time. After evaluating and comparing the performance of each of the eight models, the model constructed by the XGBoost algorithm had the best performance, with MCC values of 0.828 and 0.939 in the training and validation sets, respectively. In addition, it had good calibration and clinical utility in both the training and validation sets.

CONCLUSION

The model constructed by the XGBoost algorithm has the best performance, using this model to identify patients at high risk for transfusion early in their admission and promptly incorporating them into a patient blood management plan can help reduce the risk of transfusion.

Alternative blood transfusion triggers: a narrative review

BACKGROUND

Anemia, characterized by low hemoglobin levels, is a global public health concern. Anemia is an independent factor worsening outcomes in various patient groups. Blood transfusion has been the traditional treatment for anemia; its triggers, primarily based on hemoglobin levels; however, hemoglobin level is not always an ideal trigger for blood transfusion. Additionally, blood transfusion worsens clinical outcomes in certain patient groups. This narrative review explores alternative triggers for red blood cell transfusion and their physiological basis.

MAIN TEXT

The review delves into the physiology of oxygen transport and highlights the limitations of using hemoglobin levels alone as transfusion trigger. The main aim of blood transfusion is to optimize oxygen delivery, necessitating an individualized approach based on clinical signs of anemia and the balance between oxygen delivery and consumption, reflected by the oxygen extraction rate. The narrative review covers different alternative triggers. It presents insights into their diagnostic value and clinical applications, emphasizing the need for personalized transfusion strategies.

CONCLUSION

Anemia and blood transfusion are significant factors affecting patient outcomes. While restrictive transfusion strategies are widely recommended, they may not account for the nuances of specific patient populations. The search for alternative transfusion triggers is essential to tailor transfusion therapy effectively, especially in patients with comorbidities or unique clinical profiles. Investigating alternative triggers not only enhances patient care by identifying more precise indicators but also minimizes transfusion-related risks, optimizes blood product utilization, and ensures availability when needed. Personalized transfusion strategies based on alternative triggers hold the potential to improve outcomes in various clinical scenarios, addressing anemia's complex challenges in healthcare. Further research and evidence are needed to refine these alternative triggers and guide their implementation in clinical practice.

Explainable artificial intelligence for predicting red blood cell transfusion in geriatric patients undergoing hip arthroplasty: Machine learning analysis using national health insurance data

This study uses machine learning and population data to analyze major determinants of blood transfusion among patients with hip arthroplasty. Retrospective cohort data came from Korea National Health Insurance Service claims data for 19,110 patients aged 65 years or more with hip arthroplasty in 2019. The dependent variable was blood transfusion (yes vs no) in 2019 and its 31 predictors were included. Random forest variable importance and Shapley Additive Explanations were used for identifying major predictors and the directions of their associations with blood transfusion. The random forest registered the area under the curve of 73.6%. Based on random forest variable importance, the top-10 predictors were anemia (0.25), tranexamic acid (0.17), age (0.16), socioeconomic status (0.05), spinal anesthesia (0.05), general anesthesia (0.04), sex (female) (0.04), dementia (0.03), iron (0.02), and congestive heart failure (0.02). These predictors were followed by their top-20 counterparts including cardiovascular disease, statin, chronic obstructive pulmonary disease, diabetes mellitus, chronic kidney disease, peripheral vascular disease, liver disease, solid tumor, myocardial infarction and hypertension. In terms of max Shapley Additive Explanations values, these associations were positive, e.g., anemia (0.09), tranexamic acid (0.07), age (0.09), socioeconomic status (0.05), spinal anesthesia (0.05), general anesthesia (0.04), sex (female) (0.02), dementia (0.03), iron (0.04), and congestive heart failure (0.03). For example, the inclusion of anemia, age, tranexamic acid or spinal anesthesia into the random forest will increase the probability of blood transfusion among patients with hip arthroplasty by 9%, 7%, 9% or 5%. Machine learning is an effective prediction model for blood transfusion among patients with hip arthroplasty. The high-risk group with anemia, age and comorbid conditions need to be treated with tranexamic acid, iron and/or other appropriate interventions.

Complication Rates and Survival of Nonagenarians after Hip Hemiarthroplasty versus Proximal Femoral Nail Antirotation for Intertrochanteric Fractures: A 15-Year Retrospective Cohort Study of 113 Cases

OBJECTIVE

Intertrochanteric fracture is a very common but serious type of hip fracture in nonagenarians. The surgical treatment remains a significant challenge for orthopedists. The objective of this study was to investigate postoperative complications and survival outcomes compared between bipolar hemiarthroplasty (HA) and proximal femoral nail anti-rotation (PFNA) in nonagenarians with intertrochanteric fractures, and to evaluate the efficacy and safety of the two surgical procedures in this patient population.

METHODS

A total of 113 consecutive nonagenarians who underwent bipolar HA or PFNA for the treatment of intertrochanteric fractures from January 2006 to August 2021 were retrospectively studied in the current paper. There were 34 males and 79 females, with a mean age of 92.2 years (range 90-101 years) at the time of operation. The average duration of follow-up was 29.7 months (range 1-120 months). The full cohort was divided into bipolar HA (77 cases) and PFNA (36 cases) groups. Damage control orthopedics was used to determine the optimal surgery time and assist in perioperative management. A restrictive blood transfusion strategy was employed, along with appropriate adjustments under multidisciplinary assessment, throughout the perioperative period. Perioperative clinical information and prognostic data were analyzed. Kaplan-Meier survival curves were used for survival analysis, and landmark analysis divided the entire follow-up period into 1-12 months (short-term), 13-42 months (medium-term) and 43-120 months (long-term) according to the configurations of Kaplan-Meier survival curves.

RESULTS

Both groups had similar general variables except for the proportion of high adjusted Charlson comorbidity index (aCCI) (≥6 points) (6.5% in bipolar HA group and 22.2% in PFNA group, p = 0.024). Intraoperative blood loss and transfusion requirements were greater, and the intraoperative transfusion rates were higher in the bipolar HA group compared to the PFNA group (all p < 0.05). The complications rates, 1- to 60-month cumulative all-cause mortality, postoperative optimal Harris hip score (HHS), and Barthel index (BI) presented no significant difference between the two groups (all p > 0.05). Both groups had similar overall survival curves (p = 0.37). However, landmark analysis revealed that bipolar HA group exhibited higher survival rates in medium-term (p = 0.01), while similar survival rates were observed in the short- and long-term post-operation periods (both p > 0.05). Cox regression with survival-time-dependent covariate calculated the hazard ratio (HR) of bipolar HA was 0.41 in medium-term (p = 0.039).

CONCLUSION

Bipolar HA is equally effective and reliable as PFNA for treating intertrochanteric fractures in nonagenarians. Despite resulting in more intraoperative blood loss and transfusions, bipolar HA therapy is associated with a higher medium-term survival rate compared to PFNA treatment. The application of damage control orthopedics and precise perioperative patient blood management could contribute to the positive clinical outcomes observed in this patient population.

Influencing factors on morbidity and mortality in intertrochanteric fractures

We aimed to evaluate the effect of the patient's clinical and paraclinical condition before and after surgery on short-term mortality and complication and long-term mortality. A retrospective cohort study was conducted and multivariate logistic regression was applied to determine the effect of demographic characteristics (sex, age, AO/OTA classification, height, weight, body mass index), medical history (hypertension, ischemic heart disease, diabetes mellitus, thyroid malfunction, cancer, osteoporosis, smoking) lab data (Complete blood cell, blood sugar, Blood Urea Nitrogen, Creatinine, Na, and K), surgery-related factors (Anesthesia time and type, implant, intraoperative blood transfusion, postoperative blood transfusion, and operation time), duration of admission to surgery and anticoagulant consumption on short-term mortality and complication and long-term mortality. Three hundred ten patients from November 2016 to September 2020 were diagnosed with an intertrochanteric fracture. 3.23% of patients died in hospital, 14.1% of patients confronted in-hospital complications, and 38.3% died after discharge till the study endpoint. ΔNumber of Neutrophiles is the primary determinant for in-hospital mortality in multivariate analysis. Age and blood transfusion are the main determinants of long-term mortality, and Na before surgery is the primary variable associated with postoperative complications. Among different analytical factors Na before surgery as a biomarker presenting dehydration was the main prognostic factor for in hospital complications. In hospital mortality was mainly because of infection and long-term mortality was associated with blood transfusion.

Effect of Peri-Operative Blood Transfusion on Short and Long-Term Mortality Rates in Elderly Patients With Neck of Femur Fractures: A Retrospective Study

Introduction  The current reported mortality rate for elderly neck of femur fractures (eNOFF) is relatively high in the UK. eNOFF patients commonly suffer from associated cardiovascular co-morbidities and tend to have fragile physiological states and poor physiological reserves. Although some studies have shown a potential link between blood transfusion and mortality in eNOFF patients, there is no general consensus on this matter. Therefore, our study aims to explore the possible association between blood transfusion and length of hospital stay (LOHS) as well as short- and long-term mortality rates in eNOFF patients by reviewing the practice of blood transfusion. Methods  This retrospective study was conducted at Wrexham Maelor Hospital, which is part of the Betsi Cadwaladr University Health Board (BCUHB), Wales. The study included patients who were 65 years of age or older and presented with neck of femur fractures. Only patients who required surgical intervention were included, and those managed non-operatively were excluded from the study. The statistical analysis was performed using IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp., Armonk, New York, United States). Furthermore, unpaired t-tests and log-rank (Mantel-Cox) tests were performed to compare the groups that received blood transfusions.  Results  During the study period, a total of 501 eNOFF patients were included in the primary cohort of the study, with a mean age of 81 years (ranging from 65 to 102). The majority of the patients were female (n=340). Of the 501 patients, 79 (15.8%) received a blood transfusion during their treatment. Around 52.9% of the eNOFF patients were categorized as American Society of Anesthesiologists (ASA) III, but there was no statistically significant difference in the requirement of blood transfusion between patients in ASA III, II, and IV categories, as compared to ASA I. Additionally, the mean time to surgery was higher in patients who received a blood transfusion (35.8 hours), and this difference was statistically significant (p=0.035). Moreover, the average LOHS after surgery for eNOFF was longer in patients who needed peri-operative blood transfusion (22 days), and this difference in the means was statistically significant (p=0.022). At the one-year post-surgery mark, mortality was higher in the transfused group (33%), and long-term five-year mortality rates were also higher in this group (63.2%).  Conclusion  Peri-operative blood transfusion may confer certain benefits in the management of eNOFF ptients. However, it should not be regarded as a panacea for improving long-term outcomes. The decision to administer blood transfusion must be made on a case-by-case basis, with careful assessment of individual clinical indications, and the potential risks and benefits taken into consideration. To achieve optimal clinical outcomes, close monitoring and follow-up of eNOFF patients, both in the short-term and long-term, are essential.

Administration of Prophylactic Enoxaparin on the Morning of Surgery Does Not Increase Risk of Blood Transfusion or Wound Drainage Following Internal Fixation of Geriatric Femur Fractures

BACKGROUND

Despite standard use of chemoprophylaxis, 30-day incidence of venous thromboembolism after geriatric, those older than 60 years, femur fracture surgery is reported to be up to 10%. Missing one dose of enoxaparin has been proven to increase the risk of developing venous thromboembolism. It is commonplace to hold preoperative chemoprophylaxis the morning of surgery because of concern for intraoperative bleeding or wound drainage. We sought to determine whether administration of prophylactic enoxaparin the morning of surgery resulted in an increased rate of blood transfusion or wound drainage in geriatric patients undergoing femur fracture treatment.

METHODS

We retrospectively reviewed patients older than 60 years who underwent internal fixation of an isolated femur fracture, including femoral neck, intertrochanteric, subtrochanteric, femoral shaft, and distal femur fractures, at a Level 1 trauma center. Medical records, hospital billing data, and radiographs were reviewed to determine patient characteristics such as Charlson Comorbidity Index, enoxaparin dosing, packed red blood cell transfusion, and persistent wound drainage, defined as any drainage requiring utilization of closed incision negative pressure wound therapy. Thirty-day mortality served as the secondary outcome measure.

RESULTS

Five hundred seven patients were included. One hundred sixty-four (32%) received enoxaparin on the morning of surgery, whereas 343 (68%) did not. 27% of patients received PRBC transfusion, and this did not differ between groups (27% vs. 28%, P = 0.72). Subgroup analysis of fixation strategies revealed no difference in the frequency of blood transfusion for any fixation type as related to the timing of enoxaparin dosage. Utilization of closed incision negative pressure wound therapy for the treatment of postoperative wound drainage did not differ between dosing groups. No difference was observed in thirty-day mortality between groups (2.4% vs. 2.7%, P = 0.9).

DISCUSSION

Administration of a prophylactic dose of enoxaparin on the morning of surgery does not seem to increase the rate of postoperative blood transfusion or wound drainage after fixation of geriatric femur fracture.

LEVEL OF EVIDENCE

Level III, therapeutic.

Effect of a Patient Blood Management Program on the Appropriateness of Red Blood Cell Transfusion and Clinical Outcomes in Elderly Patients Undergoing Hip Fracture Surgery

BACKGROUND

Elderly patients with hip fractures frequently receive perioperative transfusions, which are associated with increased morbidity and mortality. This study aimed to evaluate the impact of a patient blood management (PBM) program on the appropriateness of red blood cell (RBC) transfusion and clinical outcomes in geriatric patients undergoing hip fracture surgery.

METHODS

In 2018, the revised PBM program was implemented at the Korea University Anam Hospital, Seoul, Republic of Korea. Elderly patients aged ≥ 65 years who underwent hip fracture surgery from 2017 to 2020 were evaluated. Clinical characteristics and outcomes were analyzed according to the timing of PBM implementation (pre-PBM, early-PBM, and late-PBM). Multiveriate regression analysis was used to evaluate the risk factors of the adverse outcomes, such as in-hospital mortality or 30-day readmission.

RESULTS

A total of 884 elderly patients were included in this study. The proportion of patients who received perioperative RBC transfusions decreased significantly (43.5%, 40.1%, and 33.2% for pre-PBM, early-PBM, and late-PBM, respectively; P = 0.013). However, the appropriateness of RBC transfusion significantly increased (54.0%, 60.1%, and 94.7%, respectively; P < 0.001). The duration of in-hospital stay and 30-day readmission rates significantly decreased. Multivariable regression analysis revealed that RBC transfusion (odds ratio, 1.815; 95% confidence interval, 1.137-2.899; P = 0.013) was significantly associated with adverse outcomes.

CONCLUSION

Implementing the PBM program increased the appropriateness of RBC transfusion without compromising transfusion quality and clinical outcomes. Therefore, adopting the PBM program may improve the clinical management of elderly patients following hip fracture surgery.

Blood-Transfusion Risk Factors after Intramedullary Nailing for Extracapsular Femoral Neck Fracture in Elderly Patients

BACKGROUND

Extracapsular femoral neck fractures (eFNF) are the third most common type of fracture in traumatology. Intramedullary nailing (IMN) is one of the most frequently used ortho-pedic treatments for eFNF. Blood loss is one of the main complications of this treatment. This study aimed to identify and evaluate the perioperative risk factors that lead to blood transfusion in frail patients with eFNF who undergo IMN.

METHODS

From July 2020 to December 2020, 170 eFNF-affected patients who were treated with IMN were enrolled and divided into two groups according to blood transfusion: NBT (71 patients who did not need a blood transfusion), and BT (72 patients who needed blood transfusion). Gender, age, BMI, pre-operative hemoglobin levels, in-ternational normalized ratio (INR) level, number of blood units transfused, length of hospital stay, surgery duration, type of anesthesia, pre-operative ASA score, Charlson Comorbidity Index, and mortality rate were assessed.

RESULTS

Cohorts differed only for pre-operatively Hb and surgery time (p < 0.05).

CONCLUSION

Patients who have a lower preoperative Hb level and longer surgery time have a high blood-transfusion risk and should be closely followed peri-operatively.

Preoperative Correction of Low Hemoglobin Levels Can Reduce 1-Year All-Cause Mortality in Osteoporotic Hip Fracture Patients: A Retrospective Observational Study

Purpose

Osteoporotic hip fracture surgery is associated with a risk of morbidity and mortality. Admission hemoglobin levels <10 g/dL have been documented as a strong predictor of mortality risk. This study aimed to investigate the mortality outcome between osteoporotic hip fracture patients who had preoperative hemoglobin levels raised to ≥10 g/dL and those with hemoglobin levels were <10 g/dL.

Patients and Methods

This 5-year retrospective observational study included 226 participants with osteoporotic hip fractures that required surgery and who had admission hemoglobin levels <10 g/dL. Patients were categorized into two groups: those with corrected preoperative hemoglobin ≥10 g/dL and those with either corrected or uncorrected preoperative hemoglobin <10 g/dL. Outcomes were analyzed using Cox proportional hazard regression adjusted for confounders. Results are presented as hazard ratio (HR) and 95% confidence interval (95% CI).

Results

Among 226 the patients, the overall mortality rate was 17.25% (n=39/226) of the 226 patients, 93 (41.15%) had their hemoglobin levels raised to ≥10 g/dL by red blood cell transfusion. Multivariable analysis after adjustment for confounders showed a 50% lower incidence of mortality among patients with preoperative hemoglobin levels ≥10 g/dL than among those with hemoglobin levels <10 g/dL (HR 0.50, 95% CI (0.25–0.99), p=0.048).

Conclusion

In osteoporotic hip fracture patients with admission hemoglobin <10g/dL, raising preoperative hemoglobin levels to ≥10 g/dL can significantly reduce the risk of mortality. Testing for and correction of low preoperative hemoglobin levels is of value in hip surgery patients.

Postoperative hypoalbuminemia is an independent predictor of 1-year mortality after surgery for geriatric intertrochanteric femoral fracture: A retrospective cohort study

Preoperative hypoalbuminemia from malnutrition is associated with increased morbidity and mortality after geriatric hip fracture surgery. However, little is known regarding the correlation between postoperative hypoalbuminemia and mortality. This study aimed to evaluate whether postoperative hypoalbuminemia could predict 1-year mortality after intertrochanteric femoral fracture surgery in elderly patients.The medical records of 263 geriatric patients (age ≥65 years) who underwent intertrochanteric femoral fracture surgery between January 2013 and January 2016 in a single hospital were reviewed retrospectively. The patients were allocated to 2 groups based on lowest serum albumin levels within 2 postoperative days (≥3.0 g/dL [group 1, n = 46] and <3.0 g/dL [group 2, n = 217]. Data between the non-survival and survival groups were compared. Multivariable logistic regression analysis was conducted to identify the independent predictor for 1-year mortality.The 1-year mortality rate was 16.3% after intertrochanteric femoral fracture surgery. Multivariable logistic regression analysis revealed that postoperative hypoalbuminemia was significantly associated with 1-year mortality (adjusted odds ratio, 8.03; 95% confidence interval, 1.37-47.09; P = .021). The non-survival group showed a significantly increased incidence of postoperative hypoalbuminemia (95.4% vs 80.0%, P = .015) and intensive care unit admission (11.6% vs 2.7%, P = .020), older age (82.5 ± 5.8 years vs 80.0 ± 7.2 years, P = .032), lower body mass index (20.1 ± 3.2 kg/m2 vs 22.4 ± 3.8 kg/m2, P < .001), and increased amount of transfusion of perioperative red blood cells (1.79 ± 1.47 units vs 1.43 ± 2.08 units, P = .032), compared to the survival group.This study demonstrated that postoperative hypoalbuminemia is a potent predictor of 1-year mortality in geriatric patients undergoing intertrochanteric femoral fracture surgery. Therefore, exogenous albumin administration can be considered to improve postoperative outcomes and reduce the risk of mortality after surgery for geriatric hip fracture.

Establishment and Verification of a Perioperative Blood Transfusion Model After Posterior Lumbar Interbody Fusion: A Retrospective Study Based on Data From a Local Hospital

Objective: We aimed to analyze the related risk factors for blood transfusion and establish a blood transfusion risk model during the per-ioperative period of posterior lumbar interbody fusion (PLIF). It could provide a reference for clinical prevention and reduction of the risk of blood transfusion during the peri-operative period. Methods: We retrospectively analyzed 4,378 patients who underwent PLIF in our hospital. According to whether they were transfused blood or not, patients were divided into the non-blood transfusion group and the blood transfusion group. We collected variables of each patient, including age, sex, BMI, current medical history, past medical history, surgical indications, surgical information, and preoperative routine blood testing. We randomly divide the whole population into training group and test group according to the ratio of 4:1. We used the multivariate regression analyses get the independent predictors in the training set. The nomogram was established based on these independent predictors. Then, we used the AUC, calibration curve and DCA to evaluate the nomogram. Finally, we verified the performance of the nomogram in the validation set. Results: Three or more lumbar fusion segments, preoperative low hemoglobin, with hypertension, lower BMI, and elder people were risk factors for blood transfusion. For the training and validation sets, the AUCs of the nomogram were 0.881 (95% CI: 0.865-0.903) and 0.890 (95% CI: 0.773-0.905), respectively. The calibration curve shows that the nomogram is highly consistent with the actual observed results. The DCA shows that the nomogram has good clinical application value. The AUC of the nomogram is significantly larger than the AUCs of independent risk factors in the training and validation set. Conclusion: Three or more lumbar fusion segments, preoperative low hemoglobin, with hypertension, lower BMI, and elder people are associated with blood transfusion during the peri-operative period. Based on these factors, we established a blood transfusion nomogram and verified that it can be used to assess the risk of blood transfusion after PLIF. It could help clinicians to make clinical decisions and reduce the incidence of peri-operative blood transfusion.