Transfusion Risk Factors in Low-Energy Pelvic Fractures: A Retrospective Cohort Study

Abstract

Introduction

The management of low-energy pelvic fractures poses a multifaceted challenge in orthopedic and geriatric care. With an increasing prevalence of patients taking new oral anticoagulant medications, it is unclear, which patients with pelvis fractures will have a significant life threatening bleed. The aim of this study was to assess the main risk factors leading to blood transfusion and longer hospital stay in patients with low-energy pelvic fractures.

Materials and Methods

This single center retrospective observational study was performed with reference to STROBE guidelines. Demographic data, bleeding risk factors as well as blood transfusion rates were recorded. Initially, a univariate regression analysis was performed to analyze factors associated with blood transfusion. Significant variables were selected and a multidimensional predictive model was built based on variables that showed a significant relationship with the transfusion occurrence.

Results

In total 167 patients were included in the study, there were 24 male and 143 female patients. The number of patients requiring blood transfusion was 20 (12%). Following univariate analysis, in final multivariate regression model, the only variable factors predicting the need for blood transfusion were low hemoglobin level on first hospital assessment (9.85 g/dl in transfused group vs 12 g/dl in non-transfused group, OR: 0.364, P < 0.005). First control hemoglobin check after 24 hours was also a significant predictor of transfusion (8.45 g/dl in transfused group vs 11.20 in non-transfused group, OR: 0.347, P < 0.005). The area under curve analysis for ROC showed good predictive accuracy for blood transfusion in patients with low haemoglobin level on first assessment (AUC = 0.9) and for first control haemoglobin (AUC = 0.88).

Conclusions

We suggest that all patients following low energy pelvic fractures have a haemoglobin level assessment in the emergency department. Patients with initial haemoglobin <10 g/dl should be considered for hospital admission and repeat blood tests within the first 24 hours.

Keywords

low-energy fracture fragility fractures pelvis geriatric trauma geriatric medicine

Introduction

The management of low-energy pelvic fractures poses a multifaceted challenge in orthopedic and geriatric care, necessitating a nuanced understanding of factors influencing patient outcomes. Stable pelvic ring fractures are associated with a significant bleeding risk and the need to transfuse blood.¹ This study aims to identify main risk factors leading to blood transfusion and longer hospital stay in patients with low-energy pelvic fractures.

Low-energy pelvic ring fractures include anterior ring fractures, such as those of the superior and inferior rami and symphysis, and posterior ring injuries, such as those of the ilio-sacral region and sacrum.² The total number of fragility fractures and as a result, low-energy pelvic ring fractures is on the rise due to the increasing aging population.³

In the last decade, there has been a significant increase in the use of new direct oral anticoagulant medications (DOACs) as well as new antiplatelet medications, with fewer available reversible agents and increased bleeding risk following trauma.⁴ However, the literature is limited and the effects of these new medications on bleeding following pelvic fractures has not been reported.

Currently there are no clear guidelines and risk stratification systems aimed to identify patients, who are at risk of significant bleed requiring transfusion of blood products after a low energy pelvic ring fracture. Identification of bleeding risk factors in patients with low-energy pelvic fractures who may not require hospitalization without compromising outcomes, could lead to better tailored care to individual patient needs and to decreased burden on healthcare facilities.

Materials and Methods

A retrospective observational data collection was performed to determine the outcomes for all patients admitted with low-energy pelvic fractures and assess factors contributing to longer hospital stay. All data was collected from Department of Trauma, Orthopaedics and Musculoskeletal Cancer, Copernicus Memorial Hospital, Lodz, Poland, a trauma tertiary referral center, between January 2018 and January 2024. Institutional approval was granted prior to data collection. This study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Supplemental material).⁵

Records of 492 patients consecutively admitted with pelvis fractures within the study period were assessed. Adult patients (≥18 years) were included if they sustained a stable pelvic ring fracture defined as a result of falling from standing height or less.⁶ Stable pelvis fractures were defined as fractures of the pelvis ring, which did not compromise vertical and horizontal stability of the pelvis.⁷ Fracture locations included: anterior ring fractures, such as those of the superior and inferior rami and symphysis, and posterior ring injuries, such as those of the ilio-sacral region and sacrum.² Fractures were classified based on Young and Burges classification.⁸ We further classified fractures based on the number of fractured bones within the pelvic ring. Ipsilateral superior and inferior pubic rami were considered as a single fracture rather than two separate fractures. Exclusion criteria included high-energy trauma, unstable unstable vertically or horizontally pelvis ring fractures requiring surgical intervention, or cases with an Injury Severity Score (ISS) greater than 12. Patients were also excluded with incomplete medical records. Data collection was performed by two surgeons and eligibility criteria was determined using a screening log.

Patient demographics including sex, age and associated comorbidities, such as cardiovascular disease was collected. Current use of antiplatelet medication including; acetylsalicylic acid, clopidogrel or their concurrent use was recorded. Similarly, the use of anticoagulant medication such as warfarin, acenocumarol and direct oral anticoagulants DOACs was recorded. DOACs were defined as direct oral anticoagulants taken orally to prevent blood clot formation by directly inhibiting certain coagulation factors including thrombin (factor IIa) or factor Xa.⁹ These included: apixaban, rivaroxaban, dabigatrane and endoxaban. Pre-admission haemoglobin levels were recorded within the first hour of initial assessment in the accident and emergency department. Similarly, the first control haemoglobin levels were noted; usually performed 24 hours after admission to the hospital. Additionally, platelet count, APTT and INR levels were noted. Singh scale was used to assess the severity of osteoporosis on admission.¹⁰ Pre-admission osteoporosis treatment was also noted. Psoas area to third lumbar vertebra (PM:L3) ratio was used to assess low muscle mass as an indicator of sarcopenia presence and to determine if sarcopenia is a bleeding risk factor. PM:L3 ratio was measured as described in previously published research.^11,12 CT images and psoas major area was calculated in millimeters squared (Infinitt, Phillipsburg, NJ, USA). The primary objective was to assess blood transfusion rates in patients. Blood transfusion was defined as the administration of one unit of packed red blood cells. Transfusion of other non-blood products such as fresh frozen plasma were also assessed. Other patient outcomes including prolonged hospital stay (defined as >3 days) and other intrahospital complications were recorded. Direct comparisons between transfused and non-transfused grpups of patients were performed using the Wilcoxon test or Fisher’s exact test, depending on the nature of the data. All data was obtained from the protected online hospital records. No formal sample size calculation or power analysis was performed prior to this study. The sample size was determined based on the number of eligible patients available in the institutional records during the study period.

The primary measured outcome was the need for transfusion and the main goal of the statistical analysis was to identify clinically significant factors predicting transfusion. Our local indications for transfusion were anemia with a hemoglobin level <7.5 g/dL or <8.5 g/dL in patients with previous diagnosis of cardiovascular disease. Life-threatening anemia was defined as a hemoglobin level <6.5 g/dL. In total, 19 variables were initially considered as possible predictors of blood transfusion (number of fractured pelvic bones, patient characteristics, hemoglobin levels, coagulation markers, pre-admission osteoporosis treatment status, Singh index osteoporosis severity score and presence of medical comorbidities (Supplemental material). Medical comorbidities included: cardiovascular disease, arrythmias, hypertension, diabetes, active malignancy, dementia and chronic kidney disease. Cardiovascular disease was defined as conditions including: coronary artery disease (CAD), cerebrovascular disease (CVD), peripheral artery disease, and aortic atherosclerosis.¹³ Additionally, we included patients with clear sequelae of these conditions, as documented in their clinical history. Active malignancy was defined as an ongoing oncological condition requiring current active treatment. Both dementia was defined as a pre-admission diagnosis of cognitive decline, impacting memory, reasoning, and daily activities and chronic kidney disease as pre-admission chronic kidney dysfunction.

For the initial selection of variables, a univariate regression analysis was conducted, reducing the number of statistically significant variables from 19 to 6. In the next step, a multidimensional predictive model was developed based on these 6 variables, which demonstrated either a significant (P < 0.05) or marginally significant (P < 0.1) association with transfusion occurrence in the univariate analysis. Those variables included: anticoagulation medication use, high Singh index score, active cancer presence, first admission haemoglobin level, first control haemoglobin level and high APTT level. Following the logistic regression, the variables, which were not significant predictors in the initial model, were excluded from the final analysis. Eventually, the final model was built using two most important variables: first admission haemoglobin level and first control haemoglobin. Additionally, to the main analyses, pairwise comparisons were performed using the Wilcoxon test on the predictors from the initial multivariate model. A logistic regression model was constructed using the glm function (binomial family, logit link) to predict the binary outcome. Predictors were selected via univariate analysis, followed by multivariate modeling with stepwise removal of insignificant variables. Class weights addressed outcome imbalances. Assumptions of multicollinearity and linearity in the logit were verified through predictor plots and variance inflation factors (VIF). VIF analysis confirmed no multicollinearity violations (all VIF <10).

To address concerns regarding univariate analysis as a precondition for multivariate regression, we additionally applied stepwise regression (combining forward selection and backward elimination) for predictor selection. The optimal model identified included first admission and first control haemoglobin measurements, antiplatelet medication use, INR, hypertension, diabetes, arrhythmias, and active malignancy. While this model slightly improved performance (Nagelkerke’s R² = 0.934) compared to the simpler model using only haemoglobin measurements (Nagelkerke’s R² = 0.830), removing hemoglobin measurements significantly reduced predictive power (R² = 0.228). Given the risk of overfitting with a more complex model, the simpler model was used due to its strong predictive balance while maintaining robustness.

Additionally, the Area Under the Receiver Operating Characteristic Curve (AUC) was assessed for both variables: first admission haemoglobin level and first control haemoglobin. A separate regression analysis was performed to assess the effect of sarcopenia on need to transfuse blood on a smaller cohort of patients with available CT scans for assessment. Statistical analyses and data visualizations were performed using R packages (version 4.3.2; R Core Team, 2023).

Results

492 records of patients with pelvic fractures were reviewed and 325 removed after not reaching the inclusion criteria. In total, 167 patients were included in the study of which 24 were males and 143 were females. The mean age of male patients was 83 and the mean age of female patients was 74 years. All admitted patients sustained a fall from their own standing height. The number of patients taking anticoagulation medication and antiplatelet medication prior to hospital admission was 40 (24%) and 24 (14%) respectively. Of the admitted patients, 79 (47%) were osteoporotic. On admission the average first haemoglobin measurement was 11.9 (SD: 10.9-13.0) and the second haemoglobin count 24 hours after admission was 11.0 (SD: 9.9-12.1). Mean platelet count on admission was 211 (SD: 162-255), APTT 28 (SD: 26-32) and INR 1.09 (SD: 1.00-1.21). In total 61 (36%) patients had coexisting cardiovascular disease, 49 (29%) arrythmias, 82 (49%) hypertension, 23 (14%) diabetes, 10 (6%) active malignancy, 10 (6%) dementia, 6 (4%) end stage chronic kidney disease (Table 1).

Table 1.

Baseline Characteristics, Bleeding Risk Factors and Outcomes of Patients With and Without Bleeding Requiring Blood Transfusion During Hospitalization for Low-Energy Pelvic Fractures

	No transfusion	Transfusion	P - value
	N = 147	N = 20	P - value
Age (years)	82 (72, 87)	86 (77, 92)	0.091
Fracture characteristics
Number of pelvic bones fractured			0.796
1	78 (53%)	9 (45%)
2	44 (30%)	7 (35%)
3	25 (17%)	4 (20%)
Pre-admission osteoporosis treatment	67 (46%)	12 (60%)	0.243
Singh index on admission			0.079
2	8 (5.4%)	4 (20%)
3	58 (39%)	8 (40%)
4	71 (48%)	8 (40%)
5	10 (6.8%)	0 (0%)
Bleeding risk factors
Admission platelet level	211 (161, 255)	209 (171, 247)	0.079
Admission APTT	28 (26, 32)	33 (28, 37)	0.007
Admission INR	1.08 (1.00, 1.19)	1.15 (1.07, 1.40)	0.065
Antiplatelet medications	20 (14%)	4 (20%)	0.495
Anticoagulative medications	32 (22%)	8 (40%)	0.093
First assessment haemoglobin	12.00 (11.30, 13.20)	9.85 (8.40, 10.70	<0.005
Control haemoglobin	11.20 (10.20, 12.20)	8.45 (7.50, 10.05)	<0.005
Existing comorbidities
Cardiovascular disease	55 (37%)	10 (50%)	0.331
Arrythmias	40 (27%)	9 (45%)	0.119
Hypertension	71 (48%)	11 (55%)	0.638
Diabetes	20 (14%)	3 (15%)	0.741
Active malignancy	6 (4.1%)	4 (20%)	0.020
Dementia	9 (6.1%)	1 (5.0%)	1.000
Chronic kidney disease	5 (3.4%)	1 (5.0%)	0.546
Other outcomes
Short length of stay	53 (36%)	1 (5.0%)	<0.005
Other complications	9 (6.1%)	6 (30%)	<0.005

Note. If values are numerical, the median and first and third quartile are displayed. Otherwise, group size is included. Hospital short length of stay defined as < 3 days.

The number of patients requiring blood transfusion was 20 (12%). Only one patient required angioembolization. Patients who required blood transfusion were more likely to stay longer than 3 days when compared to those that did not receive blood products (36% vs 5%, P < 0.005). Similarly, the blood transfusion group had a proportionally higher rate of other complications compared to the non-transfusion group (30% vs 6.1%, P < 0.005).

In univariate regression analysis, statistically significant variables predictive of transfusion included: anticoagulation medication use (P = 0.088), high Singh index score (P = 0.087), active cancer presence (P = 0.018), first admission haemoglobin level (P < 0.005), first control haemoglobin level (P < 0.005) and high APTT level (P = 0.031). The remaining parameters did not reach marginally significant level (P < 0.1). In final multivariate regression model, the only variable factors predicting the need for blood transfusion were first assessment haemoglobin, with average haemoglobin count of 12 g/dl in non-transfused group and 9.85 g/dl in the transfused group (OR: 0.0364, P < 0.005). Similarly, first control haemoglobin level 24 hours after hospital admission was predictive of transfusion with 11.20 g/dl in non-transfused group and 8.45 g/dl in transfused group (OR: 0.347, P < 0.005) (Table 2). The area under curve analysis for ROC showed good predictive accuracy for blood transfusion in patients with low haemoglobin level on first assessment (AUC = 0.9) and for first control haemoglobin (AUC = 0.88) (Figures 1 and 2).

Table 2.

Multivariate Linear Regression Results for the Final Model

	Estimate	Odds ratio	95% CI	SE	Statistic	P-value
Intercept	19.217	-	-	4.101	4.686	2.790 × 10^-6
First assessment haemoglobin	−1.012	0.364	0.198-0.608	0.281	−3.605	3.126 × 10^-4
Control haemoglobin	−1.058	0.347	0.184-0.589	0.291	−3.640	2.731 × 10^-4

Note. the table presents the results of a multivariate linear regression analysis with selected predictor variables, where the dependent variable is the occurrence of transfusion.

Figure 1.

Receiver Operating Characteristic Curve for First Admission Haemoglobin Level

Figure 2.

Receiver Operating Characteristic Curve for Control Haemoglobin Level 24 Hours After Hospital Admission

The number of patients with prolonged hospital stay was 54 (32%). The number of patients with other intrahospital complications was 15 (9%). Those were made up of: 8 (5%) urinary sepsis, 2 (1%) pulmonary embolism, 2 (1%) pneumonia, 1 (0.5%) stroke, 1 (0.5%) decompensated heart failure and 1 (0.5%) severe electrolyte imbalance.

Muscle mass, a sarcopenia parameter was recorded in 84 patients who had a CT pelvis performed on the day of hospital assessment. Of those patients 70 were female and 14 were male. The mean PML3 was 0.543 (SD: 0.197). In separate regression analysis, low muscle mass was not predictive of the need to transfuse blood (P = 0.262).

Discussion

This study examined risk factors for blood transfusion following low-energy pelvic fractures. Among the 167 patients included, the findings indicate that low hemoglobin at initial assessment is the strongest predictor of transfusion risk. Additionally, the use of antiplatelet and anticoagulant medications, including direct oral anticoagulants (DOACs), did not appear to impact transfusion rates or the recovery process in these patients.

Our study confirms that low-energy pelvic fractures are a complex and prevalent issue in the geriatric population. Only one previous study assessed the bleeding risk in patients with low energy pelvic fractures, which reported that 9% of the study cohort required blood transfusions, as compared to 11% in our study.¹⁴ Sng et al., 2020 reports a mean haemoglobin drop of up to 1.4 g/dl following initial assessment of patients with low-energy pelvic fractures.¹⁴ On average our study demonstrates a drop in haemoglobin of 0.9 g/dl within the first 24 hours from initial assessment in the hospital emergency department. Our demographic data revealed that a significant proportion of admitted patients had existing comorbidities, with cardiovascular disease being the most common (61%). Majority of these patients required either antiplatelet or DOAC medications to manage the underlying atrial fibrillation or decrease the risk of a thrombo-embolic event. The usage of anticoagulants and antiplatelet medications in 24% and 14% of patients, respectively, is an important consideration in managing geriatric patients with pelvic fractures, especially as the aging population is expected to have increasing use of these medications.¹⁵ Despite recent recommendations suggesting early operative management of hip fractures in patients taking DOACs, no guidelines or studies exist describing the pharmacological management of these patients with low-energy pelvic fractures.¹⁶ The results suggest that although patients in the transfused group were more likely to be taking DOACs and antiplatelet medications, multivariate analysis did not identify these as statistically significant risk factors for significant bleeding after low-energy pelvic fractures. Therefore, their use alone should not be considered an indication for hospital admission. Orthopaedic literature suggests similar results in hip fracture patients. Yang et al, 2020 metanalysis results shows unnecessary delay to surgery in order to restore platelet function in patients with hip fractures receive antiplatelet therapy.¹⁷ Similarly Kolodyczuk et al, 2023 suggests that early surgery may be safe in patients with hip fractures taking DOACs despite theoretical risk of increased bleeding.¹⁸

Previous reports highlight that low energy pelvic fractures may require transfusion acutely within hospital admission. Our study is the first to report that the most significant factor leading to transfusion is low haemoglobin level on initial assessment. Low first assessment haemoglobin (OR: 0.364, P < 0.005) as well as control haemoglobin level 24 hours after admission (OR: 0.347, P < 0.005) were predictive of transfusion need in the final regression analysis. These results indicate that the most at risk patients are the ones with low starting haemoglobin level. They become symptomatic of anemia or drop to significantly low haemoglobin levels as a result of further injury from a pelvic fracture. This included patients with a hemoglobin level of <8.5 g/dL and associated symptoms such as lethargy, weakness, and tiredness, as well as likely secondary effects, including syncope, shortness of breath, and reduced exercise tolerance.¹⁹ Previous literature suggests that life-threatening bleeding may result from low-energy pelvic fractures in the elderly and these injuries should not be ignored.^20-23 However, on the contrary unnecessary admission of elderly patients may lead to increased risk of hospital acquired infections, morbidity, mortality and increased healthcare costs.²⁴ We suggest that all patients with low energy pelvic fractures have a haemoglobin level assessment in the emergency department. After the initial trauma assessment and exclusion of co-existing injuries or acute medical conditions requiring immediate intervention, patients with an initial haemoglobin level below 10 g/dL should be considered for hospital admission for observation and repeat blood tests within the first 24 hours. However, conservative treatment consisting of analgesia and early weight-bearing remains the mainstay of treatment in geriatric patients with pelvic fragility fractures.²⁵ Patients deemed suitable for conservative treatment at home may be advised to begin mobilization and partial weight-bearing once pain subsides, ideally under the guidance of a physiotherapist. However, the literature on this topic is limited, heterogeneous, and lacks clear guidelines.²⁶

Similarly, previous orthopaedic systematic review suggests that presence of sarcopenia is a risk factor for significant bleeding and blood transfusion following orthopaedic surgery.²⁷ Of the 167 patients, 84 had CT scans of the pelvis, which allowed for ileopsoas area assessment at the level of third lumbar vertebra. In a separate regression analysis, decreasing muscle mass, which suggests a likelihood of sarcopenia presence was not predictive of transfusion and prolonged hospitalization. Our results are however based on a cohort of patients receiving conservative treatment without orthopaedic surgery, which may explain the differing results from other orthopaedic research.

Previous literature suggests that increased number of fractured bones contributes to increased hemorrhage from cancellous bony surfaces.²⁸ Interestingly, increased number of fractured pelvic bones in low-energy pelvic fractures did not influence the bleeding risk. These results may be explained by the fact that low energy pelvic fractures tend to not injure pelvic venous complex and a natural tamponade is formed within the pelvic compartments.²⁹

Despite the strengths of this study, certain limitations should be acknowledged. The retrospective design limits causal interpretations. The data was gathered from a single institution, which may affect the generalizability of the results. A limitation of this study is also the lack of an a priori power analysis. As the sample size was limited by available patient records, the study may be underpowered to detect smaller effect sizes, and nonsignificant results should be interpreted with caution. Also, some patients with pelvic fractures may have been discharged without hospital admission, which our study has not taken into consideration and as a result may be a contributing bias. Future multicenter prospective studies could enhance the robustness of findings and provide further clarity on the variables influencing hospital stay durations.

Conclusion

In conclusion, this study highlights critical factors affecting outcomes in elderly patients with low-energy pelvic fractures. In this study, use of DOACs was not a statistically significant risk factor for significant bleed following low-energy pelvic fractures. Our results suggest that patients with low hemoglobin level on first hospital assessment are most at risk of life-threatening anemia, requiring blood transfusion and should be admitted for observation.

Supplemental Material

Supplemental Material - Transfusion Risk Factors in Low-Energy Pelvic Fractures: A Retrospective Cohort Study

Supplemental Material for Transfusion Risk Factors in Low-Energy Pelvic Fractures: A Retrospective Cohort Study by Hieronim Kołodziejczyk, Filip Brzeszczyński, and Oktawiusz Bończak in Geriatric Orthopaedic Surgery & Rehabilitation.

Footnotes

Acknowledgements

Dr Kamil Myszczynski was a contributor and participated in statistical analysis of this study.

ORCID iDs

Hieronim Kołodziejczyk

Filip Brzeszczyński

Ethical Considerations

The Ethics Committee of the Copernicus Memorial Hospital, Lodz, Poland waived the need for ethics approval and the need to obtain consent for the collection, analysis and publication of the retrospectively obtained and anonymized data for this non-interventional study.

Author Contributions

H.K. and O.B. concieved the manuscript idea. H.K., F.B. collected data and performed data analysis. All authors discussed the results and contributed to the final manuscript. All authors contributed to writing the manuscript. O.B. supervised the project.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Abernathy

Schroder

Bohn

Switzer

. Low-energy pelvic ring fractures: a care conundrum. Geriatr Orthop Surg Rehabil. 2021;12:2151459320985406. doi:10.1177/2151459320985406

Soles

GLS

Ferguson

. Fragility fractures of the pelvis. Curr Rev Musculoskelet Med. 2012;5(3):222-228. doi:10.1007/s12178-012-9128-9

Hill

Robinson

Keating

. Fractures of the pubic rami. Epidemiology and five-year survival. J Bone Joint Surg Br. 2001;83(8):1141-1144. doi:10.1302/0301-620x.83b8.11709

Navar

Kolkailah

Overton

, et al. Trends in oral anticoagulant use among 436 864 patients with atrial fibrillation in community practice, 2011 to 2020. J Am Heart Assoc. 2022;11(22):e026723. doi:10.1161/JAHA.122.026723

Strobe . Strobe. https://www.strobe-statement.org/. Accessed October 2, 2022.

Diamantopoulos

Rohde

Johnsrud

Skoie

Hochberg

Haugeberg

. The Epidemiology of low- and high-energy distal radius fracture in middle-aged and elderly men and women in southern Norway. PLoS One. 2012;7(8):e43367. doi:10.1371/journal.pone.0043367

Elsissy

Ruckle

LeBrun

Johnson

. Pelvic ring injuries: stable or not? J Am Acad Orthop Surg. 2024;32(3):99-107. doi:10.5435/JAAOS-D-23-00470

Alton

Gee

. Classifications in brief: Young and burgess classification of pelvic ring injuries. Clin Orthop Relat Res. 2014;472(8):2338-2342. doi:10.1007/s11999-014-3693-8

Definition of direct oral anticoagulant - NCI drug dictionary - NCI. 2011. https://www.cancer.gov/publications/dictionaries/cancer-drug/def/direct-oral-anticoagulant. Accessed February 12, 2025.

10.

Singh

Nagrath

Maini

. Changes in trabecular pattern of the upper end of the femur as an index of osteoporosis. J Bone Joint Surg Am. 1970;52(3):457-467.

11.

Simpson

Wilson

Vimalachandran

McNicol

Magee

. Sarcopenia estimation using psoas major enhances P-POSSUM mortality prediction in older patients undergoing emergency laparotomy: cross-sectional study. Eur J Trauma Emerg Surg. 2022;48(3):2003-2012. doi:10.1007/s00068-021-01669-1

12.

Simpson

Manu

Magee

Wilson

Moug

Vimalachandran

. Measuring sarcopenia on pre-operative CT in older adults undergoing emergency laparotomy: a comparison of three different calculations. Int J Colorectal Dis. 2020;35(6):1095-1102. doi:10.1007/s00384-020-03570-6

13.

Olvera Lopez

Ballard

Jan

. Cardiovascular disease. In: StatPearls. Treasure Island, FL: StatPearls Publishing; 2025. https://www.ncbi.nlm.nih.gov/books/NBK535419/. Accessed 20 February 2025.

14.

Sng

Gentle

Asadollahi

. Bleeding risk associated with hemodynamically stable low-energy pelvic fracture. Geriatr Orthop Surg Rehabil. 2020;11:2151459320911868. doi:10.1177/2151459320911868

15.

Bonanad

Formiga

Anguita

Petidier

Gullón

. Oral anticoagulant use and appropriateness in elderly patients with atrial fibrillation in complex clinical conditions: ACONVENIENCE study. J Clin Med. 2022;11(24):7423. doi:10.3390/jcm11247423

16.

Wall

Mitchell

Kent

. Review of perioperative outcomes and management of hip fracture patients on direct oral anticoagulants. EFORT Open Rev. 2023;8(7):561-571. doi:10.1530/EOR-22-0060

17.

Yang

Long

Kuang

Gao

Tao

. Is hip fracture surgery safe for patients on antiplatelet drugs and is it necessary to delay surgery? A systematic review and meta-analysis. J Orthop Surg Res. 2020;15(1):105. doi:10.1186/s13018-020-01624-7

18.

Kolodychuk

Godshaw

Nammour

Starring

Mautner

. Early hip fracture surgery is safe for patients on direct oral anticoagulants. OTA Int. 2023;6(2):e252. doi:10.1097/OI9.0000000000000252

19.

Turner

Parsi

Badireddy

. Anemia. In: StatPearls. Treasure Island, FL: StatPearls Publishing; 2025. https://www.ncbi.nlm.nih.gov/books/NBK499994/. Accessed 20 February 2025.

20.

Coupe

Patel

McVerry

Wynn-Jones

. Fatal haemorrhage following a low-energy fracture of the pubic ramus. J Bone Joint Surg Br. 2005;87-B(9):1275-1276. doi:10.1302/0301-620X.87B9.16696

21.

Garrido -Gómez

Pena -Rodríguez

Martín-Noguerol

Hernández-Cortes

. Corona mortis artery avulsion due to a stable pubic ramus fracture. Orthopedics (Thorofare, N J). 2012;35(1):e80-e82. doi:10.3928/01477447-20111122-25

22.

Krappinger

Zegg

Jeske

El Attal

Blauth

Rieger

. Hemorrhage after low-energy pelvic trauma. J Trauma Acute Care Surg. 2012;72(2):437-442. doi:10.1097/TA.0b013e31822ad41f

23.

Sandri

Regis

Bizzotto

. Delayed bleeding and pelvic haematoma after low-energy osteoporotic pubic rami fracture in a warfarin patient: an unusual cause of abdominal pain. Case Rep Emerg Med. 2014;2014:783268. doi:10.1155/2014/783268

24.

Naouri

Yordanov

Lapidus

Pelletier-Fleury

. Cost-effectiveness analysis of direct admission to acute geriatric unit versus admission after an emergency department visit for elderly patients. BMC Geriatr. 2023;23(1):283. doi:10.1186/s12877-023-03985-0

25.

Mears

Berry

. Outcomes of displaced and nondisplaced pelvic and sacral fractures in elderly adults. J Am Geriatr Soc. 2011;59(7):1309-1312. doi:10.1111/j.1532-5415.2011.03455.x

26.

Haveman

van de Wall

BJM

Rohner

, et al. Conservative or operative therapy in patients with a fragility fracture of the pelvis: study protocol for a prospective, randomized controlled trial. Trials. 2024;25(1):513. doi:10.1186/s13063-024-08350-z

27.

Brzeszczynski

Brzeszczynska

Duckworth

Murray

Simpson

AHRW

Hamilton

. The effect of sarcopenia on outcomes following orthopaedic surgery : a systematic review. Bone Joint Lett J. 2022;104-B(3):321-330. doi:10.1302/0301-620X.104B3.BJJ-2021-1052.R1

28.

Gibbs

Champaneria

Novak

Doree

Palmer

Estcourt

. Pharmacological interventions for the prevention of bleeding in people undergoing definitive fixation of hip, pelvic and long bone fractures: a systematic review and network meta‐analysis. Cochrane Database Syst Rev. 2019;2019(12):CD013499. doi:10.1002/14651858.CD013499

29.

Biffl

. Control of pelvic fracture–related hemorrhage. Surg Open Sci. 2022;8:23-26. doi:10.1016/j.sopen.2022.01.001

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB