Preoperative Indicators for 1-year Mortality in Elderly Individuals Following Hip Fracture Surgery Under A Multidisciplinary Team Co-Management Model: A Single-Centre Retrospective Observational Study

Abstract

Background

Elderly patients have an impaired functional state and multiple comorbidities, resulting in poor postoperative rehabilitation ability and high rates of disability and mortality. However, little evidence exists on mortality predictors for geriatric hip fractures within the context of the multidisciplinary team co-management model. This study aimed to investigate the incidence and explore preoperative indicators of 1-year mortality following hip fractures in the elderly under this model.

Methods

A total of 439 elderly patients (130 men and 309 women) surgically treated for hip fractures under the multidisciplinary team co-management model between January 2018 and June 2021were included. Data regarding demographics, health state-related variables, injury- and admission-related variables, and preoperative laboratory test results were collected from medical records. Univariate and multivariate logistic regression analyses were used to identify preoperative indicators for 1-year mortality.

Results

A total of 49 patients died within 1 year of hip fracture surgery between January 2018 and June 2021, with an accumulated mortality rate of 11.16%. In univariate analysis, 14 items were found to be significant. In the multivariable logistic regression model, age >85 years, body mass index <21.0 kg/m², time from injury to admission >9.5 h, preoperative haemoglobin <117 g/L, serum albumin <33.9 g/L, lactate dehydrogenase >292 U/L, and blood urea nitrogen >8.5 mmol/L were the independent preoperative indicators for 1-year mortality after surgery in elderly patients with hip fracture under the multidisciplinary team co-management model.

Conclusions

This study establishes a novel set of preoperative predictors for 1-year mortality in geriatric hip fracture patients managed under an MDT model, distinct from previous investigations focusing on postoperative interventions. The identified indicators enable early risk stratification, facilitating timely preoperative optimization. These findings underscore the prognostic value of integrating clinical and biochemical markers before surgery, warranting validation in multicenter prospective studies. Further prospective studies should be conducted to elucidate these associations and assess the effectiveness of targeted measures.

Keywords

hip fractures aged mortality preoperative predictors multidisciplinary care team MDT

Background

In recent years, hip fractures have become increasingly common as a result of population ageing,¹ with the number of annual patients worldwide estimated to be up to 6.26 million by 2050.² While early surgery is recommended as the optimal treatment for hip fractures, elderly patients have an impaired functional state and multiple comorbidities, resulting in poor postoperative rehabilitation ability, as well as high rates of disability and mortality.³ The cumulative 30-day and 1-year mortality rates for geriatric hip fracture patients have been up to 10% and 36%, respectively.⁴

Since 2016, a multidisciplinary team (MDT) co-management model for geriatric hip fracture patients has shown great value in our hospital’s Department of Orthopaedic Trauma, for identifying high-risk elderly patients and taking targeted preventive measures to reduce the risk of death.^5,6 Previous studies have reported a series of factors associated with 1-year mortality in these patients.^7,8 However, little evidence exists on mortality predictors for geriatric hip fractures within the context of the MDT co-management model.⁹ Thus, we designed this single-centre observational study to investigate the incidence and explore the preoperative indicators of 1-year mortality in elderly patients after hip fracture surgery to provide evidence for the optimisation of preoperative assessment in this model.

Materials and methods

Objective

The objective of this study was to analyze 1-year mortality rates and preoperative predictive factors in elderly patients undergoing hip fracture surgery, with the goal of improving preoperative evaluation protocols. This study followed the STROBE guidelines for reporting observational studies. This study was approved by the Institutional Review Board of our hospital. All data were anonymized to protect patient privacy. Primary outcome was all-cause mortality within 1 year postoperatively. Secondary outcomes included time-to-death and cause-specific mortality (Supplemental Table 1).

Patients and Groups

Patients aged >65 years with hip fractures admitted to our institution under the MDT co-management model from January 2018 to June 2021 were eligible for inclusion in this single-centre observational study. All data were collected anonymously to protect patient privacy. Patients who had non-femoral neck fractures or intertrochanteric fractures, older fractures (>21 days from initial injury), conservative treatment, surgical procedures other than hip arthroplasty or intramedullary nail internal fixation, lack of clinical data, and those lost to follow-up were excluded. (To minimize heterogeneity, we included only patients undergoing hip arthroplasty or intramedullary nail internal fixation. Alternative procedures (e.g., cannulated screws for undisplaced fractures or DHS for stable trochanteric fractures) were excluded, as these are typically indicated for lower-risk fractures with distinct postoperative outcomes.)Post-discharge patient follow-up was performed in the orthopaedic outpatient clinic or by phone. This was an observational study.

MDT Co-management Model

Led by the Department of Orthopaedic Trauma with 18 participating departments, MDTs were composed of 20 core members whose job title were chief physician, a hip fracture ward, and three fast tracks (emergency, operating and the postoperative intensive care unit (ICU)). According to medical history and quick physical examination, patient severity was classified into mild, moderate, severe, and critical.¹⁰ Once admitted, the conditions of severe and critical individuals were posted on a public communication platform for the MDTs, and related departments like cardiovascular department, interventional and vascular surgery, intensive care unit (ICU), quickly consulted to make comprehensive diagnoses and treatment plans for sufficient preoperative preparation and shortening of the preoperative waiting time (PWT), coordinated and arranged by orthopaedic trauma surgeons. Post-procedure, patients assessed as having a high risk for adverse events and unfit for the general ward were sent directly to the ICU and received custody treatment for a short period to get through the crisis. Following surgery, MDTs focused on early physical and functional recovery. Active steps based on interdisciplinary cooperation, including multi-modal analgesia, airway and pulmonary management, prophylactic anticoagulation regimen, maintenance of electrolyte balance, anaemia, malnutrition management, and dementia assessment or managements were taken to prevent postoperative complications and enhance recovery. Individualised hip function rehabilitation programs and encouragement of early functional exercise helped elderly individuals restore function and return to daily life.

In brief, the concept of whole-course perioperative MDT co-management under the guidance of enhanced recovery after surgery was followed.^5,6

The primary goal of the multidisciplinary team (MDT) was to implement comprehensive perioperative co-management guided by enhanced recovery after surgery (ERAS) principles. The main intervention population consisted of patients older than 65 years with hip fractures, especially those who were elder and had multiple complications. Sample size was determined by available cases during the study period. A post-hoc power analysis indicated 80% power to detect OR ≥2.0 for primary predictors (α = 0.05).

Data Collection

Demographic, health state-related variables, injury- and admission-related variables, and preoperative laboratory test results (the first results after injury) were collected (Table 1).

Table 1.

Relative Data Collection (I Have Revised and Updated the Formation of this Table According to the Reviewer’s Comment)

	Variables
Demographic	Age; Gender; BMI
Health state-related	ASA classification History of smoking History of fracture Comorbidities: hypertension, diabetes mellitus, cerebrovascular fibrillation, CHD, LEDVT Quantity of comorbidities
Injury-and admission-related	Fracture types: intertrochanteric fracture or femoral neck fracture; TIA; PWT; TIS
^aPreoperative laboratory test results	Routine blood test: RBC, HGB, Hct, MCH, MCHC, WBC, NEUT, NEUT%, LYM, PLT Biochemical test: Tb, Alb, LDH, BUN, Cr, ALT, AST Electrolyte test: Na⁺, K⁺, Ca²⁺, Cl^-Fibrinolytic function: PT, APTT, INR, FIB-C, FDP, ATIII, D-dimer

^aThe first results after admission.

BMI: body mass index, ASA: American Society of Anesthesiologists, CHD: coronary atherosclerotic heart disease, LEDVT: lower extremity deep venous thrombosis, TIA: time from injury to admission, PWT: preoperative waiting time, TIS: time from injury to surgery, RBC: red blood cell, HGB: hemoglobin, Hct: hematocrit, MCH: mean red blood cell hemoglobin content, MCHC: mean red blood cell hemoglobin concentration, WBC: white blood cell, NEUT: neutrophile granulocyte, LYM: lymphocyte, PLT: platelet, Tb: serum total protein, Alb: albumin, LDH: lactate dehydrogenase, BUN: blood urea nitrogen, Cr: creatinine, ALT: alanine aminotransferase, AST: aspartate aminotransferase, Na⁺: sodium ions, K⁺: potassium ions, Ca²⁺: calcium ions, Cl^-: chloride ion, PT: prothrombin time, APTT: activated partial thrombin time, INR: international normalized ratio, FIB-C: fibrinogen-C type, FDP: plasma fibrin degradation products, ATIII: antithrombin III.

Statistical Analysis

The Shapiro–Wilk test was used to test data normality for continuous variables of both groups. Student’s t-test was performed for data that satisfied normality, and non-parametric tests were applied for the analysis of non-normal data. Data that satisfied normality were presented as the mean and standard deviation ( ¯x±s), and non-normal data were presented as median and first and third quartiles [M (Q₁, Q₃)]. Receiver operating characteristic (ROC) analysis was performed for continuous variables with significant differences between groups to establish the best cut-off values by calculating sensitivity, specificity, and the Youden index. Continuous variables were stratified into categorical variables based on determined cut-off values, and the area under the ROC curve (AUC), 95% confidence interval (CI), cut-off value, and P value were recorded. The chi-square test (or Fisher’s exact test) was performed for all categorical variables which were expressed as numbers and percentages (n, %). Univariate analysis was performed for all categorical variables associated with mortality within 1 year after hip fracture surgery between the 2 groups. Categorical variables with a significant association with mortality in the univariate analysis were included in the multivariate logistic regression analysis to identify the independent preoperative indicators. One-year Kaplan–Meier survival curves were plotted to stratify patients according to the cut-off value of the independent preoperative indicators. Finally, a log-rank test was performed to examine the crude differences between the 2 strata. SPSS Statistics for Windows, version 26.0 (IBM Corp, Armonk, NY, USA) was used for data analysis. All P values were two-sided, and the level of significance was set at P < 0.05.

Results

Initially, 620 patients who were diagnosed with hip fracture and treated under the MDT co-management model were eligible, and 439 were ultimately included in the analysis (Figure 1).

Figure 1.

Flow Chart of the Research

A total of 49 (11.16%) patients who died within 1 year after surgery, including 17 femoral neck fractures and 32 intertrochanteric fractures, were assigned to the death group, and the remaining 390 (88.84%) to the survival group. The main reasons for death were shown in the table (Supplemental Table 1). Univariate analysis of continuous variables associated with mortality within 1 year after hip fracture surgery between the death and survival groups is shown in Table 2.The death group had a higher age; lower body mass index (BMI); longer time from injury to admission (TIA); lower levels of red blood cell (RBC), haemoglobin (HGB), haematocrit (Hct), serum total protein (Tb), serum albumin (Alb), calcium ions (Ca²⁺); and higher levels of lactate dehydrogenase (LDH), blood urea nitrogen (BUN), and creatinine (Cr) than the survival group.

Table 2.

Univariate Analyses of Continuous Variables Associated With Mortality Within One Year Following Hip Fracture Surgery Between the two Groups

Variables	Number (%) of patients in death group(n = 49)	Number (%) of patients in survival group(n = 390)	Statistics (t, Z)	P value
Age [years, M (Q₁, Q₃)]	87 (81, 92)	82 (76, 87)	−4.382	<0.001*
BMI [kg/m², M (Q₁, Q₃)]	21.0 (19.5, 24.5)	22.7(20.0, 25.0)	−2.028	0.043*
Preoperative comorbidities
Quantities [M (Q₁, Q₃)]	4 (3, 5)	4 (2, 5)	−1.670	0.095
TIA [h, M (Q₁, Q₃)]	22.0 (6.5, 66.0)	7.0 (4.0, 24.0)	−2.914	0.004*
PWT [h, M (Q₁, Q₃)]	45.4 (25.8, 76.9)	59.2 (40.5, 86.2)	−1.725	0.085
TIS [h, M (Q₁, Q₃)]	76.0 (48.5, 134.9)	70.3 (48.6, 111.3)	−0.584	0.559
Preoperative routine blood test
RBC (×10¹²/L, ¯x ± s)	3.28 ± 0.73	3.83 ± 0.64	5.706	<0.001*
HGB [g/L, M (Q₁, Q₃)]	107 (91, 116)	122 (108, 132)	−4.506	<0.001*
Hct [%, M (Q₁, Q₃)]	31.2 (26.2, 35.1)	35.5 (31.3, 38.8)	−4.368	<0.001*
MCH [pg, M (Q₁, Q₃)]	30.4 (29.8, 31.7)	30.5 (29.5, 31.6)	−0.333	0.739
MCHC [g/L, M (Q₁, Q₃)]	334.0 (329.5, 339.0)	332.0 (327.0, 338.0)	−1.734	0.083
WBC [×10⁹/L M (Q₁, Q₃)]	9.62 (6.95, 12.38)	9.25 (7.20, 11.35)	−0.532	0.595
NEUT [×10⁹/L M (Q₁, Q₃)]	7.98 (5.71, 11.18)	8.14 (5.98, 13.05)	−0.367	0.713
NEUT [%, M (Q₁, Q₃)]	85.00 (79.91, 87.95)	83.15 (76.98, 88.50)	−0.957	0.339
LYM [×10⁹/L, M (Q₁, Q₃)]	0.77 (0.61, 1.26)	0.86 (0.63, 1.17)	−0.839	0.402
PLT [×10⁹/L, M (Q₁, Q₃)]	172 (120, 225)	180.00 (150, 224)	−1.152	0.249
Preoperative biochemical test
Tb (g/L, ¯x ± s)	65.5 ± 8.3	69.2 ± 8.9	2.771	0.006*
Alb (g/L,¯x ± s)	35.7 ± 4.6	38.9 ± 4.4	4.747	<0.001*
LDH [U/L, M (Q₁, Q₃)]	303 (246, 390)	259 (220, 314)	−3.206	0.001*
BUN [mmol/L, M (Q₁, Q₃)]	8.9 (6.3, 11.6)	6.3 (5.2, 8.1)	−4.268	<0.001*
Cr [μmol/L, M (Q₁, Q₃)]	74 (61, 103)	64 (54, 80)	−2.782	0.005*
ALT [U/L, M (Q₁, Q₃)]	22 (18, 32)	23 (16, 24)	−0.093	0.926
AST [U/L, M (Q₁, Q₃)]	28 (24, 34)	27 (23, 34)	−0.667	0.505
Preoperative electrolyte test
Na⁺ [mmol/L, M (Q₁, Q₃)]	138.2 (134.6, 140.5)	138.4 (136.0, 140.5)	−0.710	0.478
K⁺ [mmol/L, M (Q₁, Q₃)]	3.70 (3.56, 4.09)	3.73 (3.45, 4.03)	−0.652	0.515
Ca²⁺ [mmol/L, M (Q₁, Q₃)]	2.08 (1.91, 2.25)	2.21 (2.08,2.31)	−3.373	0.001*
Cl^- [mmol/L, M (Q₁, Q₃)]	104.2 (100.2, 109.5)	104.2 (101.9, 106.5)	−0.498	0.619
Preoperative fibrinolytic function
PT [s, M (Q₁, Q₃)]	11.8 (11.5, 12.7)	11.8 (11.3, 12.4)	−1.643	0.100
APTT [s, M (Q₁, Q₃)]	28.3 (27.0, 31.3)	29.9 (27.7, 32.1)	−1.685	0.092
INR [M (Q₁, Q₃)]	1.10 (1.07, 1.19)	1.11 (1.05, 1.18)	--1.276	0.202
FIB-C [g/L, M (Q₁, Q₃)]	3.7 (3.0, 4.2)	3.6 (3.1, 4.3)	−0.444	0.657
FDP [mg/L, M (Q₁, Q₃)]	30.6 (11.0, 70.7)	29.7 (9.8, 71.6)	−0.004	0.997
ATIII [%, M (Q₁, Q₃)]	90 (79, 104)	90 (79, 99)	−0.541	0.589
D-dimer [μg/L, M (Q₁, Q₃)]	4755 (1559, 9107)	5472 (1301, 10654)	−0.164	0.869

Data presented as median (IQR) or mean ± SD. Two-tailed Mann-Whitney U/t-tests used for comparisons; *P < 0.05, exact P-values reported where significant.BMI: body mass index, TIA: time from injury to admission, PWT: preoperative waiting time, TIS: time from injury to surgery, RBC: red blood cell, HGB: hemoglobin, Hct: hematocrit, MCH: mean red blood cell hemoglobin content, MCHC: mean red blood cell hemoglobin concentration, WBC: white blood cell, NEUT: neutrophile granulocyte, LYM: lymphocyte, PLT: platelet, Tb: serum total protein, Alb: serum albumin, LDH: lactate dehydrogenase, BUN: blood urea nitrogen, Cr: creatinine, ALT: alanine aminotransferase, AST: aspartate aminotransferase, Na⁺: sodium ions, K⁺: potassium ions, Ca²⁺: calcium ions, Cl^-: chloride ion, PT: prothrombin time, APTT: activated partial thrombin time, INR: international normalized ratio, FIB-C: fibrinogen-C type, FDP: plasma fibrin degradation products, ATIII: antithrombin III.

Relevant results of the ROC analysis performed to determine cut-off values are presented in the table (Supplemental Table 2). The cut-off value of age was set as 85 years as this variable was generally an integer.

Results of the univariate analysis performed for categorical variables associated with mortality within 1 year after hip fracture surgery showed that 14 items were found to have significant differences between groups (Table 3).

Table 3.

Univariate Analyses of Categorical Variables Associated With Mortality Within One Year After Hip Fracture Surgery Between the two Groups

Variables	Number (%) of patients in death group(n = 49)	Number (%) of patients in survival group(n = 390)	χ ²	P value
Age (>85 years)	29 (59.2)	121 (31.0)	15.344	<0.001^*
Gender
Male	17 (34.7)	113 (29.0)	0.683	0.408
Female	32 (65.3)	277 (71.0)	0.683	0.408
BMI (<21.0 kg/m²)	27 (55.1)	133 (34.1)	8.287	0.004^*
ASA classification
Ⅰ/Ⅱ	19 (38.8)	202 (51.8)	2.952	0.086
Ⅲ/Ⅳ	30 (61.2)	188 (48.2)	2.952	0.086
Smoking history	0 (0)	9 (2.3)	-	0.606
Fracture history	9 (18.4)	90 (23.1)	0.553	0.457
Comorbidities
Hypertension	30 (61.2)	224 (57.4)	0.256	0.613
Diabetes mellitus	15 (30.6)	94 (24.1)	0.988	0.320
Cerebral fibrillation	18 (36.7)	150 (38.5)	0.055	0.815
CHD	11 (22.4)	62 (15.9)	1.348	0.246
LEDVT	10 (20.4)	65 (16.7)	0.430	0.512
Fracture type
Femoral neck fracture	17 (34.7)	202 (51.8)	5.092	0.024^*
Intertrochanteric fracture	32 (65.3)	188 (48.2)	5.092	0.024^*
TIA(>9.5 hours)	33 (67.3)	154 (39.5)	13.818	<0.001^*
Preoperative routine blood test
RBC (<3.09 × 10¹²/L)	23 (46.9)	44 (11.3)	42.795	<0.001^*
HGB (<117 g/L)	38 (77.6)	162 (41.5)	22.762	<0.001^*
Hct (<29.5 %)	23 (46.9)	64 (16.4)	25.531	<0.001^*
Preoperative biochemical test
Tb (<69.2 g/L)	34 (69.4)	189 (48.5)	8.559	0.003^*
Alb (<33.9 g/L)	20 (40.8)	51 (13.1)	24.707	<0.001^*
LDH (>292 U/L)	27 (55.1)	118 (30.8)	12.148	<0.001^*
BUN (>8.5 mmol/L)	28 (57.1)	81 (20.8)	30.857	<0.001^*
Cr (>73 μmol/L)	27 (55.1)	135 (34.6)	7.846	0.005^*
Preoperative electrolyte test
Ca²⁺ (<2.05 mmol/L)	24 (49.0)	74 (19.0)	22.602	<0.001^*

Data presented as median (IQR) or mean ± SD. Two-tailed Mann-Whitney U/t-tests used for comparisons; *P < 0.05, exact P-values reported where significant.-: Fisher’s exact probability test.

BMI: body mass index, ASA: American Society of Anesthesiologists, CHD: coronary atherosclerotic heart disease, LEDVT: lower extremity deep venous thrombosis, TIA: time from injury to admission, RBC: red blood cell, HGB: hemoglobin, Hct: hematocrit, Tb: serum total protein, Alb: serum albumin, LDH: lactate dehydrogenase, BUN: blood urea nitrogen, Cr: creatinine, Ca²⁺: calcium ions.

Multivariate logistic regression analysis (Table 4) identified 7 independent preoperative predictors for 1-year mortality after surgery in elderly patients with hip fracture treated under the MDT co-management model: age >85 years (OR = 2.49, 95%CI 1.25-4.95, P = 0.010), time from injury to admission >9.5 h (OR = 2.41, 95%CI 1.19-4.89, P = 0.015), BMI <21.0 kg/m² (OR = 2.10, 95%CI 1.05-4.18, P = 0.035), preoperative hemoglobin <117 g/L (OR = 2.60, 95%CI 1.17-5.79, P = 0.020), serum albumin <33.9 g/L (OR = 2.89, 95%CI 1.34-6.27, P = 0.007), lactate dehydrogenase >292 U/L (OR = 2.51, 95%CI 1.25-5.06, P = 0.010), and blood urea nitrogen >8.5 mmol/L (OR = 2.53, 95%CI 1.24-5.14, P = 0.011).

Table 4.

Multivariate Logistic Regression Analysis Result of Independent Risk Factors for Mortality Within One Year After Hip Fracture Surgery in Elderly Patients

Variables	β	S.E.	Wald	df	OR	95% CI	P value
Age (>85 years)	0.910	0.351	6.706	1	2.49	1.25-4.95	0.010^*
BMI (<21.0 kg/m²)	0.740	0.352	4.432	1	2.10	1.05-4.18	0.035^*
TIA (>9.5 h)	0.879	0.361	4.570	1	2.41	1.19-4.89	0.015^*
Preoperative HGB (<117 g/L)	0.955	0.409	5.446	1	2.60	1.17-5.79	0.020^*
Preoperative Alb (<33.9 g/L)	1.063	0.395	7.246	1	2.89	1.34-6.27	0.007^*
Preoperative LDH (>292 U/L)	0.921	0.357	6.636	1	2.51	1.25-5.06	0.010^*
Preoperative BUN (>8.5 mmol/L)	0.927	0.362	6.540	1	2.53	1.24-5.14	0.011^*

*P < 0.05, statistical significance. Variables with P < 0.1 in univariate analyses were included. Results are presented as adjusted odds ratios (OR) with 95% confidence intervals (CI).

BMI: body mass index, TIA: time from injury to admission, HGB: hemoglobin, Alb: serum albumin, LDH: lactate dehydrogenase, BUN: blood urea nitrogen.

One-year Kaplan–Meier survival curves for independent preoperative indicators are shown in the chart (Supplemental Chart). The survival rates were evaluated by log-rank test and statistically significant differences were detected in all of the independent preoperative indicators (P < 0.001).

Discussion

The concept of MDT co-management was first proposed in the 1980s.¹¹ A variety of models have been developed after years of clinical practice. A previous work indicated that compared with traditional models, this approach had resulted in a significant reduction in PWT, postoperative hospitalisation stays, length of stay, surgical duration, and proportion of patients receiving ICU consultation or transfer.⁶ However, there is little data on mortality predictors for geriatric hip fractures within the context of the MDT co-management model. Thus, this study attempted to determine the preoperative indicators of 1-year mortality after surgery in elderly patients with hip fractures and to further develop this model to improve the prognosis of the target cohort.

Wu et al. (2019) evaluated a multidisciplinary co-management program for elderly hip fracture patients at Beijing Jishuitan Hospital and found it improved care indicators such as reducing the time from admission to surgery and increasing osteoporosis assessment.¹² Fan et al. (2022) compared traditional and MDT care for elderly intertrochanteric fracture patients and demonstrated that the MDT model shortened surgery waiting time, hospital stay, and decreased postoperative complications.⁹ Özel et al. (2023) identified COPD, high RDW, and low albumin levels at admission as predictors of 1-year mortality after hip fracture surgery.⁸ In our study, the 1-year mortality rate after hip fracture surgery was relatively low, and the collaborative management model in our hospital likely played a role. However, to thoroughly understand how our collaborative management model impacts mortality and its long-term efficacy, further in-house research is crucial. Existing studies offer useful insights but may not suit our patient group and healthcare setting. We should conduct large-scale, long-term follow up research. This will help optimize the model, overcome limitations, and enhance patient outcomes, thereby validating the model’s role in reducing mortality and guiding hip fracture care. The incidence of mortality following hip fracture surgery within 1 year was significantly lower than that reported in previous studies, which may be associated with the MDT co-management model.¹³ Also, the lack of a control group limits causal inference. After univariate and multivariate analysis, age >85 years, TIA >9.5 h, BMI <21.0 kg/m², preoperative HGB <117 g/L, serum Alb <33.9 g/L, LDH >292 U/L, and BUN >8.5 mmol/L were independent preoperative indicators for mortality within 1 year after surgery in elderly patients with hip fracture under the MDT co-management model.

In this study, a cut-off value of 85 years was the best predictor of mortality, and patients older than 85 years had an average of 2.49 times increased mortality rate compared to patients younger than 85 years. In patients over 85 years old, the postoperative 1-year mortality rate was 19%, which was lower than the 33% reported in a recent study.¹⁴ Although it have been argued that it is age-related comorbidities and not age that is a risk factor for postoperative adverse events,¹⁵ the frailty, as well as poor resistance and resilience brought about by age, do deserve attention in the perioperative management period and during long-term rehabilitation.

In the present study, a significant association was identified between the time from injury to admission (TIA) and mortality, whereas no such correlation was observed for the time from injury to surgery (TIS). This finding, though intriguing, is not entirely unexpected. The TIA in the death group (22.0 (6.5, 66.0)) was substantially longer compared to that in the survival group (7.0 (4.0, 24.0)). Analyzing the data dispersion, the interquartile range of TIA in the death group reached 59.5. This implies that a notable proportion of patients exhibited a wide variance in admission times post-injury, potentially attributed to factors like remote injury sites and delayed emergency transfers. Conversely, the survival group demonstrated a relatively minor variance, with an interquartile range of merely 20.0. This strongly indicates that the marked fluctuation in TIA could be a critical determinant contributing to the elevated mortality risk. Simultaneously, the TIS values of the 2 groups were relatively close, with 76 (48.5, 134.9) in the death group and 70.3 (48.6, 111.3) in the survival group. The mean TIA in the death group approximated a full day. During this period, patients received no medical intervention and were likely in a state of suboptimal resuscitation and continuous bed rest, thereby exacerbating the risks of pneumonia, venous thromboembolism, and pressure ulcers. Moreover, the wider interquartile range of TIA in the death group suggested that some patients were admitted 2-3 days after injury, which is undoubtedly linked to mortality.

The optimal timing of surgery remains a subject of debate. Leer et al. posited that performing surgery within 48 hours following a fracture could mitigate mortality and intraoperative complications.¹⁶ The findings of this study are, to some extent, congruent with this perspective. A shorter TIA might enhance the likelihood of patients undergoing surgery within 48 hours post-fracture, thereby reducing the mortality risk. In contrast, Castellanos et al. found no conclusive evidence suggesting that delayed surgery increases the incidence of mortality or general postoperative complications.¹⁷ The discrepancy from the results of this study might be ascribed to differences in the patient populations under investigation, including variations in comorbidities and injury severity. Keohane et al. advocated for a 36-48-hour time frame for TIS.¹⁸ In this study, the TIS of both groups fluctuated within this time window, which could potentially account for the lack of a significant correlation between TIS and mortality. However, for patients who have undergone appropriate medical optimization, the benefits of early surgery are well-established, supported by robust evidence.

Based on the outcomes of this study, future research should focus on multi-center, large-scale investigations to further elucidate the impact of TIA and TIS on the mortality of hip fracture patients across diverse regions and under varying medical resource scenarios. Additionally, endeavors should be directed towards exploring strategies to optimize the emergency transfer protocols and the preoperative assessment mechanisms of the multidisciplinary team (MDT). By minimizing TIA and accurately determining the optimal surgical timing, it is possible to effectively reduce the mortality rate among elderly patients with hip fractures. These data necessitate a more in-depth exploration. In our study, we initially postulated the existence of an “obesity paradox” within our sample.¹⁹ However, a more meticulous examination of the data reveals that this postulation demands further elucidation. The median body mass index (BMI) values of the death group (21.0 (19.5, 24.5)) and the survival group (22.7 (20.0, 25.0)) both fell within the normal range (Table 2). The receiver operating characteristic (ROC) analysis classified patients into low-BMI and normal-BMI groups. Prima facie, these data do not seem to corroborate the “obesity paradox” as previously discussed.

Indeed, a lower BMI in elderly patients with hip fractures is frequently associated with malnutrition and sarcopenia, which are well-recognized risk factors for elevated mortality.^20,21 Our data strongly indicate that elderly patients with a low BMI are at a heightened risk of death. This finding aligns with prior research demonstrating that hip fracture patients with low body weight have increased mortality rates and higher perioperative transfusion requirements.^22-24 Although a recent meta-analysis¹⁹ has shown an “obesity paradox” where overweight hip fracture patients exhibit a reduced late-mortality (≥1-year) risk compared to those with a normal BMI, it is crucial to note that this paradox may not be applicable to all patient populations.

Notably, although BMI was a statistically significant variable for 1-year mortality in our study, its clinical utility might be restricted. The ROC-derived AUC for BMI was <0.60, indicating low discriminative ability. Thus, BMI alone is not a highly reliable predictor of 1-year mortality in elderly hip fracture patients. Given that other unaccounted factors may also influence mortality risk, BMI should be interpreted cautiously within the overall patient evaluation.

In our current study, given that no participants with a BMI exceeding 30 kg/m² succumbed within 1 year post-operation, the data distribution might have skewed our initial interpretation. Future investigations with larger sample sizes and more heterogeneous patient populations are imperative to precisely evaluate the relationship between BMI and mortality in elderly hip-fracture patients. Such studies will facilitate the clarification of whether the “obesity paradox” truly exists in this particular patient cohort or if other factors, such as underlying health conditions and nutritional status, play a more preponderant role in determining mortality risk. Several biomarkers were found to be predictive of 1-year mortality. Hip fractures result in significant blood loss, and HGB levels of extracapsular fractures from admission to surgery have been shown to decrease by 31 g/L.²⁵ Worapaka et al. reported that preoperative correction of HGB to >10 g/dL would reduce one-half of the 1-year mortality in osteoporotic hip fracture patients and suggested that preoperative haemoglobin levels may be considered for preoperative planning in osteoporotic hip fracture surgery patients.²⁶ In the present study, preoperative HGB <117 g/L was regarded as an independent risk factor for 1-year mortality. However, there was no accordant cut-off value of HGB used to define anaemia or recommendation regarding patients with hip fracture who have to receive transfusion preoperatively.²⁷ Therefore, intravenous or oral tranexamic acid, intravenous iron supplementation, blood transfusion, and other effective measures with proven safety should be considered to optimise preoperative haemoglobin levels.

Recent work by Vural et al. (2025) proposed the HALP score (hemoglobin, albumin, lymphocyte, platelet) as a composite biomarker for mortality risk in hip fractures, supporting the prognostic value of combining hematologic and nutritional markers.²⁸

In addition to low BMI, Alb level <35 g/L or hypoalbuminaemia is also often regarded as a marker of malnutrition. Over 60% of patients with hip fractures were in a malnourished state, which could increase mortality risk.^21,29 In our model, preoperative Alb <33.9 g/L was found to bring an average almost 3-fold increase in postoperative 1-year mortality. Multidisciplinary and interdisciplinary approaches have been proven to decrease the incidence of malnutrition and effectively improve outcomes for patients with hip fractures.³⁰ It is worth mentioning that there was no significant difference detected in the last serum Alb level before discharge between the 2 groups, while patients with low preoperative Alb levels were still at a higher risk of death, indicating that preoperative Alb is a marker for poor nutritional status without medical monitoring; thus, long-term nutritional guidance after discharge for elderly hip fracture patients is important.

Interestingly, preoperative LDH >292 U/L was regarded as an independent risk factor for postoperative 1-year mortality in elderly hip fracture patients in our model. However, few studies have reported an association between LDH levels and hip fracture-related events. In a recent machine learning prediction study, LDH was indicated to be an important predictor of mortality and a contributing variable in classifying patients into sub-populations based on mortality risk in the ICU.³¹ Consistent with these findings, elderly hip fracture patients should also be treated as critically ill patients, which was one of the reasons why we implemented the MDT co-management model. Preoperative LDH as a potential predictor seems to be rational, and a possible explanation is that the degree of multi-organ dysfunction triggered by systematic inflammation following fracture might be associated with elevated LDH.³² This finding should be interpreted with caution, and we should pay more attention to these patients and determine why they have a higher 1-year mortality risk from a variety of different perspectives.

Impaired renal function (defined as an estimated glomerular filtration rate (eGFR) < 55.4 mL/min) has been found to be an independent risk factor for all-cause 1-year mortality in hip fracture patients.³³ In this research, we also found that poor renal function (expressed as BUN >8.5 mmol/L) was an important preoperative indicator for postoperative 1-year mortality in elderly patients with hip fractures under the MDT co-management model. For elderly patients with hip fractures, rational drug use and haemodynamic support may be useful to avoid worsening renal function, and the involvement of pharmacists in MDT is necessary. Perioperative impaired renal function especially acute kidney injury following surgery has recently been focused on hip fracture studies.³⁴ Fluid supplementation starting in the emergency department was regarded as a ratinal explanation for reduced acute kidney injury (AKI) rate,³⁵ which provided evidence for the necessary of establishing emergency fast track in MDT co-managemnet model. Further research is needed to determine the mechanism of renal-insufficiencyrelated high mortality and develop corresponding measures.

This study had several limitations. First, we focused on preoperative factors; therefore, some important intraoperative and postoperative variables were not included in the analyses. However, we believe that the difference in patient outcomes depends more on themselves rather than other variables, considering the standardisation of the management model. Second, owing to the retrospective design, some selection bias might have been introduced, and nearly all analysed factors were objective. Third, the absence of a non-MDT control group prevents direct attribution of mortality reduction to the MDT model. Fourth, as this was a single-centre study, the sample size was small. Finally, the usage of P < 0.05 as the threshold for including the variables in the multivariate analysis was strict and may have induced the loss of potentially significant variables.

While our findings align with known risk factors in hip fracture patients, generalizability may be limited to similar tertiary care centers with MDT resources. Further multicenter studies are needed to validate these predictors across diverse healthcare settings.

Conclusions

This study provided evidence for independent preoperative indicators of 1-year mortality following hip fracture surgery in elderly patients under the MDT co-management model, including age >85 years, BMI <21.0 kg/m², preoperative HGB <117 g/L, Alb <33.9 g/L, LDH >292 U/L, and BUN >8.5 mmol/L. MDT will pay more attention to patients with the indicators mentioned above. Consideration of these indicators when classifying patient severity and optimisation of the perioperative management for modified indicators would assist in the rational allocation of medical resources and benefit patients. Further multicentre studies with larger sample sizes are warranted to confirm these associations and to help optimise the preoperative assessment of the MDT co-management model.

Supplemental Material

Supplemental Material - Preoperative Indicators for 1-year Mortality in Elderly Individuals Following Hip Fracture Surgery Under A Multidisciplinary Team Co-Management Model: A Single-Centre Retrospective Observational Study

Supplemental Material for Preoperative Indicators for 1-year Mortality in Elderly Individuals Following Hip Fracture Surgery Under A Multidisciplinary Team Co-Management Model: A Single-Centre Retrospective Observational Study by Yucheng Gao, Shaoyang Zhou, Wang Gao, Yuanwei Zhang, Liu Shi, Tian Xie, Chuwei Tian, Hui Chen, and Yunfeng Rui in Geriatric Orthopaedic Surgery & Rehabilitation

Supplemental Material

Supplemental Material - Preoperative Indicators for 1-year Mortality in Elderly Individuals Following Hip Fracture Surgery Under A Multidisciplinary Team Co-Management Model: A Single-Centre Retrospective Observational Study

Footnotes

ORCID iDs

Shaoyang Zhou

Yunfeng Rui

Ethical Approval

This was an observational study. This study received ethical approval from the the independent ethics committee (IEC) for Clinical Research of Zhongda Hospital Affiliated to Southeast University (2022ZDSYLL183-P01).

Author Contribution

Y.G, S.Z and W.G designed the study and wrote the manuscript. Y.G performed the main data analyses. C.T participated in data collection. Y.Z, L.S, T.X, H.C and Y.R put forward constructive suggestions for this study. All authors have contributed to the manuscript and approved the submitted version.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Winfast Charity Foundation (YL20220525). The funding source(s) had no such involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

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.

Data Availability Statement

The patients’ data were collected in Zhongda Hospital Affiliated to Southeast University. *

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Yan

Low

, et al. Predictors of poor functional outcomes and mortality in patients with hip fracture: a systematic review. BMC Musculoskelet Disord. 2019;20:568.

Cooper

CCG

Melton

3rd . Hip fractures in the elderly_ A world-wide projection. Osteoporos Int. 1992;2(6):285-289.

Loggers

SAI

Van Lieshout

EMM

Joosse

, et al. Prognosis of nonoperative treatment in elderly patients with a hip fracture: a systematic review and meta-analysis. Injury. 2020;51:2407-2413.

Bhandari

Swiontkowski

. Management of acute hip fracture. N Engl J Med. 2017;377:2053-2062.

Zhou

Rui

, et al. [Research progress of multidisciplinary team co-management models for geriatric hip fracture treatment]. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi. 2020;34:132-138.

Xie

Rui

Liu

, et al. [Preliminary application of postoperative fast track transfer to intensive care unit for the geriatric hip fractures under enhanced recovery after surgery]. Zhonghua Yixue Zazhi. 2020;100:2897-2902.

Sing

Lin

Bartholomew

, et al. Global epidemiology of hip fractures: secular trends in incidence rate, post-fracture treatment, and all-cause mortality. J Bone Miner Res. 2023;38:1064-1075.

Özel

Altıntaş

Tatlıparmak

. Predictors of one-year mortality following hip fracture surgery in elderly. PeerJ. 2023;11:2.

Fan

, et al. The efficacy of multidisciplinary team Co-management program for elderly patients with intertrochanteric fractures: a retrospective study. Front Surg. 2022;8:1.

10.

China MoHotPsRo . Guiding principles for the classification of emergency patients' conditions (draft for comments). Chin J Crit Care Med (Electronic Edition). 2011;4:241-243. (in Chinese).

11.

Gilchrist

Newman

Hamblen

, et al. Prospective randomised study of an orthopaedic geriatric inpatient service. Bmj. 1988;297:1116-1118.

12.

Tian

Zhang

, et al. The effect of a multidisciplinary co-management program for the older hip fracture patients in Beijing: a “pre- and post-” retrospective study. Arch Osteoporosis. 2019;14:2.

13.

Turgut

Unal

. Standard and newly defined prognostic factors affecting early mortality after hip fractures. Cureus. 2022;14:e21464.

14.

De Luca

Murena

Zanetti

, et al. Should the early surgery threshold be moved to 72 h in over-85 patients with hip fracture? A single-center retrospective evaluation on 941 patients. Arch Orthop Trauma Surg. 2022.

15.

Fukui

Fujioka

Yamasaki

, et al. Risk factors for postoperative complications among the elderly after plastic surgery procedures performed under general anesthesia. Plast Surg Int. 2018;2018:7053839.

16.

Leer-Salvesen

Engesæter

Dybvik

, et al. Does time from fracture to surgery affect mortality and intraoperative medical complications for hip fracture patients? An observational study of 73 557 patients reported to the Norwegian Hip Fracture Register. Bone Joint Lett J. 2019;101-b:1129-1137.

17.

Castellanos

Marco

Domínguez

, et al.

Does delaying surgery for “healthy” hip fracture patients have increased complications and mortality?

Injury. 2022.

18.

Keohane

Downey

Sheridan

, et al. Hip fracture surgery within 36 hours reduces both 30-day and 1-year mortality rates. Surgeon. 2022;20:262-267.

19.

Wang

, et al. The impact of body mass index on mortality rates of hip fracture patients: a systematic review and meta-analysis. Osteoporos Int. 2022.

20.

Malafarina

Reginster

Cabrerizo

, et al. Nutritional status and nutritional treatment are related to outcomes and mortality in older adults with hip fracture. Nutrients. 2018;10:555.

21.

Wilson

Boissonneault

Schwartz

, et al. Frailty and malnutrition are associated with inpatient postoperative complications and mortality in hip fracture patients. J Orthop Trauma. 2019;33:143-148.

22.

Akinleye

Garofolo

Culbertson

, et al. The role of BMI in hip fracture surgery. Geriatr Orthop Surg Rehabil. 2018;9:2151458517747414.

23.

Meltzer-Bruhn

Esper

Herbosa

, et al. The role of smoking and body mass index in mortality risk assessment for geriatric hip fracture patients. Cureus. 2022;14:e26666.

24.

Amin

Raad

Rao

, et al. Survival bias may explain the appearance of the obesity paradox in hip fracture patients. Osteoporos Int. 2021;32:2555-2562.

25.

Wang

, et al. The efficacy and safety of intravenous tranexamic acid in hip fracture surgery: a systematic review and meta-analysis. J Orthop Translat. 2019;19:1-11.

26.

Manosroi

Atthakomol

Isaradech

, et al. Preoperative correction of low hemoglobin levels can reduce 1-year all-cause mortality in osteoporotic hip fracture patients: a retrospective observational study. Clin Interv Aging. 2022;17:165-173.

27.

Patel

Rahim

Davidson

, et al. Society of interventional radiology consensus guidelines for the periprocedural management of thrombotic and bleeding risk in patients undergoing percutaneous image-guided interventions-Part II: recommendations: endorsed by the Canadian association for interventional radiology and the cardiovascular and interventional radiological society of Europe. J Vasc Intervent Radiol : J Vasc Intervent Radiol. 2019;30:1168-1184.

28.

Vural

Dolanbay

Yagar

. Hemoglobin, albumin, lymphocyte and platelet (HALP) score for predicting early and late mortality in elderly patients with proximal femur fractures. PLoS One. 2025;20:2.

29.

Fujimoto

Setoguchi

Ishidou

, et al. Low geriatric nutritional risk index is a risk factor for death within 1 year following hip fracture. J Orthop Surg. 2022;30:10225536221103360.

30.

Hoekstra

Goosen

de Wolf

, et al. Effectiveness of multidisciplinary nutritional care on nutritional intake, nutritional status and quality of life in patients with hip fractures: a controlled prospective cohort study. Clin Nutr. 2011;30:455-461.

31.

Iwase

Nakada

Shimada

, et al. Prediction algorithm for ICU mortality and length of stay using machine learning. Sci Rep. 2022;12:12912.

32.

Zhu

Guo

, et al. Incidence and risk factors associated with surgical site infection after surgically treated hip fractures in older adults: a retrospective cohort study. Aging Clin Exp Res. 2022;34:1139-1148.

33.

Gulin

Kruljac

Kirigin

, et al. Advanced age, high beta-CTX levels, and impaired renal function are independent risk factors for all-cause one-year mortality in hip fracture patients. Calcif Tissue Int. 2016;98:67-75.

34.

Zhan

Xie

Yang

, et al. Incidence and risk factors of acute kidney injury after femoral neck fracture in elderly patients: a retrospective case-control study. BMC Musculoskelet Disord. 2022;23:7.

35.

Agar

Gulabi

Sahin

, et al. Acute kidney injury after hip fracture surgery in patients over 80 years of age. Arch Orthop Trauma Surg. 2022;142:2245-2252.

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

0.22 MB

0.09 MB