Echocardiography Does Not Delay Surgery in Elderly Patient with Hip Fracture: Pulmonary Hypertension and Decreased Left Ventricular Ejection Fraction Are Associated with in Hospital Mortality

1. Introduction

Routine preoperative echocardiography before non-cardiac surgery has not been demonstrated to improve outcomes in large cohort studies [1,2], although specific abnormalities such as degree of mitral regurgitation or aortic stenosis, LVEF and E/e’ ratio may be associated with increased mortality in patients undergoing intermediate- or high-risk noncardiac surgery [3]. Cardiovascular complications, mainly heart failure and myocardial infarction/damage, are the leading causes of death and prolonged hospitalization after hip surgery [4]. Risk stratification according to the AHA/ACC or ESC guidelines in patients with known heart disease may be difficult since at least 35% of patients have cognitive impairment and/or moderate/severe functional limitations [5,6]. Symptoms related to heart failure or chronic coronary artery disease may be underestimated. Early surgery, within 24-36 hours from trauma, is associated with a significant improvement in clinical outcomes [7,8]. According to SIGN guidelines echocardiography is recommended to better stratify surgical risk and define anesthesiologist strategy in suspected aortic stenosis if examination does not delay surgery [9]. Substantial changes in the perioperative management has been reported in not less than 20% of hip fracture patients after echocardiogram. Moderate to severe valvular heart disease (such as aortic stenosis or mitral regurgitation) are not rarely found by chance at hospital admission or have not been reevaluated since a long term [10,11]. Moreover, in patients with known heart failure or unexplained poor functional capacity, the assessment of left ventricular function and filling pressures may guide therapeutic optimization to limit postoperative complications rate. Results of previous investigations about preoperative echocardiography are contrasting. The main concern with preoperative echocardiography in patients with hip fracture is related to unacceptable delay in treatment and related increase in complications and mortality [12,13,14]. Canty et al. [15] compared high-risk cardiac patients underwent a preoperative transthoracic echocardiography (TTE) toa randomized cohort with similar cardiac risk but not undergoing preoperative TTE. Mortality at both 30 days (4.7% vs. 15.2%, p=0.047) and 12 months postoperatively (17.1% vs. 33.3%, p=0.031) was lower in patients underwent echocardiography. Among 354 patients underwent TTE before hip surgery age, history of CAD, presence of moderate-severe aortic stenosis and LVEF <50% were independent predictors of mortality [16]. Higher average E/e’ values, lower left ventricular (LV) ejection fraction, and higher prevalence of significant mitral regurgitation were related to the risk of major cardiovascular events or death in 1453 patients (51% male; age, 67 ± 16) who underwent intermediate- or high-risk major abdominal surgery or orthopedic surgery [3]. A large retrospective study did not show in-hospital mortality differences between echocardiography screened versus no screened in more than 25,000 propensity score matched patients. A greater average time to surgery without significant advantage on in hospital outcome were reported by AbuSharar et al. [17]. The study however, included only 42 patients, 22% of those had ACC/AHA indications for pre-operative echocardiogram. Aim of the present study was to evaluate whether in setting of a multidisciplinary hip fracture management time to surgery was delayed by echocardiography, when needed. Secondary endpoints were the effects of examination on in hospital mortality and length of stay. Finally, we searched for the relation between specific echocardiographic abnormalities and in-hospital mortality.

2. Materials and Methods

The study is part of a project of Italian Health Ministry and Regione Toscana –RF-2010-2316600- and was approved by Ethical Committee of Regione Toscana. Written informed consent to treatment and collection of clinical data for research purposes was obtained at admission. The study was conducted according to STROBE statements and performed in line with the principles of the Declaration of Helsinki.

The aim of the study was to evaluate the relationship between echocardiographic alterations and intra-hospital mortality in elderly patients hospitalized for hip fragility fracture. As a secondary objective, the difference in outcome between patients who had undergone pre-operative echocardiography and those who were referred for surgery without echocardiographic examination was compared.

2.1. Preoperative Evaluation

In the study were enrolled all patients with hip fracture admitted to a tertiary teaching hospital between January 1, 2019 and December 31, 2024. The patients were followed by a multidisciplinary group for fragility fractures. At admission each patient was carefully evaluated by the Internal Medicine-geriatric specialist (general clinical conditions, presence of comorbidities and ongoing medical treatment). For each patient , demographic data (age and sex), functional capacity prior to trauma (using the Basic Activities of Daily Living Assessment Scale, BADL), and the presence of comorbidities (arterial hypertension, atrial fibrillation, ischemic heart disease, heart failure, COPD, diabetes, both peripheral and cerebral vasculopathy, neoplasms, renal failure, anemia -Hb < 10 g/dl, osteoporosis and a number of ≥ 2 disease, and cognitive Impairment) were recorded in the database. Echocardiography was performed usually within 24 hours by a single operator with an ESAOTE MY LAB 40 echocardiograph. In the protocol were included patients with newly detected or known systolic murmurs, ECG abnormalities, known heart disease or presence of >3 coronary risk factors. Parameters were collected according to criteria of the European Association of Echocardiography (EAE) /American Society of Echocardiography (ASE) {18].

Left ventricular ejection fraction was measured using Simpson rule and patients were stratified into four groups. LVEF < 30% severe impairment of systolic function, LVEF between 30-39% moderate systolic dysfunction, LVEF between 40-55% mild systolic dysfunction, LVEF > 55% normal systolic function.

Systolic pulmonary artery pressure was calculated according to standardized criteria. In relation with measured values patients were stratified in: mild pulmonary hypertension (PAPs 25-29 mmHg), moderate pulmonary hypertension (PAPs 30-40 mmHg), severe pulmonary hypertension (PAPs >40 mmHg). Severity of valve regurgitation (aortic and mitral) was assessed by semiquantitative method and stratified as mild, moderate, and severe. Patients with aortic stenosis were divided into 3 groups according to peak gradient: mild aortic stenosis (gradient between 30 and 49 mmHg), moderate aortic stenosis (gradient between 50 and 70 mmHg) and severe aortic stenosis (gradient > 70 mmHg). Right ventricular function evaluated by Tricuspid Annular Plane Systolic Excursion (TAPSE) was defined as normal, (TAPSE > 20mm), mild dysfunction (TAPSE between 20 and 15 mm) and severe dysfunction (TAPSE < 15 mm).

2.2. Statistical Analysis

The parameters considered were expressed as mean values and standard deviations. Categorical variables were expressed as distribution frequencies. In the comparison between groups continuous parameters were compared by Student Test for unpaired data while the χ² or the Fisher Exact Test were used for categorical variables. The influence of the different parameters on mortality was evaluated by multivariate logistic regression analysis.

3. Results

The study included 2274 patients admitted to a tertiary teaching hospital between January 1, 2019, and December, 31, 2024 with diagnosis of low energy hip fracture. Mean age was 83.8 years. More than seventy percent were females (72.6%) with a female-to-male ratio of 2.65:1. Types of fracture are reported in table. A significant functional impairment, expressed as loss of 2 or more BADL, was found in 28.2%. Twenty-one percent had cognitive impairment. More than half of the patients (62.4%) had at least 2 comorbidities. Time to surgery was less than 48 hours from trauma in 76.5% of patients. Only 24 (about 1%) were treated conservatively. Characteristics of patients are reported in Table 1.

According to study protocol 1593 patients had indication to perform preoperative echocardiography. Echocardiography was performed before surgery in 1502 patients. Ninety-one despite clinical indication were not examined, mainly due to organization problems. The other 681 without clinical indications, “NO ECHO” group, underwent only routine ECG without further preoperative cardiological evaluation. 86% of echocardiograms was performed within 24 hours from trauma. In Figure 1 is reported the flow chart of the study.

Mean age was higher in the group with indication to echocardiography (85.4 ± 7.2 years ECHO group vs 80.4 ± 11.2 the NO ECHO group). Both groups had a higher female prevalence (71.3% ECHO and 74.87% NO ECHO respectively) however the proportion was lower in those with indication to echo who did not perform examination. In patients not evaluated with preoperative echocardiogram time to surgery was on average 2.1 ± 2.3 days in comparison with 2.7 ± 4.1 days in those who underwent echocardiogram. Similar time to surgery was found patients with indication to examination in which the exam was not performed.

Detailed data about type of fracture and surgical treatment are reported in Table 2.

Functional impairment defined as BADL < 4 was found in 22.8% of patients who did not undergo echocardiogram and 31.09% of patients who underwent echocardiography. Two or more comorbidities were found in 72% patients whit indication echocardiography in comparison to 42% in patients who did not. The frequencies of the different comorbidities found in the three groups are shown in the Table 3:

In the echocardiographic group, left ventricular dysfunction was appreciable in 24.6%, while severely depressed ventricular function in 19 patients. Moderate-severe aortic stenosis and mitral regurgitation were found respectively in 7 and 23%. Moderate-to-severe pulmonary hypertension was observed in 40.6%, while only 18 patients showed a depressed RV function (TAPSE < 15). Twenty five percent had abnormalities of diastolic pattern. Echocardiogram led to a modification of the anesthetic strategy (general vs. spinal anesthesia) in 13% of patients and in 20% to the need of ICU stay in the 24 hours after surgery. Aortic stenosis accounted for 70% of cases, a marked reduction in left ventricular function in 27% to and finally severe mitral valve disease in the remaining 3%.

Figure 2. Frequency distribution of moderate-severe echocardiographic abnormalities.

Figure 2. Frequency distribution of moderate-severe echocardiographic abnormalities.

In hospital mortality was 2% in patients who did not undergo echo, 7.3% in patients who underwent echo and finally 10% in patients with indication to echo in whom examination was not performed. In each group patients discharged alive were younger and more frequently female. Time to surgery was longer in patients died in hospital as well they had a lover functional capacity (Table 4). Anemia and cognitive impairment were the only factors associated with mortality in patients who did not undergo echocardiographic examination (Table 5). In patients underwent echocardiographic examination anemia, atrial fibrillation, heart failure, COPD, cognitive impairment, >2 comorbidities and finally renal failure were significantly more frequent in patients died in hospital in comparison to discharged alive

In Table 5 and Table 6 are reported the different comorbidities rate in patients of non-echo group and respectively of patients with indication underwent or not echocardiography.

Figure 3. Distribution of LVEF and pulmonary artery pressure in patients discharged alive and died in hospital.

Figure 3. Distribution of LVEF and pulmonary artery pressure in patients discharged alive and died in hospital.

Regard echocardiographic parameters moderate and severely depressed left ventricular function was found in 5.5% and 1.0% of patients discharged alive in comparison to 17.3% and respectively 4.5% of patients died in hospital (p<0.0001) (Figure 1). Moderate-to-severe mitral regurgitation was found in 22.8% of patients discharged alive and in 31.8% of patients went to death (p=0.0014). Thirty-nine percent of patients discharged alive had moderate to severe pulmonary hypertension in comparison to 59.1% of patients who died. (p=0.0018) We did not find any statistically significant difference in rate of aortic stenosis and regurgitation, diastolic abnormalities and finally TAPSE between patients discharged alive and patients died in hospital. At multivariate analysis age, time to surgery, anemia at admission, and cancer were independently related to in-hospital mortality while preserved functional capacity expressed as maintained BADL was related with survival. Among echocardiographic parameters an increased in-hospital mortality was associated with reduced EF (OR 1.45 – 95% CI 1.06- 1.93 p=0.02) and increased pulmonary pressure (OR 1.22 – 95% CI 1.02- 1.47, p=0.03) (table

	Odds Ratio	95% CI	P
Age	1.07	(1.031 to 1.11)	0.0005
Time to surgery	1.03	(1.01 to 1.07)	0.0423
BADL >4	0.85	(0.74 to 0.98)	0.0286
Anemia<10 g/dl Hb	3.41	(1.82 to 6.37)	0.0001
Cancer	1.95	(1.18 to 3.21)	0.0084
Left Ventricular ejection fraction	1.45	(1.07 to 1.99)	0.0189
Pulmonary hypertension	1.22	(1.02 to 1.47)	0.0257

4. Discussion

AHA/ACC guidelines for cardiac evaluation in patients who need non-cardiac surgery suggest assessment of left ventricular function (class I, N-RB) in patients with new dyspnea, physical examination findings of HF, or suspected new/worsening ventricular dysfunction, as well in patients with aortic stenosis to help guide perioperative management [6]. Otherwise, ESC guidelines recommend echocardiography with in patients with poor functional capacity and/or high NT-pro-BNP/ BNP or if murmurs are detected before high-risk (class I B) and in patients with suspected new CVD or unexplained signs or symptoms before high-risk NCs (class IIa,B) [5]. Patients with fragility hip fracture represent a major challenge in preoperative evaluation since significant cardiopathy may be masked by poor functional capacity related to age and frailty as well to cognitive impairment therefore they may benefit most from preoperative echocardiography. Results of previous investigations about preoperative echocardiography in patients with hip fracture are contrasting. Since both survival and functional results are related to time to surgery, within 24-48 hours from trauma, the main concern with preoperative echocardiography is related to unacceptable delay in treatment waiting examination and related increase in complications and mortality [12,13,14]. Canty et al. [15] compared patients at increased cardiac risk underwent preoperative transthoracic echocardiography (TTE) to a randomized cohort with similar cardiac risk but not undergoing preoperative TTE. In the TTE group mortality was lower at both 30 days and 12 months postoperatively ((4.7% vs. 15.2%, p=0.047 and respectively 17.1% vs. 33.3%, p=0.031). In the study by Chen et al. [19] mortality rate was 2.8% while postoperative cardiac complications were 7.6%. History of coronary artery disease (CAD) and presence of aortic stenosis were independent predictors of postoperative cardiac complications. Age history of CAD, aortic stenosis and LVEF <50% were independent predictors of mortality. A retrospective study included 1453 (51% male; age, 67 ± 16) underwent TTE before intermediate- or high-risk major abdominal surgery or orthopedic surgery [20]. The primary endpoints were major adverse events (MAEs), i.e., all-cause mortality and major adverse cardiovascular-cerebral events (MACCEs) at a follow-up of 56 days. Mortality was 2.4% and MACCEs rate 1.2%. Higher average E/e’ values, lower Left ventricular (LV) ejection fraction, and higher prevalence of significant mitral regurgitation (MR) other than moderate-advanced chronic kidney disease (CKD) were associated with MAEs. A recent case control matched 113 hip fracture patients who did or did not undergo preoperative TTE [21]. Indication for TEE was adequate in 71%. TTE group had longer time to surgery and LOS of average 20 hours and respectively 2 days in comparison to non-TTE group (p < 0.0001). There was higher 90-day mortality for the TTE group (odds ratio 4.4, 95% confidence interval 1.3-14.7, p < 0.03) but no difference at 2-years. In a large retrospective study form Japan 34,679 (52.1%) of 66,620 hip fracture patients underwent preoperative echocardiography screening [16]. Propensity score matching created a matched cohort of 25,205 pairs of patients. There were no in-hospital mortality differences between the 2 groups (screened versus no screened: 417[1.65%] vs 439 [1.74%]). Preoperative echocardiography was not associated with reduced postoperative complications and intensive care unit admissions. In our study echocardiography was performed according to protocol in patients with newly detected or known systolic murmurs, ECG abnormalities, known heart disease or presence of >3 coronary risk factors. Ninety-four % of those with clinical indications underwent examination. We did not find difference in time to surgery between patients underwent or not to echocardiography, since our organization allowed to perform bed-side examination at time of the admission in multidisciplinary ward. In more than 20% of those underwent echocardiography perioperative strategy was changed according to ultrasonographic findings: in 13% anesthesiologic strategy was modified from spinal to general anesthesia, in the remnants post-operative ICU was required. TTE had a higher mortality, as expected since inclusion criteria identified a higher-risk surgical group. Mortality in patients with indication but who do not undergo echocardiography had 10% mortality in comparison to 7.3% of those underwent examination. The small number of patients in the group who did not made echo and the absence of randomization do not allow to draw conclusions. At multivariate analysis in-hospital mortality was associated with reduced EF (OR 1.45 – 95% CI 1.06- 1.93 p=0.02) and increased pulmonary pressure (OR 1.22 – 95% CI 1.02- 1.47, p=0.03). Other factors related to mortality were age, time to surgery, anemia at admission, functional impairment, and cancer. Careful preoperative treatment with the aim to decrease congestion may improve outcome in these patients.

What is known about preoperative echocardiography in patients who need hip fracture surgery is the result of observational studies with different protocols and inclusion criteria and do not give definitive indication. An ongoing multicenter trial started in 2025 will enroll 2000 adults with hip fractures [22]. Participants will be randomized before surgery to either receive focused cardiac ultrasound as part of their preoperative assessment or to routine care without FCU. Aim of the study is to determine whether preoperative focused ultrasound examinations of hip fracture patients reduce serious postoperative complications, improves the quality of recovery, improves life quality, and is cost-effective.

5. Conclusions

Results from our study confirm that in a large volume well organized multidisciplinary hip fracture unit the execution of echocardiography does not affect time to surgery. Moreover, we demonstrated that impaired left ventricular ejection fraction and pulmonary hypertension are associated with an increased mortality. At present several evidence exist that echocardiographic findings have direct implications for perioperative management and risk stratification. Further investigations are needed to evaluate whether these changes may affect clinical outcome. Moreover, the ongoing randomized study had to establish the usefulness of focused echocardiography to improve clinical outcomes.