This retrospective cohort study included 997 consecutive patients over 65 years of age with a diagnosis of traumatic hip fracture admitted to our hospital from January 2019 to January 2024. A total of 361 patients were excluded based on the screening criteria and 636 patients were included in the final analysis. Patients were categorized into three groups based on the time from fracture onset to surgery: less than 3 days for the early group, 3–7 days for the intermediate group, and more than 7 days for the late group (Figure 1). Surgery was scheduled by the medical staff after assessing the health status of the patients. The study complied with the requirements of the Declaration of Helsinki, and all patients and their families signed an informed consent for surgery.

Inclusion criteria: (1) age ≥65 years; (2) conscious; (3) fresh hip fracture diagnosed by X-ray or CT and treated surgically; (4) complete clinical data; (5) normal hip function before the fracture. Exclusion criteria: (1) pathologic fracture due to tumor metastasis; (2) multiple fractures or open hip fractures; (3) unspecified type of fracture; (4) previous ipsilateral history of previous ipsilateral hip surgery; (5) fracture of unknown ASA grade; (6) typical hemolytic reaction after blood transfusion; (7) loss to follow-up or follow-up time <12 months.

This study used a strict exclusion method to handle missing data during the cohort selection process, and the final cohort consisted of 636 patients with complete data. Therefore, a complete case analysis was used without estimation. Data were collected retrospectively from chart reviews by two investigators using the same study protocol, which was approved by the institutional review board. Demographic and surgery-related data included age, gender, BMI, patient comorbidities, time from fracture onset to surgery, fracture type, American society of Anesthesiologists (ASA) classification, and type of surgery. Postoperative in-hospital status included complications, ICU transfer rate, ICU retention time, and length of stay. Post hospital status included follow-up time, follow-up Harris score, and survival outcome observations.

(1) Patients’ clinical baseline data. (2) General postoperative in-hospital conditions: ICU transfer rate, ICU retention time, and hospitalization time. (3) Incidence of postoperative complications: pulmonary infection: pneumonitis caused by pathogens, physicochemical factors, and other elements; urinary tract infection: positive urine culture (≥10⁵ CFU/mL) with symptoms or pyuria; incision infection: purulent drainage from the incision, or positive culture from aseptically obtained fluid, with local signs of inflammation; lower extremity deep vein thrombosis: abnormal coagulation of venous blood in the deep veins of the lower extremities, blocking the lumen and causing venous return disorders; stress ulcers: overt gastrointestinal bleeding (hematemesis or melena) confirmed by endoscopy; anemia: the number of red blood cells, or the concentration of hemoglobin in them, is lower than normal; hyperproteinemia: total serum protein > 80 g/L; delirium: acute onset of fluctuating mental status and inattention, with either disorganized thinking or altered level of consciousness; pressure ulcer: damage to the skin and subcutaneous tissue due to nutrient deficiencies, ulcers and even necrosis. (4) Details of in-hospital deaths (5) Efficacy and prognosis: Harris score at the 1st and 3rd postoperative months, total score of 100, the higher the score, the better the efficacy (26). Mortality rate of patients at 1 month, 3 months, 6 months and 1 year postoperatively.

Data were analyzed and plotted using GraphPad Prism 8.0. Measurement data were expressed as mean ± standard deviation, and count data were expressed as [n (%)]. Comparison of measurement data between groups was performed by independent samples t-test, and comparison of count data between groups was performed by chi-square test or Fisher’s exact test, and p < 0.05 indicated that the difference was statistically significant.

A total of 636 patients with a mean age of 76.8 years were included in the analysis, of whom 431 (67.8%) were female. Of the 636 patients, 257 (40.4%) were operated on within 3 days of fracture (early group), 224 (35.2%) were operated on within 3–7 days (intermediate group), and 155 (24.4%) were operated on after at least 7 days (late group). The 3 groups of patients were compared with respect to their age, gender, body mass index, type of fracture, ability to walk on admission, comorbidities, baseline data such as type of surgery and ASA classification were compared, and the differences were not statistically significant (p > 0.05) (Table 1).

All patients underwent surgery successfully. After surgery, the number of ICU transfers was 8 (3.6%) and 12 (7.7%) in the intermediate and late groups, respectively, while the number of ICU transfers in the early group was only 5 (1.9%). ICU retention time in the early group was significantly shorter than that in the intermediate group (1.6 ± 0.4 d vs. 2.6 ± 0.3 d, p < 0.001), whereas the difference in ICU retention time between the intermediate group and the late group was not statistically significant (2.6 ± 0.3 d vs. 2.9 ± 0.4 d, p = 0.070). In addition, as the preoperative waiting time increased, patients’ postoperative hospitalization time increased, and the hospitalization time in the early group was significantly shorter than that in the intermediate group (9.5 ± 4.2 d vs. 11.9 ± 3.7 d, p = 0.047), whereas there was no statistically significant difference in the comparison of postoperative hospitalization time in the intermediate group versus the late group (11.9 ± 3.7 d vs. 13.3 ± 4.5 d, p = 0.082) (Table 2).

The difference in the incidence of postoperative urinary tract infection, incision infection, anemia, hyperproteinemia, and delirium was not statistically significant in the three groups (p > 0.05). And the postoperative pulmonary infection (2.7% vs. 6.3% vs. 8.4%, p = 0.035), deep vein thrombosis (3.9% vs. 6.7% vs. 11.6%, p = 0.010), stress ulcers (1.9% vs. 3.6% vs. 7.1%, p = 0.029), and pressure sores (2.3% vs. 6.7% vs. 7.7%, p = 0.025) all showed an increasing trend in incidence with increasing preoperative waiting time, with the lowest in the early group, the second highest in the intermediate group, and the highest in the late group, and the trends were all statistically significant (Table 3).

The total number of in-hospital deaths in the 3 groups was 43 (6.8%), including 9 (3.5%) in the early group, 18 (8.0%) in the intermediate group, and 16 (10.3%) in the late group. The top three causes of death were pneumonia in 9 cases (20.9%), heart failure in 6 cases (14.0%) and arrhythmia in 6 cases (14.0%), in that order. No one in the early group died of aortic aneurysm rupture, and there were more deaths from pneumonia and heart disease (heart failure and arrhythmia) in the intermediate and late groups than in the early group (Table 4).

Patients in all 3 groups received at least 1 year of postoperative follow-up. Harris scores tended to decrease at 1 month postoperatively with increasing preoperative waiting time, and were significantly higher in the early group than in the intermediate group (87.1 ± 5.3 vs. 82.2 ± 5.6, p < 0.001), and in the intermediate group than in the late group (82.2 ± 5.6 vs. 78.4 ± 5.0, p = 0.008), with no significant difference between the 3 groups at 3 months postoperatively (p > 0.05) (Figure 2A). After excluding the number of in-hospital deaths, the cumulative number of deaths at 1 year postoperatively in 248 patients in the early group was 6, with a cumulative mortality rate of 2.4%, the cumulative number of deaths at 1 year postoperatively in 206 patients in the intermediate group was 14, with a cumulative mortality rate of 6.8%, and the cumulative number of deaths at 1 year postoperatively in 139 patients in the late group was 10, with a cumulative mortality rate of 7.2% (Figure 2B). At 1 year postoperatively, the trend of increasing mortality was significantly higher in the intermediate group than in the early group (p = 0.022), whereas the difference in trend with the late group was not statistically significant (p = 0.906) (Figure 2C). The survival curves showed that the one -year survival rates of the early, intermediate and late groups were 97.6, 93.2 and 92.8%, respectively, and the differences between the three groups were statistically significant (p = 0.046) (Figure 2D).

This study is a retrospective cohort study, which may have some selection bias and information bias. Patients in better condition may be selected for early surgery, which could bias the results in favor of this group of patients. When patients were grouped, although they were grouped according to the time from fracture to surgery, there may have been certain unmeasured or under considered confounding factors, such as the patient’s family economic status and the accessibility of healthcare resources, which may affect the choice of the timing of surgery and the patient’s prognosis. Unadjusted comparisons of baseline characteristics showed a significant benefit of early surgery, but the lack of multivariate regression or propensity score analyses prevented full consideration of all potential confounding influences. In addition, the sample size of the study was relatively limited, and some of the data in the study relied on medical records, which may have been incomplete or inaccurate, thus affecting the results of the study to some extent. Future studies can carry out prospective randomized controlled trials to randomly group elderly hip fracture patients according to the timing of surgery to more accurately assess the impact of different timing of surgery on postoperative complications and prognosis. At the same time, the research indexes can be further refined, such as analyzing different types of fractures (femoral neck fracture, intertrochanteric fracture, etc.) separately, to explore whether there is a difference in the impact of surgery timing on patients with different types of fractures. In addition, new technologies such as biomarkers and genetic testing can be combined to study in depth the relationship between surgical timing and patients’ body functions, inflammatory response, immune function, etc., to provide a basis for making more precise decisions on surgical timing. Through these in-depth studies, the treatment strategy of elderly hip fracture patients can be further optimized to improve the treatment effect and quality of life of patients.