Clinical effects and safety of proximal femur bionic nail versus proximal femoral nail anti-rotation or InterTAN for the treatment of intertrochanteric femoral fracture: a systematic review and meta-analysis

Inclusion and exclusion criteria

Search strategy

A search was conducted in databases including PubMed, WOS, Embase, Cochrane, CNKI, Wanfang, and VIP for clinically controlled studies on the treatment of intertrochanteric femoral fracture (IFF) with PFBN published prior to 16 December 2025. “Proximal femur bionic nail” and “PFBN” were used as search terms. The search strategy is shown in Appendix S1.

Literature screening and data extraction

After the studies were retrieved, they were imported into EndNote, and duplicate references were removed. Two researchers (Zhang and Ma) independently screened the literature on the basis of the inclusion and exclusion criteria. The titles and abstracts of the remaining studies were read after removing duplicates and preliminarily excluding irrelevant studies, after which the full texts of the remaining studies were read. Finally, data from the literature, including basic features of the literature (author names, country, publication year, etc.) and the patients (intervention and control measures, sample size, sex ratio, age, follow-up time, etc.), trial results, and quality assessment methods, were extracted. In the case of any disagreements, the final determination will be made by a senior researcher (Hu). If only the median (interquartile range) is given in the study, the conversion is made via online data conversion (https://www.math.hkbu.edu.hk/ tongt/papers/median2mean.html). After the extraction was complete, the data were entered into Review Manager 5.4 (RevMan 5.4) software (Cochrane Collaboration, 2020) for analysis.

Quality evaluation of the literature

The quality of cohort studies was assessed according to the Newcastle-Ottawa Quality Assessment Scale (NOS) (Stang, 2010). A funnel plot was used to assess the presence of publication bias in the included studies.

Statistical methods

Data analysis was performed via RevMan 5.4. Binary variables are expressed as the relative risk (RR) and 95% confidence interval (CI), whereas continuous variables are expressed as the standardized mean difference (SMD) and 95% CI. For heterogeneity testing, the chi-square test and I² test were used. When I² ≤ 50% and P ≥ 0.1, low heterogeneity was indicated, and a fixed effects model was selected. Conversely, if I² > 50% or P < 0.1, high heterogeneity was suggested. If the source of heterogeneity could not be identified, a random effects model was chosen for analysis, subgroup analysis and sensitivity analysis was conducted when necessary. A difference was considered statistically significant when P < 0.05.

Literature search results

In the initial assessment, 231 studies were identified. Following a rigorous preliminary review, the full-text examination led to the retention of 15 studies that met the inclusion criteria of this research (Liu et al., 2023; Yang et al., 2023; Fu et al., 2023; Jin et al., 2024; Long & Li, 2022; Ling et al., 2023; Li et al., 2022; Lin et al., 2022; Zhang & Peng, 2022; Wang et al., 2023; Wu et al., 2023; Wan et al., 2024; Jia et al., 2023; Fu et al., 2024; Zhang, Zhang & Zhang, 2024). Notably, all 15 retained studies had a cohort design. The literature screening process is depicted in Fig. 1.

A total of 15 studies with 990 patients were included, of which 401 were treated with PFBN, 476 with PFNA, and 113 with InterTAN. The essential characteristics of the studies included in this study are provided in Table 1.

Quality assessment results

Among the included cohort studies, exposure determination and cohort selection were distinct and comparable, with no pre-existing outcomes. The outcome index measurement method was suitable. Eight studies (Liu et al., 2023; Yang et al., 2023; Long & Li, 2022; Ling et al., 2023; Li et al., 2022; Wang et al., 2023; Wu et al., 2023; Fu et al., 2024) reported follow-up periods of less than one year. One study (Zhang & Peng, 2022) did not detail the follow-up times or procedures and thus was not scored. All studies scored over seven on the NOS scale, ensuring quality.

Meta-analysis results

A summary of the meta-analysis results is shown in Table 2.

Partial weight bearing time

A total of 10 studies (Liu et al., 2023; Yang et al., 2023; Fu et al., 2023; Jin et al., 2024; Long & Li, 2022; Li et al., 2022; Zhang & Peng, 2022; Wang et al., 2023; Wan et al., 2024; Fu et al., 2024) reported partial weight-bearing time after surgery. Considerable heterogeneity was observed across the studies. As sensitivity analyses (e.g., leave-one-out method) failed to pinpoint a specific source of this heterogeneity, a random-effects model was therefore applied for the meta-analysis. The results revealed that the partial weight-bearing time in the PFBN group was significantly shorter than that in the control group (PFNA: SMD = −3.60, 95% CI [−4.70∼−2.51], P < 0.00001; InterTAN: SMD = −3.20, 95% CI [−5.62∼−0.79], P = 0.009; Overall: SMD = −3.49, 95% CI [−4.47∼−2.52], P < 0.00001). As shown in Table 2 and Fig. 2.

Full weight bearing time

Four studies (Yang et al., 2023; Lin et al., 2022; Wang et al., 2023; Wan et al., 2024) reported the full weight-bearing time after surgery, with the PFBN group showing a significantly shorter duration than the control group did (PFNA: SMD = −1.61, 95% CI [−2.88∼−0.33], P = 0.01; InterTAN: SMD = −2.34, 95% CI [−3.23∼−1.46], P < 0.00001; Overall: SMD = −1.78, 95% CI [−2.86∼−0.70], P = 0.001). Considerable heterogeneity was observed across the studies. As sensitivity analyses (e.g., leave-one-out method) failed to pinpoint a specific source of this heterogeneity, a random-effects model was therefore applied for the meta-analysis. As shown in Table 2 and Fig. 3.

Fracture healing time

A total of 11 studies (Liu et al., 2023; Yang et al., 2023; Fu et al., 2023; Jin et al., 2024; Long & Li, 2022; Ling et al., 2023; Li et al., 2022; Lin et al., 2022; Wu et al., 2023; Jia et al., 2023; Fu et al., 2024) reported fracture healing times. Considerable heterogeneity was observed across the studies. As sensitivity analyses (e.g., leave-one-out method) failed to pinpoint a specific source of this heterogeneity, a random-effects model was therefore applied for the meta-analysis. The results revealed that the fracture healing time in the PFBN group was significantly shorter than that in the control group (PFNA: SMD = −0.56 95% CI [−0.89∼−0.23], P = 0.0008; InterTAN: SMD = −0.76, 95% CI [−1.06∼−0.46], P < 0.00001; Overall: SMD = −0.61, 95% CI [−0.86∼−0.37], P < 0.00001). As shown in Table 2 and Fig. 4.

Postoperative complication rate (safety assessment)

A total of nine studies (Liu et al., 2023; Yang et al., 2023; Jin et al., 2024; Ling et al., 2023; Zhang & Peng, 2022; Wu et al., 2023; Jia et al., 2023; Fu et al., 2024; Zhang, Zhang & Zhang, 2024) reported postoperative complication rates, known as safety assessments. The heterogeneity among the studies was relatively small, and a fixed-effect model was used for analysis. The results revealed that the postoperative complication rate in the PFBN group was significantly lower than that in the control group (PFNA: RR = 0.32, 95% CI [0.18∼0.56], P < 0.0001; InterTAN: RR = 0.58, 95% CI [0.22∼1.54], P = 0.28; Overall: RR = 0.36, 95% CI [0.22∼0.59], P < 0.0001). Thus, to a certain extent, patients with IFF are relatively safer in choosing PFBN treatment. As shown in Table 2 and Fig. 5.

Fracture reduction quality

Eight studies (Liu et al., 2023; Long & Li, 2022; Ling et al., 2023; Li et al., 2022; Wang et al., 2023; Wu et al., 2023; Wan et al., 2024; Jia et al., 2023) reported the fracture reduction quality, that is, the rates of good and excellent reduction quality. The heterogeneity among the studies was relatively small, and a fixed-effect model was used for analysis. The results revealed that the quality of fracture reduction in the PFBN group was greater than that in the control group, and the difference was statistically significant (PFNA: RR = 1.08, 95% CI [1.01∼1.16], P = 0.02; InterTAN: RR = 1.01, 95% CI [0.89∼1.15], P = 0.87; Overall: RR = 1.07, 95% CI [1.01∼1.13], P = 0.02). As shown in Table 2 and Fig. 6.

Sensitivity analysis

In this study, sensitivity analysis was carried out on the above studies with high heterogeneity via leave-one-out and change effect models. The results revealed that all the results were relatively stable and reliable with no significant changes, and no source of heterogeneity was found. The analysis results of the random effects model were acceptable.

Publication bias

The operation duration was used as an example to assess publication bias. The funnel plot, which displays a symmetrical funnel shape, indicates that the risk of publication bias in the included studies was not significant. As shown in Fig. S6.

Findings of this study

This study conducted a systematic review and meta-analysis of the clinical efficacy and safety of PFBN compared with PFNA or InterTAN. Compared with previous related research findings, the following characteristics are presented:

In terms of clinical efficacy, this study indicates that PFBN may have certain advantages in some aspects. Compared with PFNA and InterTAN, PFBN demonstrates superior performance in maintaining the stability of fracture reduction and reducing the time required for fracture healing. It allows for earlier ambulation postoperatively without increasing the risk of internal fixation failure (Atzmon et al., 2022; Jia et al., 2021). This not only aids in patients’ quicker recovery of physical functions and enhances their independence and autonomy in daily life but also has a positive and profound impact on the overall rehabilitation process. These benefits include the prevention of complications and a reduction in mortality rates (Atzmon et al. 2022). This may be attributed to its unique biomimetic design, which better conforms to the anatomical structure of the proximal femur, restores the physiological fulcrum position of the proximal femur, provides a more appropriate stress distribution, and offers more sustained and robust fixation (Wang et al., 2022c; Wang et al., 2022a; Cheng et al., 2023). PFNA and InterTAN may not provide ideal fixation for the lateral wall of the IFF, and the spiral blade of its anti-rotating screw is not supported effectively, which may result in a significant increase in cutting stress. This can result in displacement of the femoral head, followed by the four major challenges of ”retrograde nail, rotation, cutting, and instability”. Subsequently, complications such as failure of internal fixation and varus collapse of the hip may arise (Mavrogenis et al., 2016; Nie et al., 2021; Zhang et al., 2021b; Shi et al., 2021). Since these shortcomings were identified, efforts have been continuously made to explore solutions (Wu et al., 2022; Wang et al., 2022a; Chen et al., 2020b; Chen et al., 2023).

In terms of safety, the amount of intraoperative blood loss in PFBN is relatively lower than that in PFNA, while there is no significant difference compared with that in InterTAN. The incidence of postoperative complications also exhibited the same characteristics. Some studies have also indicated that PFBN has a lower mortality rate than hip replacement does (Sun et al., 2024), suggesting that PFBN may have a potential advantage in reducing surgical risk. However, there is also complexity and specificity between different treatment methods, a characteristic that warrants in-depth analysis. PFBN may enhance internal fixation stability and reduce the risk of complications through its more optimized structural design. By triangulating the tension screws and pressure screws, the fractures of the main pressure bone trabeculae are repaired, the shear forces on the main tension bone trabeculae are reduced (Zhu et al., 2021a; Zhang et al., 2021c), and the center of the force arm is closer to the physiological fulcrum (Lin et al., 2022). This design aligns the forces more closely with the body’s natural mechanics, which can improve the stability of internal fixation and reduce the risk of complications such as loosening or failure of internal fixation (Zhu et al., 2021b). This not only reduces the intraoperative risk but also may have a positive impact on patients’ early postoperative recovery and reduce the incidence of complications. For example, it can lead to a decrease in anemia-related complications, accelerate postoperative weight bearing, reduce postoperative pain, and mitigate inflammatory responses.

Compared with PFNA and InterTAN, PFBN did not significantly differ in terms of surgical duration, incision length, or postoperative hip range of motion (Table 2). These findings suggest that PFBN does not have a clear advantage over PFNA or InterTAN in terms of certain surgical and postoperative recovery indicators and that the overall efficacy of the three internal fixation methods is essentially comparable. As seen from the forest plot, the operative times of PFNA and InterTAN were numerically shorter than that of PFBN, although the difference was not statistically significant. The reason might be that the insertion of transverse support screws has, to some extent, increased the surgical time (Fu et al., 2023; Jin et al., 2024; Long & Li, 2022). Additionally, the surgical duration and length of hospital stay may also vary due to the interactive effects of various other factors, such as the surgeon’s experience and technical skill, the patient’s physical condition, and postoperative care and rehabilitation management. These factors can significantly influence patient outcomes and should be taken into account when evaluating the overall surgical process and recovery. The length of the surgical incision is likely to be primarily determined by the choice of surgical approach rather than the internal fixation device itself. The number of intraoperative fluoroscopies is likely more closely related to the complexity of the surgery and the surgeon’s experience rather than being solely dependent on the type of internal fixation used. Postoperative hip range of motion is more closely associated with factors such as postoperative rehabilitation exercises.

In summary, PFBN has advantages in terms of early postoperative weight bearing time, fracture healing time, postoperative complication rates, length of hospital stay, and intraoperative blood loss. However, in other aspects, the differences between PFBN and PFNA or InterTAN are not significant. This provides an important reference for the selection of clinical treatment plans. On the basis of the patient’s specific conditions and the surgeon’s experience, the unique advantages of PFBN can be weighed against the relative consistency in other aspects to formulate the most appropriate surgical strategy.

Limitations and strengths