© The Authors, published by EDP Sciences, 2025

Introduction

Osteogenesis Imperfecta (OI) is a genetic disorder caused by mutations in the COL1A1 or COL1A2 genes, leading to alterations in type I collagen [1, 2]. Its incidence ranges from 1 in 10,000 to 1 in 20,000 live births [3]. Patients with OI typically experience early-onset recurrent fractures and progressive bony deformities, often requiring realignment osteotomies and internal fixation [4].

Anterograde femoral nailing is the gold standard for femoral deformity correction in OI patients [4]. Since the original description of the Bailey and Dubow telescopic rod system in 1963 [5], numerous implant designs have been introduced to improve fixation and longevity [6, 7]. However, complications remain common, including gluteal muscle weakness, Trendelenburg gait, insufficient correction of the neck-shaft angle (NSA), eccentric nail positioning, femoral head osteonecrosis, nonunion, nail migration, and infection [8–13]. Many of these complications stem from the anterograde approach itself, which poses challenges in achieving optimal implant positioning while preserving surrounding structures.

To address these limitations, we developed a retrograde femoral nailing technique using the Bailey and Dubow telescopic rod system. Originally described by co-author GF in 1989, this technique has been the standard approach for OI patients in our institution. It was designed to optimize nail positioning along the femur, improve deformity correction, and reduce complications associated with anterograde nailing, particularly those related to hip mechanics, growth plate disruption, and femoral head vascularization.

Despite the long-standing use of this technique in our institution, no studies have systematically evaluated its outcomes compared to anterograde nailing. This study aims to fill this gap by assessing the clinical and radiographic outcomes of retrograde femoral nailing in OI patients. By doing so, we seek to determine whether this approach offers a viable alternative with fewer complications and improved deformity correction.

Materials and methods

Study design

This retrospective study was approved by the ethical committee of our institution. We reviewed medical records and radiographs of OI patients who underwent retrograde intramedullary nailing with Bailey and Dubow telescopic rods for femoral fracture or deformity between 2004 and 2019 at a pediatric university hospital. All surgeries were performed by three senior orthopedic surgeons.

Patient population

A total of 31 patients met the inclusion criteria, which required that they had no prior femur surgeries and underwent retrograde femoral nailing for fracture or deformity. The minimum follow-up period was 48 months or until an adverse event, such as a fracture around the nail requiring exchange, occurred.

Among these patients, 13 (42%) were female, with a mean age at surgery of 3.1 years and a mean follow-up of 2.7 years. The identified genetic mutations were primarily COL1A1 and COL1A2. Table 1 summarizes the demographic and clinical data.

Hardware description

The implant used was a modified Bailey and Dubow telescopic nail (Medicalex). The nail consists of a female component (Figure 1A) with threading at one end to ensure secure attachment to the drill bit, facilitating insertion into the femur. A T-piece (Figure 1B) is secured by screwing it into the female component, and forceps are used to tightly clamp it, preventing loosening. Additionally, the male component (Figure 1A) also ends in a T-shaped piece. The nail is available in diameters of 3.5, 4, 4.5, 5, and 6 mm.

Figure 1 Description of the telescopic nail utilized in this study. A: Red arrow: The female rod; Blue arrow: The male rod. B: The T-piece (black arrow) is secured by screwing it into the female rod (Blue star). C: The drill bit.

The primary instrument for this retrograde technique is a long drill bit (Figure 1C). One end is reamed 0.2 mm larger than the nail diameter, allowing for preparation of the medullary canal and precise, perpendicular osteotomies aligned with the knee joint. The opposite end of the bit is threaded to secure the female part of the nail, guiding its insertion. Additional impactors for the T-pieces complete the instrument set. Custom nail planning is based on an AP or lateral X-ray of the femur, using a radiopaque index for accurate magnification assessment.

Surgical technique

The procedure was performed under general anesthesia with the patient in the lateral decubitus position. The affected limb was prepared and draped up to the hip. The image intensifier was positioned horizontally to allow full visualization of the femur.

The drill bit is introduced at the knee through a percutaneous transpatellar tendon approach, using a T-handle. After confirming correct positioning on both anterior and lateral views with the image intensifier, the drill bit is advanced into the center of the intercondylar notch. This manual drilling is performed perpendicular to the joint line in both anterior and lateral views (Figure 2A). During drilling, it is crucial to remain perpendicular to the distal femoral joint line, performing osteotomies as needed (Figure 2B). These can be done either via percutaneous osteoclasis or through a lateral open approach. For surgical exposure, dissection and periosteal stripping are primarily performed with electrocautery to limit bleeding, especially in severe cases. The osteotomy is performed with an oscillating saw to prevent splintering of the bone and to ensure a precise osteotomy.

Figure 2 This figure illustrates the surgical technique. A: Manual drilling is performed perpendicular to the joint line in both anterior (left) and lateral (right) views. B: During drilling, it is crucial to remain perpendicular to the distal femoral joint line, performing osteotomies as needed. These can be done either via percutaneous osteoclasis or through a lateral open approach (this figure was created by the authors to be used in this article). C: The exit point for the drill bit is at the femoral neck, allowing for valgus fixation to prevent coxa vara deformity and the risk of lateral nail exit or femoral neck fracture. To facilitate this, a subtrochanteric osteotomy may be performed. D: The male component is introduced percutaneously through the knee and impacted into the distal femoral epiphysis.

The exit point for the drill bit is at the femoral neck, allowing for valgus fixation to prevent coxa vara deformity and the risk of lateral nail exit or femoral neck fracture. To facilitate this, a sub trochanteric osteotomy may be performed (Figure 2C). The drill bit is then retracted percutaneously at the level of the gluteal region. The female component of the telescopic nail, with the male component inside, is screwed onto the drill, guiding it to the proximal exit point. The T-tool is screwed and secured in the female component to prevent unscrewing, and the entire assembly is impacted at the superior border of the femoral neck. The male component is introduced percutaneously through the knee and impacted into the distal femoral epiphysis (Figure 2D).

Postoperatively, the limb was immobilized with a hip spica cast for 4 weeks in cases of percutaneous osteotomy and 6 weeks for open osteotomy, particularly in valgus correction cases. This was followed by intensive physiotherapy focusing on hip and knee range of motion. Radiographs were obtained at 6 weeks, 12 weeks, 24 weeks, 48 weeks, and annually thereafter.

Outcome measures

Clinical outcomes

Knee range of motion (ROM) was assessed preoperatively and at the last follow-up. A ROM between 0° and 130° was considered normal [15].

Pain levels were evaluated using the Visual Analog Scale (VAS) [16].

Radiological outcomes

Radiographic assessments were performed preoperatively, after cast removal (4–6 weeks), and at the last follow-up.

The neck-shaft angle (NSA) was measured on AP radiographs, with values between 120° and 140° considered normal. Coxa vara was defined as an NSA < 120°, while coxa valga was defined as an NSA > 140° (Figure 3A) [17].

Figure 3 This figure illustrates the measured angles which are utilized as radiological outcomes. A: The neck-shaft angle (NSA) measured on AP radiographs. Normal range: 120–140 degrees. B: mLDFA was defined as the angle measured between the mechanical axis of the femur and the knee joint line. Normal range: 85–90 degrees. C: The femoral deformity angle (DA) was defined as the angle measured between two lines drawn along the bone shaft axis and intersecting at the apex of the angulation on AP (left) and Lateral (right) view.

The mechanical lateral distal femoral angle (mLDFA) was assessed, with normal values ranging between 85° and 90° (Figure 3B) [18]. The deformity angle (DA) was determined by measuring the intersection of two shaft axis lines on both AP and lateral views (Figure 3C) [19].

The positioning of the nail was examined at the hip and knee joints on AP and lateral radiographs. The distal positioning was deemed optimal when the male rod passed through the middle compartment of the epiphysis (Figure 4A), while the proximal positioning was ideal when the female rod exited through the femoral neck (Figure 4B). Any deviation from these ideal positions was categorized as eccentric. The number of osteotomies required for deformity correction was also recorded based on postoperative X-ray analysis.

Figure 4 This figure shows optimal positioning of the hardware in the proximal and distal femur. A: In order to define the optimal position of the nail distally, we virtually divided the distal femoral epiphysis into three compartments. A good positioning was defined as the distal end of the male rod passing through the middle compartment of the epiphysis on AP (left) and lateral (right) X-rays. B: Proximally, the position was considered optimal when the female rod exited through the femoral neck.

Statistical analysis

Statistical analyses were conducted using R software (version 3.5.0), accessed on September 5, 2024, at https://www.R-project.org. Quantitative variables were expressed as means and standard deviations, while categorical variables were presented as frequencies and percentages. Comparisons between two groups were performed using the Mann–Whitney test, and comparisons among more than two groups were carried out using the Kruskal–Wallis test. A p-value of less than 0.05 was considered statistically significant.

Results

Clinical outcomes

Postoperatively, knee range of motion (ROM) was preserved in all patients. Among the 34 femurs without fractures at the time of surgery, preoperative ROM was within the normal range of 0 to 130 degrees. However, ROM assessment was not possible for the 20 femurs with fractures preoperatively, so these were excluded from analysis.

Postoperative pain, assessed using the Visual Analog Scale (VAS), showed a significant reduction over time. The mean VAS score was 6 in the early postoperative period and decreased to 0 at the final follow-up, indicating complete pain resolution. These clinical outcomes are summarized in Table 2.

Radiological outcomes

The correction of femoral deformity was reflected in significant improvements in radiographic parameters. The NSA increased from a preoperative mean of 132 degrees to 142 degrees immediately postoperatively. At the final follow-up, the NSA was measured at a mean of 138 degrees. The difference between preoperative NSA and early postoperative NSA was statistically significant (p = 0.0001), as was the difference between preoperative NSA and final follow-up NSA (p = 0.018). However, the difference between early postoperative and final follow-up NSA was not statistically significant (p = 0.07).

Similarly, the mLDFA showed a significant correction, decreasing from a preoperative mean of 101 degrees to 89 degrees postoperatively. At the final follow-up, the mean mLDFA was 91 degrees. The difference between preoperative and early postoperative mLDFA was highly significant (p = 8.4e–9), as was the difference between preoperative and final follow-up mLDFA (p = 4.3e–7). However, no significant change was observed between the early postoperative and final follow-up mLDFA values (p = 0.49).

Femoral deformity was also evaluated using the deformity angle (DA), which was 29 degrees on AP X-rays and 40 degrees on lateral X-rays preoperatively. Postoperatively, no residual deformity was observed in any patient on both AP and lateral X-rays through the final follow-up.

Implant positioning was satisfactory in the majority of cases. On postoperative AP X-rays, the nail position was deemed satisfactory in 48 out of 54 femurs (89%), and on lateral X-rays, it was satisfactory in 50 out of 54 femurs (93%). The mean number of osteotomies required to achieve proper alignment of the femoral deformity was 1.6 per femur. A summary of these radiological outcomes is presented in Table 2.

Complications

Complications were observed in a subset of patients. The most common complication was an inability of the nail to telescope, occurring in 28 out of 54 femurs (52%). Proximal hardware migration was reported in 7 out of 54 femurs (13%). Importantly, no cases of intraarticular migration into the knee joint were observed. Additionally, there were no recorded instances of growth arrest, aseptic necrosis, nonunion, or infection.

Discussion

Osteogenesis imperfecta (OI) is a rare genetic disorder characterized by bone fragility, recurrent fractures, and progressive deformities [1, 2]. The primary goal of surgical management in these patients is to correct deformities, stabilize fractures, and improve functional mobility while minimizing complications. The main finding of this study is that retrograde nailing of the femur in patients with OI achieved effective deformity correction and proper fracture alignment. The procedure ensured optimal positioning and secure anchoring of the nail at both ends of the femur. Notably, there was no risk of growth arrest in growing children.

The primary limitation of this study is its retrospective design and the absence of a control group, which limits the ability to establish causality and compare outcomes with a standardized baseline. However, despite this limitation, our study population is notably heterogeneous, encompassing a diverse range of OI patients across different age groups, varying levels of walking and functional abilities, and multiple OI phenotypes. This diversity adds to the generalizability of the findings but also introduces variability that could affect the outcomes. Another limitation of this study is that we did not analyze the results following the replacement of the first nail. Additionally, our focus was not on disease-specific factors inherent to osteogenesis imperfecta (OI), but rather on the surgical technique itself.

The clinical outcomes were satisfactory in our study population with an improvement in motor function and the achievement of a normal ROM following surgery. Conversely, previous studies [20, 21] reported that retrograde femoral nailing has been demonstrated to be a safe alternative to anterograde nailing, although it is known to cause knee pain, which is reported as one of the most common adverse effects of retrograde nailing, with incidence rates as high as 70%. In our series, no knee pain was reported during long-term follow-up. We believe this is likely due to the small diameter of the telescopic nail, femoral notch enter point and the high potential for fibrocartilage tissue formation in children. Moreover, the drill bit pass through only once, with no reaming required, eliminating the risk of reaming debris entering the knee joint, as the nail is directly connected to the drill bit. Knee pain was not evaluated at the 6-week follow-up, as spica cast removal is often associated with knee and hip apprehensions in children. No knee pain was reported at the last follow-up.

We hypothesize that restoring the proper mechanics of the hip and knee joints will reduce the risk of pathological fractures and enhance functional outcomes.

Regarding the hip joint, we realized corrective osteotomies of the coxa vara with a target NSA of more than 135 degrees in order to prevent deformity recurrence over time. To achieve this goal, the nail was fixed proximally through femoral neck. We measured the NSA as an indicator of optimal correction and as an indicator of the static and mechanical properties of the joint. Our findings indicate that this procedure results in normal to relatively increased NSA, which is crucial in preventing postoperative complications and unfavorable outcomes. As per Aarabi et al. [12], the clinical consequences of coxa vara are well established; a reduction in the neck–shaft angle shortens the abductor lever arm, resulting in abductor insufficiency and a positive Trendelenburg sign. Notably, telescopic nails are straight; therefore, in cases of antegrade nailing with a greater trochanteric entry point, aiming for the center of the distal femoral physis can result in fixation of the femoral neck in varus, increasing the risk of worsening the deformity. Conversely, if the femoral neck is positioned in valgus, the distal nail may end up in the medial femoral condyle, raising the risk of epiphysiodesis.

Furthermore, the retrograde telescopic nailing procedure does not compromise the gluteal muscles, which are vital for optimal hip function.

In our study, no cases of femoral head osteonecrosis were observed, likely due to several key aspects of our surgical technique. By passing through the femoral neck without reaming and advancing the drill bit solely through the neck, we minimized mechanical and thermal stress on the surrounding bone and vascular structures. This approach helped preserve the femoral head’s blood supply by preventing thermal effects and avoiding ischemia that could result from motorized drilling. Additionally, the use of a direct connection between the nail and the drill bit further reduced the risk of vascular disruption and thermal damage. These procedural factors likely contributed to the preservation of femoral head vascularity, potentially explaining the absence of osteonecrosis in our patient cohort.

The most widely recognized surgical method for treating femoral fractures or deformities in patients with OI is based on the antegrade femoral nailing technique [4–6]. However, this method is associated with several complications, including Trendelenburg gait, persistent hip pain, proximal rod migration, difficulties in correcting distal femoral deformities, and eccentric positioning of the distal end in the knee due to the trochanteric entry point, which can result in growth disturbances, rod bending, and fractures [8, 9, 13, 22, 23]. A crucial finding of this study was the successful achievement of centered nail positioning on both AP and lateral X-rays in the vast majority of patients.

To the best of our knowledge, this is the first study to describe the retrograde nailing technique using a telescopic rod in patients with OI. Previous research has demonstrated improved motor skills and walking abilities, along with satisfactory correction of deformities, using anterograde nailing [4–7, 22]. However, these studies have reported various complication rates and have questioned the efficacy of these surgical interventions in enhancing functional status and reducing the risk of fractures [8, 9, 24]. In our study, postoperative complications such as intraarticular migration, growth arrest, and aseptic necrosis of the femoral head were absent, and the only reported complications were related to the hardware utilized in this study population.

Furthermore, retrograde nailing demonstrates greater efficiency and ease in correcting distal deformities, although it may lead to distal femoral growth arrest. Our findings suggest that the retrograde technique achieves satisfactory correction. The correction was easier from distal to proximal, allowing for precise alignment. Moreover, the drill bit serves both as a correction tool and as a guide for nail insertion.

It is crucial to highlight the significance of an accurate entry portal, which ensures optimal implant placement, thereby facilitating effective correction of distal deformities and enabling subsequent proper osteotomies and alignment. Proper alignment of the nail within the center of the knee and femoral neck is essential for stability, achieving satisfactory correction, and preventing growth plate arrest or disturbances. In our study, no growth disturbances were observed and good correction was achieved (Figure 5).

Figure 5 Pre-operative (A), post-operative (B) and 6-year-follow-up (C). X-rays of one of this study population patients showing no growth arrest or aseptic femoral head necrosis along as good nail positioning and femur alignment.

Previous studies [9, 19] have reported that revisions were necessary in 13–53% of patients who received anterograde telescopic nails for femoral fractures, due to rod bending and rod migration. In our series, we achieved excellent anchoring with only 13% migration. However, we observed 52% inability to telescope of the nail. Additionally, we observed no rod bending or non-traumatic femoral fractures at the last follow-up.

The lack of rotational control of the femur and the thin diameter of OI long bones make intramedullary rodding insufficient for providing optimal fixation. Therefore, spica casting is crucial to counteract this mechanical disadvantage and control rotational forces. It also prevents malrotation of bony unions while reducing the postoperative pain. It is important to note that mechanics alone will not prevent fractures, as OI is a congenital disorder caused by mutations in the COL1A1 or COL1A2 genes [1, 2], resulting in abnormal collagen cross-linking and a decrease in type I collagen. This collagen disorder increases the risk of fragility fractures and bone deformities. Therefore, a multidisciplinary approach, including medical treatment with bisphosphonates, is necessary to help reduce fracture risk [25].

In conclusion, the present study demonstrated that retrograde femoral nailing utilizing the Bailey and Dubow telescopic rod system is a safe and effective method for managing fractures and deformities in patients with OI. Satisfactory clinical and radiological outcomes were achieved while no major complications were reported. Modifying and adapting the Bailey and Dubow nail to minimize hardware-related complications has the potential to improve clinical outcomes in the OI population.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

The authors declared no conflicts of interest.

Data availability statement

No data allowing patient identification is published.

Author contribution statement

Author 1: Contributed to the study design and article writing.

Author 2: Participated in article writing, data collection, data analysis and critical feedback.

Author 3: Participated in study design and critical feedback.

Author 4: Participated in data collection.

Author 5: Participated in study design and critical feedback.

Author 6: Participated in data collection, study design and provided critical feedback.

Ethics approval

Referenced under the number: 2023-11-09. IRB Number: IORG0010044.

Informed consent

Informed consent was obtained from all patients included in this study.

References

All Tables

All Figures

	Figure 1 Description of the telescopic nail utilized in this study. A: Red arrow: The female rod; Blue arrow: The male rod. B: The T-piece (black arrow) is secured by screwing it into the female rod (Blue star). C: The drill bit.
In the text

Figure 2 This figure illustrates the surgical technique. A: Manual drilling is performed perpendicular to the joint line in both anterior (left) and lateral (right) views. B: During drilling, it is crucial to remain perpendicular to the distal femoral joint line, performing osteotomies as needed. These can be done either via percutaneous osteoclasis or through a lateral open approach (this figure was created by the authors to be used in this article). C: The exit point for the drill bit is at the femoral neck, allowing for valgus fixation to prevent coxa vara deformity and the risk of lateral nail exit or femoral neck fracture. To facilitate this, a subtrochanteric osteotomy may be performed. D: The male component is introduced percutaneously through the knee and impacted into the distal femoral epiphysis.

In the text

Figure 3 This figure illustrates the measured angles which are utilized as radiological outcomes. A: The neck-shaft angle (NSA) measured on AP radiographs. Normal range: 120–140 degrees. B: mLDFA was defined as the angle measured between the mechanical axis of the femur and the knee joint line. Normal range: 85–90 degrees. C: The femoral deformity angle (DA) was defined as the angle measured between two lines drawn along the bone shaft axis and intersecting at the apex of the angulation on AP (left) and Lateral (right) view.

In the text

Figure 4 This figure shows optimal positioning of the hardware in the proximal and distal femur. A: In order to define the optimal position of the nail distally, we virtually divided the distal femoral epiphysis into three compartments. A good positioning was defined as the distal end of the male rod passing through the middle compartment of the epiphysis on AP (left) and lateral (right) X-rays. B: Proximally, the position was considered optimal when the female rod exited through the femoral neck.

In the text

	Figure 5 Pre-operative (A), post-operative (B) and 6-year-follow-up (C). X-rays of one of this study population patients showing no growth arrest or aseptic femoral head necrosis along as good nail positioning and femur alignment.
In the text