1 Intro

During the past couple of decades, the use of telescoping intramedullary lengthening nails (ILNs) has advanced the field of limb lengthening and deformity correction beyond many of its formerly limiting factors.^1–4 These were primarily related to conventional external fixators, which initially suffered from significant complication rates, including pin site infection, deep sepsis, neurovascular injury, joint stiffness and contracture.^2,5 ILNs have since become the mainstay treatment offering for many limb lengthening surgeons, particularly in the femur.^4–6 As with any medical device, and perhaps most particularly those that are intended for implantation, it is crucial to grasp how the physical properties of ILNs interact with the surrounding physiological tissues in order to properly plan procedures, predict outcomes, and otherwise optimize patient care. Lengthening nails are uniquely affected by multidirectional forces putting them at risk of elastic and plastic deformity. Finite element analysis tells us that these forces encompass such load-sharing conditions as created by kinesthesis, such as adductor muscle motion exerting superomedial forces on the femur and the embedded nail,⁷ compressive forces exerted by gravity and the patient's weight,⁸ or a complex combination of both. On the other hand, studies have shown that ILN diameter is directly related to the risk of screw breakage, with smaller diameter nails being more prone to fatigue-induced failure of the interlocking screws.⁹ Moreover, and most prescient to the line of inquiry herein, nail diameter was found to play a pivotal role in the device's susceptibility to plastic deformity. In a laboratory study, Maai et al.¹⁰ found that the nail with the smallest diameter offered the least resistance against plastic deformity. Corroborating findings are echoed in other studies as well.^11,12 Bending failure during lengthening or consolidation was observed to be the most common reason for nail failures.¹³ In one series bending of more than 5° was deemed to necessitate intervention, such as revision surgery.¹³

For these reasons, newer lengthening nails like the PRECICE nail carry best practice recommendations of very limited weight bearing until consolidation where patients must avoid full weight-bearing throughout the entire lengthening phase. No more than 20 % or 30 lbs of the patient's body weight should be loaded onto the leg with the implanted PRECICE nail.¹⁴ This has not been the trend for previous ILNs like the Albizzia or the Betzbone, that allowed for toe-touch bearing after surgery with full weight bearing after lengthening,¹⁵ or full postoperative weight bearing, respectively.¹⁶ Direct comparison regarding safety and efficacy has not been done and conservative guidelines were adhered.

This study aims to investigate the degree of nail bending and plastic deformity of the antegrade femoral PRECICE nail in patients.

2 Methods

537 intramedullary nail assisted femoral limb lengthening were retrospectively reviewed. 130 antegrade PRECICE nails in 100 patients were selected after accommodating for the exclusion criteria (Flowchart). We excluded patients that underwent concomitant osteotomies or tibial lengthening, as well as those who had malunion, non-union, mechanical failure, or revision surgery for any reason. Radiographs that were unclear or misaligned were deemed unusable, leading to patients being either requested to undergo a repeat imaging session or excluded from the study. A well-aligned anterior-facing patella served as a landmark for obtaining a high-quality AP X-ray of the lower limb. The average age at the time of surgery was 20.5 years, with a range from 18 to 59. All inserted nails were inserted either through the greater trochanter (66 cases) or piriformis (64 cases).

Preoperative, postoperative (at end of lengthening), approximately 3 months postoperatively, and at end of consolidation (around 6 months post op), long-leg standing X-rays were taken, in case of delayed union, postoperative end of lengthening xrays were used instead. Consolidation, signifying “full healing,” was defined by the radiographic confirmation of bone bridging at the osteotomy site. All X-rays underwent calibration using a 1-inch magnification ball at the bone level, administered by a certified technician and assessed by appropriately trained personnel. Visual evaluations comparisons were performed for pre- and post-operative X-rays. In order to assess intra- and interobserver reliability, a second round of all radiologic assessments was conducted one week later. The intraclass correlation coefficient (ICC) for intra-observer reliability yielded a value of 0.973, while the ICC for interobserver reliability was 0.913. A P value below 0.05 was considered statistically significant.

3 Results

3.3

3.3 Nail characteristics and findings

A total of 130 nails were evaluated. A slight majority of the nails, (52 %), were inserted at the tip of the greater trochanter, whereas 48 % were inserted via the piriformis region (Table 3). 3 different diameters of nails were available, 8.5 mm (13 %), 10.7 mm (60 %)and 12.5 mm (27 %). The ratio of patient weight (in kg) to nail diameter (in mm) was computed, yielding a mean value of 5.91 kg/mm (SD 1.34), and ranging from 2.92 kg/mm to 9.64 kg/mm (Table 4). 14 patients had nail bending of equal or more than 5° at consolidation or end of lengthening, these patients had all underwent bilateral concomitant limb lengthening except for 1, while 3 of them had 12.5 mm nails, 10 had 10.7 mm nails inserted and 1 had an 8.5 mm nail inserted, which equates to 8.5 % of 12.5 nails, 12.8 % of 10.7 nails, 6 % of 8.5 nails (Table 9)

Table 3 Nail distribution by insertion site.

Nail insertion site	numbers of nails
Tip of greater trochanter	63
Piriformis	69

Table 4 Nail characteristics and weight over diameter ratio.

Nail Characteristics	Mean	St. Dev	Min	Max
Diameter	10.9	1.2	8.5	12.5
Weight(Kg)/Diameter(mm)	5.9	1.3	2.9	9.6

Analysis of nail diameter revealed a noteworthy distribution across varying dimensions. Approximately 60.9 % (79) of the nails had a diameter of 10.7 mm, accompanied by an average nail bend of 2.41° (SD 2.1). In contrast, 26.6 % of ILNs had a larger diameter of 12.5 mm, with a lower average nail bend of 1.8°. A smaller percentage (approximately 12.5 %) had the thinnest nail diameter of 8.5 mm, which correlated with a mean nail bend of 2.17° (Table 5).

Table 5 Mechanical access deviation, Pre-op, Post-op at consolidation, and overall shift.

MAD	Mean	St. Dev	Min	Max
Pre-Op Lateral	7.6	6.06	0	25.7
Pre-Op Medial	9.1	8.5	0	38.5
Post-Op Lateral	7.8	5.4	1	27
Post-Op Medial	10.4	9.2	1.1	37
Lateral MAD shift	3.8	3.7	0.4	18.8
Medial MAD shift	5.5	3.6	0.6	15.7

Among the quartiles of nail bending results for our cohort, distinct patterns emerged regarding associated weight metrics and their ratios. In the first quartile (Q1) the corresponding mean patient weight and the weight-to-diameter ratio were 59.5 kg and 5.56 kg/mm, respectively. The second quartile (Q2), which was characterized by a an increase in mean patient weight (64.5 kg) and weight-to-diameter ratio (5.74 kg/mm). In the third quartile (Q3), with mean patient weight advancing to 66.7 kg and the weight-to-diameter ratio rising noticeably to 6.37 kg/mm. Lastly, in the fourth quartile (Q4), where patient mean weight is 67.1 kg weight-to-diameter ratio had stabilized at 6.18 kg/mm (Table 6).(see Table 7–9)

Table 6 Nail diameter distribution and average nail bend for each.

Nail Diameter (mm)	Percentage of cohort	Avg
8.5	12.5	2.17
10.7	60.9	2.41
12.5	26.6	1.8

Table 7 Nail bend quartiles, with mean weight and mean weight over nail diameter (Q1 <0.75, Q2<2.4,Q3 >3,Q4 <3.

Nail bend quartiles	mean weight kg	mean weight kg/Nail diam mm
Q1	59.5	5.56
Q2	64.5	5.74
Q3	66.7	6.37
Q4	67.1	6.18

Table 8 Multiple linear regression for the effect of femoral lengthening and weight over nail diameter ratio and if bilateral or not on nail bend.

	Dependent variable:
	Nail.Bend
`Weight(Kg)/Diameter(mm)`	0.429b
	(0.193)
Bilateral	2.374c
	(0.505)
Constant	−1.111
	(1.187)
Observations	72
R2	0.283
Adjusted R2	0.262
Residual Std. Error	2.143 (df = 69)
F Statistic	13.595c (df = 2; 69)

b p < 0.05.

c p < 0.01.

Table 9 Details of patients that had a nail bend of >5°.

Degrees of Nail Bend	Nail Diameter mm	Weight(Kg)/Diameter(mm)	Lengthening (cm)	Bilateral nail	Delayed healing
5	8.5	5.08	5.3	No	Yes
5	11.5	5.91	7.7	Yes	No
5	10.7	5.08	8.3	Yes	No
5	10.7	8.41	5.1	Yes	No
7	12.5	4.72	7.7	Yes	Yes
7	12.5	4.72	8.3	Yes	Yes
7	10.7	5.72	8	Yes	No
7	10.7	5.93	2	Yes	No
8	10.7	6.02	4.7	Yes	No
8	10.7	6.02	4.9	Yes	No
8	10.7	5.93	1.4	Yes	No
8	10.7	7.84	6.9	Yes	No
9	10.7	5.72	7.9	Yes	No
9	10.7	7.84	7.2	Yes	No
Mean		6.1	6.1

Furthermore, bilateral concomitant lengthening lead to significant nail bending (P < 0.05) when compared with unilateral lengthening procedures, with a mean of 3.7° (SD 2.9) ranging between 0 and 9 for the bilateral subgroup and 1.3° (SD 1.14) ranging between 0 and 5 for the unilateral subgroup. With a difference of 2.4 degrees of bend between both groups.

Results of the multiple linear regression analysis investigating the determinants of nail bend are presented in Table 6. Two predictor variables, weight(kg)/diameter(mm) and if the patient had bilateral or unilateral limb lengthening were included in the model.

A significant positive association was observed between nail bend and weight(kg)/diameter(mm) (P < 00.05), indicating that for each additional kilogram per diameter(mm), the expected increase in nail bend is 0.43°. Along with an increase of 2.3 degrees of nail bend attributed to undergoing concomitant bilateral limb lengthening (P < 00.05).

The overall model was statistically significant (F-statistic = 13.6, P < 00.05), explaining approximately 26 % of the variance in nail bend.(Plot 1, Plot 2)(seeFig. 4).

Close

Image 1 showing the effects of nail bending on MAD, with Pre-MAD of 0 mm (image on the left) and MAD of 4 mm medial on post-op image on the left with a nail bend of 7°.

Export to PPT

Close

Plot 1 Scatter Plot showing nail bend as a function of Weight/Diameter. Red line is the overall trend in nail bend with increasing weight over nail diameter ratio.

Export to PPT

Close

Plot 2 boxplot showing the difference in Nail bend between unilateral and bilateral limb lengthening cases.

Export to PPT

Close

Chart 1 Flowchart for exclusion criteria.

Export to PPT

Three patients had delayed union documented in their charts, all of them had a nail bend of equal or more than 5° documented on their post-lengthening xrays taken directly after lengthening phase was over.

Whereas for MAD shift no statistically significant correlation was appreciated between MAD shift and nail bend (Table 5, Plot 3, Plot 4, Image 1).

4 Discussion

This study investigated nail bending outcomes for the PRECICE ILN as utilized in femoral lengthening to correct stature or limb length discrepancy, as well as additionally evaluating potentially contributory variables therein. Of these factors, weight to nail diameter ratio (kg/mm) and concomitant bilateral nail insertion demonstrated a statistically significant linear relationship with nail bending. No statistically significant relationship was appreciated between nail bend and MAD shift reflecting minimal effect of nail bend on coronal limb alignment (Image 1).

Previous research by Lee et al. reporting on the PRECICE 1 and 2 nails (NuVasive, Inc.) considered nail bending to be significant at 2 or more degrees.¹⁸ With an average of lengthening of 34 mm using PRECICE 2, almost half (49 %) of the nails in the cohort had a bend reaching that benchmark or greater, considerably surpassing the 12.5 % noted in those findings.¹⁸ Whereas Lee et al.¹⁸ included both tibial and femoral lengthening without distinguishing complication differences between the groups, the study focused on femoral lengthening only. Whereas tibial anatomic and mechanical axes are usually parallel, this is not the case for the femur, where a significant difference between mechanical and anatomic axes is appreciated with the AMA angle, this structural difference between both bones could explain the higher rates of nail bend perceived in femurs when compared to tibias. Forces may be exerted on the femoral nail that are not parallel to its axis, leading to more nail bending, and perhaps explaining the relatively higher rates of nail bending observed in the cohort. Hlukha et al.,¹³ looking at PRECICE nails inserted in femurs, humerus and tibias, defined bending of more than 5° to be a mechanism failure possibly necessitating revision surgery. 14 nails fit the criteria, all of which, except for 1, were inserted in patients doing bilateral limb lengthening. Furthermore, bilateral limb lengthening was a statistically significant risk factor of nail bending, our study found that concomitant bilateral limb lengthening increases nail bend on average by 2.3°. This could be attributed to the fact that patients with bilateral limb lengthening will have to exercise their full body weight on the nail while standing, whereas unilateral nail insertion gives the ability to ambulate on crutches and stand on one leg for activities of daily life.

Patient compliance with partial and non-weight bearing restrictions has also been implicated in bending failure.^13,19,20 Likewise, we suspect the potential influence of muscular rotational pulling forces on the observed nail bend. Specifically, the adductor longus and brevis muscles are considered significant contributors to femoral nail bending. This assumption is rooted in their anatomical attachments either at the osteotomy site or in close proximity to it, causing bending tension whenever these muscles contract. This underscores the intricate interplay between musculature and the mechanical dynamics of implanted devices (ILNs) in limb lengthening procedures, providing a nuanced perspective on the factors contributing to nail bend in the context of surgical outcomes. Further investigation into the precise biomechanical forces exerted by these muscles could shed light on ways to optimize surgical techniques and minimize the risks of elastic and plastic deformity.

While a direct relationship between nail bend and nail diameter has been described in the literature,^10,18 our results offer an important refinement, suggesting a direct relationship between weight to nail diameter ratio and nail bending as well. We are the first study to establish such a relationship. This introduces an argument for considering not only the patient's femoral canal width when choosing nail size, but also, factoring in the patient's weight. Moreover, a recent meta-analysis found that reamed intramedullary nailing for femur fracture patients may lead to significantly lower risks of non-union, screw failure, implant exchange, and dynamization, without increasing operative complications.⁹ Although bone reaming has been shown to be beneficial in fracture patients, aggressive reaming compared to minimal reaming showed no statistically significant benefit in terms of healing and return to work.^7,21Together with these findings, our results stress the importance of adequate bone reaming that allows for a larger diameter nail, which in turn facilitates healing and decreases the chances of nail bending. Adequate reaming has shown to be beneficial for healing in fracture patients, nonetheless, previous studies do not discuss the optimal degree of reaming needed. Furthermore, the literature discusses reaming in fracture patients,^22–24 but no study has looked at the benefits or degree of reaming for limb lengthening. Furthermore, our study showed that 10.7 mm nails had the highest percentage of nail bend of more than 5°, while 12.7 mm nails had second and 8.5 had the least number of nails with bend more than 5°, these results strengthen our hypothesis that nail bend is more affected by the ratio of nail diameter over patient weight rather than solely nail diameter.

Our study did not find a relationship between nail bending and MAD shift, since 9 degrees of nail bend was the highest in our cohort, image 1 shows a clear example on how even with severe nail bend MAD shift is still minimal. Furthermore, all of our delayed union patients had a nail bend of more than 5°, although not statistically significant, increase in delayed union in femurs with high nail bend may be an indication of inadequate stability.

Although our model explained 26 % of the variance in nail bending for our cohort, its purpose was primarily to establish a relationship between novel variables and nail bend, expanding beyond current literature citing patient non-compliance as the main factor.¹⁸ This model indicates that the primary factors in bending are not measured in this set. Likely factors include patient compliance, activity level and passive muscle forces.

This data advances our understanding of the interrelation between the nail biomechanical profile and the patient's physical attributes, offering important implications for limb lengthening. In conclusion, bilateral limb lengthening, increase in weight over nail diameter ratio and concomitant bilateral limb lengthening are all risk factors that lead to increased nail bending. Nonetheless, our study indicates that other factors are also significant, though not yet defined. On the other side, our cohort had only 3 delayed union cases, no nonunion or malunion and no statistically significant MAD shift, strengthening the fact that intramedullary limb lengthening is generally a safe and effective procedure.

Source of funding

No funding was received in support of this investigation.

Potential conflicts of interest and funding sources

Dr. McClure's disclosures:

Biocomposites: Other financial or material support

MHE Coalition: Other financial or material support

Novadip: Paid consultant; Research support

Orthofix, Inc.: Other financial or material support; Paid consultant

OrthoPediatrics: Other financial or material support; Paid consultant

Stryker: Other financial or material support

Zimmer: Other financial or material support

Dr. Assayag's disclosures:

Johnson & Johnson (unpaid consultant)

OrthoPediatrics (unpaid consultant)

Orthofix (unpaid consultant)

Consent

No guardian or Patient consent required (waived my institutional review board)

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Lifebridge health.

Funding

None.

CRediT authorship contribution statement

Akram Al Ramlawi: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Michael Assayag: Conceptualization, Supervision, Project administration. Philip McClure: Conceptualization, Supervision, Project administration, Writing – review & editing.