Do Anatomical Differences of the Volar Rim of the Distal Radius Affect Implant Design? A Three-Dimensional Analysis of Its Anatomy and Need for Personalized Medicine

Abstract

The distal radius, one of the frequent sites of upper extremity fractures, includes unique anatomy referred to as the volar rim. Few studies have addressed its interindividual differences. Additionally, implants for osteosynthesis must match the anatomical structures to prevent soft tissue invasion, but no implants have focused on that so far. In this study, three-dimensional surface models were created from CT images of 101 cases. Analysis of the distal radius, including the volar rim anatomy, was performed to design plates to minimize the discrepancy between the bone anatomy and the implant. The results showed that there were considerable interindividual differences in the morphology of the distal radius, particularly in the degree of palmar protrusion of the volar rim. A moderate correlation between the width of the distal radius and the shape of the volar rim was demonstrated. Considering that variations in plate width are available for treatment of normal distal radius fractures and that the shape of the volar rim changes in correlation with the width of the bone, we infer that simply adding volar rim shape information to the current variations should suffice in preventing complications. We conclude that individualized design according to the shape of the volar rim is unnecessary in fracture surgery.

1. Introduction

The anatomical morphology of bones is determined by various parameters such as height, weight, sex, age, nutritional status, and systemic diseases. The diaphysis is generally similar in shape [1], although there are variations in width and length [2], and the epiphysis and metaphysis near the joints show variations in three-dimensional shapes between individuals.

The volar rim, a characteristic morphology of the distal radius, is a structure with significant individual anatomical variation [2]. The volar rim projects palmarly, shifting the load-bearing axis from the lunate in the sagittal plane volarly, thus, transmitting the load of the hand to the elbow. This occurs largely through the palmar cortex, where the bone strength is high (Figure 1) [3]. The proximal part of the volar rim is covered with the pronator quadratus and the distal part is covered with fibrous tissue, forming the sliding floor of the flexor tendon located just above the rim [4,5].

The metaphyseal and epiphyseal areas of the radius, including this volar rim, are the sites where fractures frequently occur [6]. It is the area where strong forces are exerted during load transmission. Hence, a strong plate is used to maintain the restored position of the fracture. Since most plates used are ≥2 mm thick and are placed between the flexor tendon and bone where there is not much space, the influence on the flexor tendon and other soft tissues on the palmar side is substantial. Plates placed distally across the watershed line, the most prominent part of the volar rim, and plates that protrude more than 2 mm palmarly in the sagittal plane have been reported to be at risk for flexor tendon injury [7,8].

In recent years, plates designed to have a smaller palmar protrusion than the volar rim protrusion and ones with three-dimensional bending to match the shape of the volar rim have been introduced to reduce this effect on the palmar soft tissue (Figure 2) [4]. However, the only option for plates for osteosynthesis is bone width, which does not correspond to the anatomical variation of the volar rim.

Few studies have investigated individual variations in the volar rim. Sometimes plates with shapes significantly different from the anatomical shape are placed and can cause the rupture of the flexor tendon [2,9,10,11,12]. In this study, we examined three-dimensional variations in the shape of the volar rim. Additionally, we discussed whether a personalized implant shape design corresponding to the individual morphology of the volar rim is necessary for the treatment of fractures.

2. Materials and Methods

This study was performed after the approval of the institutional review board of our hospital. The patient age and gender were extracted from the medical records. We used computed tomography (CT) images of the wrist taken at our hospital over the past 10 years. The CT scanner was an Acquilion (Canon Medical Systems, Tokyo, Japan) and the image data were outputted with a slice thickness of 0.4 mm, segmented using Mimics 21.0 (Materialise, Leuven, Belgium), and then again outputted as a 3D surface model (Standard Triangulated Language, STL file) of the bone. Patients with inflammatory diseases such as rheumatoid arthritis or osteoarthritis of the hand and patients with a history of fractures were excluded from the analysis. Patients younger than 20 years of age were excluded. The surface models with less than 7 cm were excluded.

Measurements were performed using the 3D analysis software Rhinoceros 6.0 (Robert McNeel & Associates, Seattle, WA, USA). Concentric circles were created at 4 cm and 6 cm from the articular surface and the line connecting the centers of the circles was defined as the Z-axis (Figure 3). If the Z-axis was apparently different from the long axis of the radius, the Z-axis was adjusted to match its long axis. The distal radioulnar joint was identified in a cross-section perpendicular to the Z-axis and included the most proximal part of the radiocarpal joint. In the joint, the X-axis was defined as the line connecting the two points on the palmar and dorsal sides of the articular surface. The Y-axis is the axis perpendicular to it in the cross-section. The planes perpendicular to the X-, Y-, and Z-axes were defined as coronal, sagittal, and transverse planes, respectively.

The tangent line of the bone cortex at the most prominent part of the palmar side in a transverse section, 4 cm from the articular surface, was set as the U-axis (Figure 4). The T-axis was then defined as a straight line passing through the most prominent part of the palmar volar rim and the prominent part of the palmar cortex of the scaphoid fossa in a transverse section passing through the most prominent part of the volar rim (Figure 4). The angle between the T-axis and the U-axis was measured as the angle of rotation of the metaphysis relative to the diaphysis [13].

Measurements for the quantitative analyses were performed in the transverse and sagittal planes. In the transverse plane, we measured the angle of metaphysis rotation as described above and the bone width of the radius in the S plane. We also measured the angle between the Y-axis and the tangent line of the most prominent part (axial angle) and the prominent distance from the concave surface of the palmar side to the most prominent point of the volar rim (Figure 5). In the sagittal plane, as per the method of Yoneda et al., the angle of the inflection point on the cortex was taken as the teardrop inclination angle and the distance to the inflection point as the teardrop length (Figure 6) [2]. The teardrop inclination angle cannot be used for designing plates because the position of the inflection point varies in each case (Figure 7). To design the plate, the distance between the watershed line and inflection point was set to the average of the teardrop length and the sagittal angle, which corresponds to the bending angle of the plate in the sagittal plane, was measured. Finally, the transverse section was tilted based on the sagittal angle and the axial angle was measured as the modified axial angle.

3. Results

The number of cases for image analysis was 108, of which seven were excluded based on the exclusion criteria. Ultimately, surface models for the analysis were created using CT images of 101 cases (50 males and 51 females). A summary of the quantitative measurements is shown in

Table 1. Summary of the quantitative measurements.

		Average ± SD (CV)
	Male	Female	Total
Number of patients	50	51	101
Age [y]	58.9 ± 14.0	61.1 ± 12.0	60.0 ± 13.0
Bone width [mm] *	29.1 ± 1.6 (0.05)	25.2 ± 2.3 (0.09)	27.1 ± 2.8 (0.10)
Rotation of metaphysis [°]	8.8 ± 6.1 (0.69)	8.8 ± 5.9 (0.67)	8.8 ± 6.0 (0.68)
Teardrop inclination angle [°]	25.2 ± 4.7 (0.19)	20.4 ± 6.7 (0.33)	22.8 ± 6.2 (0.27)
Teardrop length [mm] *	11.7 ± 2.7 (0.23)	12.5 ± 2.7 (0.22)	12.1 ± 2.7 (0.23)
Sagittal angle [°] *	28.9 ± 4.5 (0.16)	24.4 ± 7.4 (0.30)	26.6 ± 6.5 (0.24)
Prominence distance [mm] *	6.2 ± 1.4 (0.47)	5.3 ± 1.6 (0.30)	5.76 ± 1.6 (0.27)
Axial angle [°]	17.9 ± 8.4 (0.22)	14.9 ± 8.0 (0.10)	16.4 ± 8.3 (0.51)
Modified axial angle [°]	12.0 ± 2.9 (0.24)	11.8 ± 2.7 (0.23)	12.0 ± 2.8 (0.23)

SD, standard deviation; CV, coefficient of variation. * Significant difference between males and females by Welch test (p < 0.05).

. In the transverse section, the bone width averaged 27.1 mm (SD 2.8), the rotation angle averaged 8.8° (SD 6.0), and the axial angle was 16.4 (SD 8.3). The mean teardrop inclination angle was 22.8 (SD 6.2) and the mean teardrop length was 12.1 (SD 2.7). Then, when the teardrop length was set to 12 mm, the average sagittal angle was 26.6° (SD 6.5) and the modified axial angle was 12.0° (SD 2.8). The coefficients of variation were more than 0.2 except for the bone width, suggesting individual differences in the anatomical morphology of the radius.

Among the measured items, the mean values of bone width, teardrop inclination angle, sagittal angle, and prominence distance differed between males and females (Table 1). No correlation with age was observed for any of the items. There was no correlation between bone width and angle of rotation, but moderate correlations were found between the sagittal angle, teardrop inclination angle, and axial angle (

Table 2. Correlations between measured parameters.

Parameters	Coefficient of Correlation
Age—Bone width	−0.27
Age—Rotation of metaphysis	−0.18
Age—Teardrop inclination angle	−0.13
Age—Teardrop length	0.09
Age—Sagittal angle	−0.09
Age—Axial angle	−0.23
Age—Prominence distance	−0.28
Age—Modified axial angle	−0.19
Bone width—Rotation of metaphysis	0.16
Bone width—Sagittal angle	0.46	*
Sagittal angle—Axial angle	0.53	*
Teardrop inclination angle—Axial angle	0.53	*
Sagittal angle—Modified axial angle	0.46	*
Teardrop inclination angle—Modified axial angle	0.44	*

* p < 0.05.

). Moderate correlations were also observed between modified axial angles.

The subgroup analysis was performed by dividing all models into three groups based on the bone width: lower 1/3 (range from 21.3 to 25.6 mm; small), middle 1/3 (range from 25.7 to 28.0 mm; medium), and upper 1/3 (range from 28.1 to 35.3 mm; large) with 33, 35, and 33 cases in the small, medium, and large groups, respectively. In the small and medium groups, a moderate correlation was observed between the sagittal and modified axial angles. However, the correlation between the sagittal and modified axial angles was lower in the large group (Figure 8). These results indicate that in bones with moderate or smaller widths, it is possible to create implants that match the shape of the rim. This can be achieved by preparing several variations correlating the angle of bending in the sagittal plane with the angle in the transverse plane. Conversely, in bones with large widths, the variation of the volar rim in the transverse and sagittal planes is large, causing it to be difficult to design an implant matching the shape of the volar rim even if multiple variations can be prepared. Possibly, the ulnar distal part of the plate should be adjustable to match the anatomy of the individual volar rim.

4. Discussion

It is widely known that there are individual differences in the morphology of the volar rim that cannot be explained by gender, extension, or weight alone. Several quantitative reports have been published previously and implants have been developed based on these findings. Oura et al. found that the amount of palmar protrusion in the transverse section was maximized at 4–6 mm from the articular surface, corresponding to the volar rim area [13]. In an analysis of reconstructed images from CT images, Andarmahr et al. found that the volar rim protrusion averaged 6 mm in the sagittal plane and Yoneda et al. showed that the angle between the protruding portion in the sagittal plane and the bone axis was 29° [2,14]. Kwon et al. found that the slope of the volar rim in the sagittal plane is steeper in males than in females [9].

A variation in the amount of palmar projection of the volar rim is one of the factors that concerns surgeons during plate placement in distal radius fracture surgery. Suppose the palmar protrusion of the volar rim is extensive, in that case, the plate may protrude similarly, preventing reconstruction of the flexor tendon gliding floor, and cause irritation of the flexor tendons or the median nerve. Therefore, the current treatment consensus requires that the plate not be placed distal to the watershed line and that it should not protrude palmarly more than 2 mm beyond the palmar aspect of the volar rim in the sagittal plane. Although it is desirable to use implants designed for individual geometries, current fracture implants offer only options for width and length and few implants offer possibilities for the bending angle of the plate that considers the shape of the volar rim.

The use of angularly stable locking plates to assemble fractures using implants as templates has become common [15,16]. In contrast, osteotomies for malunions are being performed based on preoperative planning using 3D image analysis technology and implants have been introduced that are placed according to the individual geometry of the osteotomy [17]. For fracture treatment, the use of implants that match the individual anatomy is ideal because a large discrepancy between the anatomical shape to be restored and the shape of the plate can cause malunion. However, because fractures are operated on a semi-emergency basis, individualized treatment is usually not feasible. Therefore, it is realistic to prepare multiple variations of implants and semi-customize the selection at the time of surgery and it is necessary to clarify the shape of the implant to accommodate these variations.

This study exhibited that gender is a factor that can affect bone width and volar rim anatomy, but age is unrelated to it. Previous studies have reported an effect of gender on fracture plate fit and the findings of this study support this [18]. The sagittal and transverse protrusions of the volar rim are moderately correlated. Furthermore, the individual differences in volar rim shape were found to be significant for bones with large widths. These results suggest that for bones with bone widths approximately less than 28 mm, plates should be designed with bending angles in the transverse plane that correlate to the bending angle in the sagittal plane. In bones with large bone widths, the variation in the shape of the volar rim tends to be significant, so it is desirable to be able to individually adjust the shape of the plate to match the rim shape. We conclude that for most bones, a plate with minimal soft tissue invasion can be designed using the above information and that custom-made plates are not necessary for treating fractures.

Next, the same investigation should be performed for other metaphyseal sites, such as distal femoral, proximal tibia, and distal tibia fractures. Similar to the distal radius fractures, the fixation strength of fractures with implants has been generally satisfactory in these sites and the next issue to be solved is the prevention of soft tissue problems caused by discrepancies between bone anatomy and implant shape [19,20]. The need for semi-custom implant design for fracture treatment based on individual bone anatomy for the treatment of any epiphyseal fractures needs to be clarified as well.

5. Limitation

This study had some limitations. First, because the analysis was limited to the bones of Asians, the external validity of the results for other races is uncertain. Second, the number of bones examined was small and multivariate analysis could not be performed to identify the factors affecting shape. Third, the thickness of the plate was not considered in this study, so the invasion of soft tissue could not be accurately predicted.

6. Conclusions

In conclusion, there are individual differences in the shape of the radius epiphysis and for bones of moderate or lower grade the size of the volar rim has a moderate correlation with the bone width in both the sagittal and transverse planes. As per the results of this study, we propose the development of several types of semi-custom plates with correlated sagittal and transverse bending angles. This allows for osteosynthesis with minimal soft tissue invasion while accommodating for the anatomical diversity of the volar rim.