- Email: [email protected]
Computed tomography study of radial head morphology
Computed tomography study of radial head morphology John M. Itamura, MD,a Nikolaos T. Roidis, MD, PhD,b Albert K. Chong, MD,a Suketu Vaishnav, MD,a Stamatios A. Papadakis, MD,c and Charalampos Zalavras, MD, PhD,a Los Angeles, CA; and Larisa and Athens, Greece
Computed tomography scans of 22 cadaveric adult elbows were obtained in 3 forearm positions: full supination, neutral, and full pronation. The radial head dimensions, the radiocapitellar joints, and the proximal radioulnar joints were measured. Multivariate analysis of variance was used to determine which portions of each articulation were the most congruent. The results showed that the radial head tended to become uncovered at the radial lip (P <.001). The radiocapitellar joint was tighter in pronation than in supination (P ¼ .001). The proximal radioulnar joint was most congruent at the middle proximal radioulnar joint, at the midportion and posterior aspects rather than the anterior aspect (P < .001). The proximal radioulnar joint coverage was between 69 and 79 . Prosthesis trial sizing should be judged by the articulations providing the most congruency: (1) the ulnar lip or trough of the radiocapitellar joint in pronation and (2) the posterior or midportion of the middle proximal radioulnar joint. (J Shoulder Elbow Surg 2008;17:347-354.)
T
he elbow is a complex joint that provides constrained motion in the sagittal plane, nearly free motion in the axial plane, and stability in the coronal plane. The radiocapitellar joint, with the proximal radioulnar joint, allows approximately 180 of motion in the axial plane. The radial head also acts as a secondary stabilizer to valgus stress and to posterior subluxation in the flexed elbow. Radial head fractures account for approximately 33% of all adult elbow injuries. They are a challenging problem to treat. Options include open reduction with internal fixation, radial head resection, or radial head replacement. Radial head replacements have proven useful with unreconstructible radial head fractures From the aDepartment of Othopaedics, Keck School of Medicine, University of Southern California; bDepartment of Othopaedics, University of Thessaly; and cDepartment of Othopaedics, Thriasio General Hospital. Reprint requests: Roidis Nikolaos, Consultant Orthopaedic Surgeon Orthopaedic Department, University of Thessaly, 34 Akronos Str, Ippokratis 41447, Larissa, Greece (E-mail: [email protected]). Copyright ª 2008 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2008/$34.00 doi:10.1016/j.jse.2007.07.019
because they help restore stability to the elbow. In addition, they prevent proximal radial migration in Essex-Lopresti type lesions.5,9,10-12 The basic shape of the radial head is a cylindrical saucer. Intraoperatively, however, the radial head is often more elliptical rather than perfectly circular. It is difficult to choose which size of radial head to implant during a radial head replacement. As evidenced by the different prosthetic designs available for use, there is a great deal of variability in how to recreate and replace the native radial head. The prostheses differ in radial head diameter, head height, and neck length.1,2,4,6-8,13,14 The design differences are likely due to the paucity of literature about the radial head morphology. The purpose of this study was to define the dimensions of the radial head better as well as those of the radiocapitellar and proximal radioulnar joints. Computed tomography (CT) scans of cadaveric elbows were used to define quantitatively and to measure various parameters of the proximal radius. The most congruent portions of the radial head articulations were determined. MATERIALS AND METHODS The study used 22 elbows from 12 cadavers (6 men, 6 women). The average age was 84.6 years (range, 74-98 years). All of these elbows were normal, with no previous traumatic (osseous or soft tissue) or osteoarthritic lesions. Our methodology regarding specimen positioning was (1) full pronation with the elbow flexed 90 , full pronation with the thumb parallel to the floor pointing to the medial side, (2) neutral with the thumb up, perpendicular to the level of the floor, and (3) full supination with the elbow in full extension, full supination, and thumb parallel to the floor pointing the lateral side. The elbows were scanned using a Philips-Marconi MX8000 CT Scanner (Philips Medical Systems, Bothell, WA) in 3 different forearm positions (full supination, neutral, and full pronation). The data were processed through the Marconi 3D Viewer (Marconi Medical Systems Inc., Highland Heights, OH), a software program that allows manipulation and reformatting of the images in 3 planes of freedom (axial, coronal, and sagittal) (Figure 1). The coronal plane was referenced off the transepicondylar axis of the distal humerus. The sagittal plane was referenced off the long axis of the proximal radius. Finally, the axial plane was referenced as the perpendicular to
347
348
Itamura et al
J Shoulder Elbow Surg March/April 2008
Figure 1 This computed tomography scan of cadaveric elbow specimen shows the measurements at the level of the radial head.
Figure 3 Levels of proximal radius measurements. A, trough; B, flare; C, head–neck interface; D, neck–tuberosity interface.
Figure 2 Elbow diagram.
the sagittal axis. Distances were measured using the builtin software ruler, which is calibrated to the nearest tenth of a millimeter. For faulty or incomplete scans, measurements were taken on the well-visualized areas and included in the statistical analyses. The radial head dimensions were measured on the axial cuts (Figures 1 and 2). They were taken at the level of the trough of the radial head (level A), the level of the widest part (flare) of the head (level B), and the level of the head–
neck interface (level C). If the head–neck interface was not an abrupt step-off, the level of the change in curvature from the head to neck was used. On the axial view, orthogonal lines were drawn to find the center of the head. The maximum and minimum diameters, which intersected the center point, were found, and the diameter difference was then calculated. The medullary neck diameter was measured, in a similar fashion, at the narrowest portion of the neck (isthmus; level D). Three independent reviewers used this single method, with each reviewer measuring only 1 forearm position. A single examiner performed all the remaining measurements (Figure 3). The radial head depth of curvature was measured as the perpendicular distance from the radial head trough to a line drawn from the radial lip to the ulnar lip. The radial head height (RHH) was measured as the axial distance from the radial lip to the head–neck interface. The radial neck height (RNH) was calculated as the axial distance from the head–neck interface to the neck–bicipital tuberosity interface, again using either an abrupt step-off or the change in curvature at this transition. A midcoronal slice of the radiocapitellar joint was taken. The radiocapitellar distance was measured at the radial lip, the trough of the radial head, and the ulnar lip (Figure 4). The offset of the trough in the radial head was measured. The total distance from the radial lip to the ulnar lip was measured. The offset distance was measured as the sagittal distance from the radial lip to trough. The percentage offset was calculated as the offset distance divided by the total distance. For instance, a 50% offset would indicate a trough centered
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
349
Figure 5 Radiocapitellar measurements. RL, radial lip; TR, trough; UL, ulnar lip.
Adjusted means and standard errors were derived. Adjustment for multiple-pair comparisons was made using the Scheffe method. The software program SAS 8.2 (SAS Institute, Inc., Cary, NC) was used for all statistical analysis.
RESULTS Radial head
Figure 4 Other proximal radius measurements. DC, depth of curvature; RHH, radial head height; RNH, radial neck height.
on the radial head. A 10% offset would indicate a trough closer to the radial lip, and a 90% offset would indicate a trough closer to the ulnar lip. The proximal radioulnar joint (PRUJ) was examined (Figure 5). On the midcoronal slice, the proximal PRUJ (PPRUJ) was defined as the point where the radial head engages the PRUJ, the distal PRUJ as the point where the radial head disengages the PRUJ, and the middle PRUJ as the midpoint between the two. Axial slices were then made through these 3 points. The radioulnar distance was measured on these axial slices at the anterior lip of the ulna, the posterior lip of the ulna, and the midportion between the anterior and posterior lips. The amount of coverage at the PRUJ (Figure 6) was also determined on the same 3 axial slices. Coverage was defined as the angle between the anterior lip of the ulna to the center of the radial head to the posterior lip of the ulna (Figure 7).
Statistical analysis Two- and 3-factor nonparametric analysis of variance was used to examine the significance of difference by position, location, and area, as well as their interactions.
The measurements at level A and D were statistically similar for maximum diameter, minimum diameter, and diameter difference across the 3 forearm positions (P > .5; Table I). However, the measurements at levels B and C varied significantly across forearm position (P <.05) and were not used in the final calculations. At the level of the radial head trough (level A), the maximum diameter was 22.3 mm and the minimum diameter was 20.9 mm, a difference of 1.4 mm. This difference represented only 6.3% of the overall maximum diameter. The depth of curvature of the radial head trough was 2.3 mm, the radial head length was 9.8 mm, and the radial neck length was 10.7 mm. At the isthmus of the medullary canal (level D), the maximum diameter was 9.7 mm and the minimum diameter was 8.2 mm, a difference of 1.5 mm. This difference represented 15.6% of the overall maximum diameter. The average trough offset was 8.2 mm. Overall, the average percentage offset was 50%, ranging from 51% in supination to 49% in neutral and to 49% in pronation (P ¼ .35); however, the range was large (39%–63%). Radiocapitellar joint
The average radiocapitellar distance (Table II) at the radial lip was 4.0 mm, nearly double the distance at the trough (2.4 mm) and the ulnar lip (2.2 mm; adjusted P < .0001; Table II). The radiocapitellar distance in pronation tended to be less than in supination (adjusted P ¼ .0008).
350
Itamura et al
J Shoulder Elbow Surg March/April 2008
Figure 6 Proximal radioulnar joint (PRUJ) measurements from the (left) coronal and (right) transverse views. PPRUJ, proximal PRUJ; MPRUJ, middle PRUJ; DPRUJ, distal PRUJ; ANT, anterior; MID, midportion; POST, posterior.
justed P < .0001). The PRUJ also tended to be more congruent in the midportion or the posterior aspect rather than the anterior aspect (adjusted P < .0001). Proximal radioulnar joint coverage
The overall PRUJ coverage was consistent across forearm position (adjusted P > .05; Table IV). When looking solely at the distal PRUJ, however, the measured coverage varied significantly across forearm positions (P ¼ .0169). DISCUSSION
Figure 7 Proximal radioulnar joint coverage (degrees).
Proximal radioulnar joint
At the PRUJ, significantly different interactions were found among position, location, and area (adjusted P < .0001; Table III). The one exception was the interaction between position and area (P ¼ .11). The PRUJ tended to be most congruent at the middle PRUJ (ad-
A review of the literature yields a handful of studies about radial head morphology.1,2,4,6-8,13,14 Many different methods have been used to assess radial head morphology, including calipers, magnetic resonance imaging (MRI), and coordinate measuring machines. The literature also supports the mismatch of radial head replacements and the native anatomy.1,6,11-13 To our knowledge, no study has assessed the radiocapitellar and proximal radioulnar joint distances. Beredjiklian et al1 reviewed 56 MRI scans of patients with lateral epicondylitis. All the scans were performed in full extension and forearm supination. Using fine calipers and a metric ruler, 2 authors performed multiple measurements on the same scans.
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
351
Table I Mean and SD of maximum and minimum diameters of proximal radius by position and level (mm) Forearm position Level
Supinated (n ¼ 22)
Neutral (n ¼ 22)
Pronated (n ¼ 19)
P
Overall (n ¼ 63)
Range
22.4 6 2.4 21.0 6 2.1 1.5 6 0.6
22.6 6 2.4 21.1 6 2.1 1.5 6 0.5
21.9 6 2.4 20.5 6 2.0 1.4 6 0.5
.505 .652 .565
22.3 6 2.4 20.9 6 2.1 1.4 6 0.5
18.6-26.6 17.9-25.1 0.6-2.5
22.6 6 2.3 21.3 6 2.0 1.3 6 0.6
22.6 6 2.3 21.3 6 1.8 1.4 6 0.6
21.6 6 2.2 20.2 6 1.8 1.5 6 0.7
.006 .001 .045
22.3 6 2.3 20.9 6 1.9 1.4 6 0.6
18.5-26.4 17.6-24.9 0.4-2.9
17.9 6 3.3 15.7 6 2.7 2.2 6 1.2
16.9 6 2.1 15.1 6 2.0 1.8 6 0.7
18.6 6 2.7 16.5 6 2.2 2.1 6 1.0
.024 .022 .074
17.8 6 2.8 15.7 6 2.4 2.0 6 1.0
12.5-23.7 11.2-21.5 0.5-5.0
9.7 6 2.0 8.2 6 1.7 1.5 6 0.7
9.8 6 2.1 8.4 6 1.9 1.4 6 0.6
9.7 6 2.0 8.1 6 1.8 1.6 6 0.6
.284 .512 .787
9.7 6 2.0 8.2 6 1.8 1.5 6 0.6
6.0-14.1 5.1-12.1 0.4-3.0
Level A Max Min Diff Level B Max Min Diff Level C Max Min Diff Level D Max Min Diff
Table II Mean and SD of radiocapitellar distance by position and level (mm) Forearm position Level
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
Radial lip Trough Ulnar lip P values For 2-factor ANOVA Position Location Position vs location interaction Adjusted for multiple comparisons Radial lip vs trough Radial lip vs ulnar lip Trough vs ulnar lip Supinated vs neutral Supinated vs pronated Neutral vs pronated
4.6 6 1.3 2.7 6 0.7 2.4 6 0.6
4.2 6 0.9 2.4 6 0.5 2.1 6 0.4
3.2 6 1.0 2.1 6 0.5 2.2 6 0.4
P .0003 .0013 .0228
Overall (n ¼ 63)
Range
4.0 6 1.2 2.4 6 0.6 2.2 6 0.5
1.3 – 7.1 0.9 – 3.8 1.2 – 3.5
.0008 <.0001 .0864 <.0001 <.0001 .1777 .1307 .0008 .1944
ANOVA, Analysis of variance.
The measurements were compared with the dimensions of commercially available radial head prostheses. The prostheses were often found to overestimate the size of the native radial head. In addition, in those radial necks large enough to accommodate a prosthetic stem, there would be an average shortening of 4 mm had the implant been placed. This study highlighted the difficulty in re-creating the proximal radial anatomy using radial head replacements. Captier et al2 used a caliper to measure maximum and minimum diameters of the radial head. Of the 96 radial head specimens, 57% were elliptical, defined
as having a diameter difference of greater than 1 mm. The elliptical radial heads had a maximum diameter of 22 6 2.9 mm and a minimum diameter of 20 6 2.8 mm. The other 43% of radial heads were circular, defined as having a difference in diameters of less than 1 mm. Circular heads had a maximum diameter of 21.2 6 2.4 mm and a minimum diameter of 20.4 6 2.4 mm. These values correspond well with our own measurements of 22.3 and 20.9 mm. Cone et al3 used CT scans with 5-mm-thick slices at 5-mm intervals. The maximum and minimum diameters were measured at neutral, 45 of supination, and 45
352
Itamura et al
J Shoulder Elbow Surg March/April 2008
Table III Mean and SD of proximal radioulnar joint distances by position, location, and area (mm) Forearm position Variable PPRUJ Anterior Midpoint Posterior MPRUJ Anterior Midpoint Posterior DPRUJ Anterior Midpoint Posterior
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
4.2 6 1.9 2.6 6 0.8 1.7 6 0.6
4.5 6 2.3 2.7 6 0.7 2.3 6 0.6
4.1 6 2.4 2.6 6 0.7 2.2 6 0.7
2.5 6 0.8 2.0 6 0.4 1.3 6 0.5
2.4 6 0.9 1.9 6 0.4 1.9 6 0.6
2.0 6 0.7 1.8 6 0.4 2.6 6 1.0
3.4 6 1.4 2.5 6 1.1 2.7 6 1.1
P
Overall (n ¼ 63)
Range
.3394 .6634 .0003
4.3 6 2.2 2.6 6 0.7 2.0 6 0.7
0.7 - 9.0 1.1 - 5.2 0.8 - 3.6
2.0 6 1.0 1.6 6 0.5 1.8 6 0.5
.0238 .0019 .0002
2.3 6 0.9 1.8 6 0.4 1.7 6 0.6
0.7 - 4.5 0.7 - 2.7 0.4 - 3.5
4.3 6 1.5 3.8 6 0.9 4.2 6 1.3
<0.0001 <0.0001 <0.0001
3.2 6 1.6 2.7 6 1.2 3.1 6 1.3
1.0 - 7.9 0.8 - 5.5 0.9 - 5.9
P values for 3-factor analysis of variance Position (supinated, neutral, pronated) Location (PPRUJ, MPRUJ, DPRUJ) Position location interaction Area (anterior, midpoint, posterior) Position area interaction Location area interaction Position location area interaction Adjusted P values for multiple comparisons Position Supinated vs neutral Supinated vs pronated Neutral vs pronated Location PPRUJ vs MPRUJ PPRUJ vs DPRUJ MPRUJ vs DPRUJ Area Anterior vs midpoint Anterior vs posterior Midpoint vs posterior
<.0001 <.0001 <.0001 <.0001 0.1141 <.0001 <.0001
.0016 .0001 .087 .0001 .8998 .0001 .0001 .0001 .7433
DPRUJ, Distal proximal radioulnar joint; MPRUJ, middle proximal radioulnar joint; PPRUJ, proximal proximal radioulnar joint.
of pronation to assess rotation of the radius in relation to the radial articular fossa of the ulna. They also concluded that the radial head was ovoid, with a diameter difference of 2.5 mm at the level of the proximal radioulnar joint. They found that in supination, the axis of the maximum diameter is roughly parallel to the radial articular fossa of the ulna. King et al6 measured 28 cadaveric samples and 40 radiographs with a tactile probe coordinate measuring machine. They found that the radial head was variably offset on the axis of the neck (4.2 6 2.5 mm). They also found a poor correlation between radial head and medullary canal diameter, with a diameter difference of 1.7 to 0.7 mm. This again corresponded well with our calculated difference of 1.4 to 2.0 mm, depending on the level of the radial head measurements. Swieszkowski et al13 measured 34 cadaveric proximal radii with a coordinate measuring machine. The
mean 6 SD measurements were maximum diameter, 23.36 6 1.14 mm; radial head height, 10.14 6 1.38; and the radial head depth, 1.92 6 0.32 mm. This again correlated well with our findings of 22.3mm maximum diameter, 9.8-mm head length, and 2.3-mm depth of curvature. In our study, the maximum and minimum diameters of the radial head did not change with forearm rotation at the trough (level A) and the isthmus (level D). However, at levels B and C, the values varied significantly across forearm position. This may reflect the inconsistency and difficulty in determining the level of the flare and head–neck interface compared with the ease in identifying the trough and isthmus as bony landmarks. Therefore, our conclusions are based only on the values obtained at levels A and D. The radial head is, indeed, not perfectly circular. At the radial head trough, however, there was only a 6%
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
353
Table IV Mean and SD of proximal radioulnar joint coverage by position and location (degrees) Forearm position Location PPRUJ MPRUJ DPRUJ
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
P
Overall (n ¼ 63)
Range
77.7 6 5.9 78.8 6 6.4 72.6 6 9.4
76.4 6 8.4 79.3 6 9.1 69.6 6 13.0
77.0 6 9.9 79.6 6 7.6 64.4 6 11.1
.7768 .148 .0169
77.76 8.0 79.2 6 7.6 69.0 6 11.6
55-99 61-100 47-97
P values for 2-factor analysis of variance Position (supinated, neutral, pronated) Location (PPRUJ, MPRUJ, DPRUJ) Position*location interaction Adjusted P values for multiple comparisons Position Supinated vs neutral Supinated vs pronated Neutral vs pronated Location PPRUJ vs MPRUJ PPRUJ vs DPRUJ MPRUJ vs DPRUJ
.6277 <.0001 .4755
.7353 .2556 .6835 .4124 <.0001 <.0001
DPRUJ, Distal proximal radioulnar joint; MPRUJ, middle proximal radioulnar joint; PPRUJ, proximal proximal radioulnar joint.
difference in the maximum and minimum diameters. To produce elliptical radial head components would also introduce rotational angulation. The data for the trough offset were inconclusive. The trough is roughly in the center of the radial head (50%), but with a large range of 39% to 63%. Again, manufacturing radial heads incorporating this wide range of trough offset would be very difficult. The medullary canal is even more elliptical, with a 16% difference between maximum and minimum diameters. Creating elliptical radial necks would introduce rotational angulation of the head on the stem and the stem into the canal. This would greatly increase the technical difficulty of performing a radial head replacement. At the radiocapitellar joint, the trough and ulnar lip are more congruent, with the joint distance increasing at the radial lip. Therefore, implants should be matched with either the trough or the ulnar lip rather than the radial lip. The joint is also more congruent in pronation than in supination; therefore, implants should fit tighter in pronation than in supination. In other words, they will naturally be looser in supination. The PRUJ is most congruent at the middle PRUJ, rather than the proximal or distal portion. It is also more congruent at the midportion or posterior aspects, with the joint distance increasing anteriorly. Implants should therefore match up with the middle PRUJ at either the middle or posterior aspect. The study had a few sources of possible bias. The cadaveric specimens were older and may have had arthritic changes. Osteophytes may have caused the measurements to overestimate the size of the true radial head. Cartilage wear may have caused the mea-
surements to underestimate the true joint distances. Being cadaveric specimens, the lack of in vivo soft tissue tensioning also may have affected the measurements. There was also investigator bias in the measurements of the radial head diameters. Three different people measured 1 forearm position each; however, this was addressed by comparing the measurements across forearm position. Overall, the measurements at level A and D were consistent across forearm positions (P > .05), as would be expected in areas with more consistent landmarks. Measurement error was also minimized by using the built-in software ruler. With the roughly 10-fold magnification of the specimens on the computer screen, one would need to make a full 1-cm error in plotting data points on the screen to create a 1-mm error in measurement. Clinical relevance
At our institutions, radial heads are implanted using a modified Boyd and Anderson posterior approach to the forearm. The proximal portion of the anconeus muscle and the elbow capsule are lifted off the ulna to allow later repair with suture anchors. This approach allows visualization of the ulnar lip and trough of the radial head as well as the posterior portion of the proximal radioulnar joint.11,12 These areas correspond to the most congruent portions of the radial head articulations. In conclusion, the elliptical shape of the native radial head is not replicated by the current cylindrical radial head prostheses. This difference is relatively small, however. The trough is roughly centered on the
354
Itamura et al
radial head, but with a large range. The elliptical medullary canal also has implications for radial head replacement. Creating elliptical radial heads and stems would introduce rotational angulation of the head on the stem and the stem into the canal. Therefore, introducing angulation, rotation, and offset would exponentially increase in technical difficulty in performing radial head replacements. Given all these factors, it is probably not cost-effective or technically feasible to manufacture elliptical radial heads and stems. Size mismatch is another common problem in radial head replacements. Radial heads sizes tend to run large and stem size choices tend to run large for the native canals. Potential problems include overstuffing the elbow or fracture of the radial neck and shaft. Shape mismatch can cause camming at the radiocapitellar and the proximal radioulnar joints. This may decrease stability, cause pain or stiffness, and lead to earlier prosthetic failure. Because it may not be feasible to create anatomic radial head replacements, it is important to maximize surgical technique. With the current cylindrical radial head prostheses, matching up only the most congruent portions of the articulations can avoid mismatch. This study points to 2 main areas: (1) the ulnar lip or trough of the radiocapitellar joint in pronation and (2) the posterior or midportion of the middle PRUJ. The clinical significance of this technique remains to be seen. We acknowledge the assistance of Chuck Nguyen, Alejandro Vasquez, Neil Parikh, Linda Chan, PhD, and Emily Ramicone, MS. REFERENCES
1. Beredjiklian PK, Nalbantoglu U, Potter HG, Hotchkiss RN. Prosthetic radial head components and proximal radial morphology: a mismatch. J Shoulder Elbow Surg 1999;8:471-5.
J Shoulder Elbow Surg March/April 2008
2. Captier G, Canovas F, Mercier N, Thomas E, Bonnel F. Biometry of radial head: biomechanical implications in pronation and supination. Surg Radiol Anat 2002;24:295-301. 3. Cone RO, Szabo R, Resnick D, Gelberman R, Taleisnik J, Gilula LA. Computed tomography of the normal radioulnar joints. Invest Radiol 1983;18:541-5. 4. Gupta GG, Lucas G, Hahn DL. Biomechanical and computer analysis of radial head prostheses. J Shoulder Elbow Surg 1997;6:37-48. 5. Itamura J, Roidis N, Mirzayan R, Vaishnav S, Learch T, Shean C. Radial head fractures: MRI evaluation of associated injuries. J Shoulder Elbow Surg 2005;14:421-4. 6. King GJ, Zarzour ZD, Patterson SD, Johnson JA. An anthropometric study of the radial head: implications in the design of a prosthesis. J Arthroplasty 2001;16:112-6. 7. Mahaisavariya B, Saekee B, Sitthiseripratip K, et al. Morphology of the radial head: a reverse engineering based evaluation using three-dimensional anatomical data of radial bone. Proc Inst Mech Eng [H] 2004;218:79-84. 8. Popovic N, Djekic J, Lemaire R, Gillet P. A comparative study between proximal radial morphology and the floating radial head prosthesis. J Shoulder Elbow Surg 2005;14:433-40. 9. Ring D. Open reduction and internal fixation of fractures of the radial head. Hand Clin 2004;20:415-27. 10. Roidis N, Stevanovic M, Martirosian A, Abbott DD, McPherson EJ, Itamura JM. A radiographic study of proximal radius anatomy with implications in radial head replacement. J Shoulder Elbow Surg 2003;12:380-4. 11. Roidis NT, Papadakis SA, Karachalios TS, Mirzayan R, Itamura JM. Radial head fractures. In: Mirzayan R, Itamura JM, editors. Shoulder and elbow trauma. New York, NY: Thieme Medical Publishers Inc; 2004. p. 22-35. 12. Roidis NT, Papadakis SA, Rigopoulos N, et al. Current concepts and controversies in the management of radial head fractures. Orthopedics 2006;29:904-16; quiz 917–8. 13. Swieszkowski W, Skalski K, Pomianowski S, Kedzior K. The anatomic features of the radial head and their implication for prosthesis design. Clin Biomech (Bristol, Avon) 2001;16:880-7. 14. van Riet RP, Van Glabbeek F, Neale PG, Bortier H, An KN, O’Driscoll SW. The noncircular shape of the radial head. J Hand Surg Am 2003;28:972-8.
Computed tomography scans of 22 cadaveric adult elbows were obtained in 3 forearm positions: full supination, neutral, and full pronation. The radial head dimensions, the radiocapitellar joints, and the proximal radioulnar joints were measured. Multivariate analysis of variance was used to determine which portions of each articulation were the most congruent. The results showed that the radial head tended to become uncovered at the radial lip (P <.001). The radiocapitellar joint was tighter in pronation than in supination (P ¼ .001). The proximal radioulnar joint was most congruent at the middle proximal radioulnar joint, at the midportion and posterior aspects rather than the anterior aspect (P < .001). The proximal radioulnar joint coverage was between 69 and 79 . Prosthesis trial sizing should be judged by the articulations providing the most congruency: (1) the ulnar lip or trough of the radiocapitellar joint in pronation and (2) the posterior or midportion of the middle proximal radioulnar joint. (J Shoulder Elbow Surg 2008;17:347-354.)
T
he elbow is a complex joint that provides constrained motion in the sagittal plane, nearly free motion in the axial plane, and stability in the coronal plane. The radiocapitellar joint, with the proximal radioulnar joint, allows approximately 180 of motion in the axial plane. The radial head also acts as a secondary stabilizer to valgus stress and to posterior subluxation in the flexed elbow. Radial head fractures account for approximately 33% of all adult elbow injuries. They are a challenging problem to treat. Options include open reduction with internal fixation, radial head resection, or radial head replacement. Radial head replacements have proven useful with unreconstructible radial head fractures From the aDepartment of Othopaedics, Keck School of Medicine, University of Southern California; bDepartment of Othopaedics, University of Thessaly; and cDepartment of Othopaedics, Thriasio General Hospital. Reprint requests: Roidis Nikolaos, Consultant Orthopaedic Surgeon Orthopaedic Department, University of Thessaly, 34 Akronos Str, Ippokratis 41447, Larissa, Greece (E-mail: [email protected]). Copyright ª 2008 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2008/$34.00 doi:10.1016/j.jse.2007.07.019
because they help restore stability to the elbow. In addition, they prevent proximal radial migration in Essex-Lopresti type lesions.5,9,10-12 The basic shape of the radial head is a cylindrical saucer. Intraoperatively, however, the radial head is often more elliptical rather than perfectly circular. It is difficult to choose which size of radial head to implant during a radial head replacement. As evidenced by the different prosthetic designs available for use, there is a great deal of variability in how to recreate and replace the native radial head. The prostheses differ in radial head diameter, head height, and neck length.1,2,4,6-8,13,14 The design differences are likely due to the paucity of literature about the radial head morphology. The purpose of this study was to define the dimensions of the radial head better as well as those of the radiocapitellar and proximal radioulnar joints. Computed tomography (CT) scans of cadaveric elbows were used to define quantitatively and to measure various parameters of the proximal radius. The most congruent portions of the radial head articulations were determined. MATERIALS AND METHODS The study used 22 elbows from 12 cadavers (6 men, 6 women). The average age was 84.6 years (range, 74-98 years). All of these elbows were normal, with no previous traumatic (osseous or soft tissue) or osteoarthritic lesions. Our methodology regarding specimen positioning was (1) full pronation with the elbow flexed 90 , full pronation with the thumb parallel to the floor pointing to the medial side, (2) neutral with the thumb up, perpendicular to the level of the floor, and (3) full supination with the elbow in full extension, full supination, and thumb parallel to the floor pointing the lateral side. The elbows were scanned using a Philips-Marconi MX8000 CT Scanner (Philips Medical Systems, Bothell, WA) in 3 different forearm positions (full supination, neutral, and full pronation). The data were processed through the Marconi 3D Viewer (Marconi Medical Systems Inc., Highland Heights, OH), a software program that allows manipulation and reformatting of the images in 3 planes of freedom (axial, coronal, and sagittal) (Figure 1). The coronal plane was referenced off the transepicondylar axis of the distal humerus. The sagittal plane was referenced off the long axis of the proximal radius. Finally, the axial plane was referenced as the perpendicular to
347
348
Itamura et al
J Shoulder Elbow Surg March/April 2008
Figure 1 This computed tomography scan of cadaveric elbow specimen shows the measurements at the level of the radial head.
Figure 3 Levels of proximal radius measurements. A, trough; B, flare; C, head–neck interface; D, neck–tuberosity interface.
Figure 2 Elbow diagram.
the sagittal axis. Distances were measured using the builtin software ruler, which is calibrated to the nearest tenth of a millimeter. For faulty or incomplete scans, measurements were taken on the well-visualized areas and included in the statistical analyses. The radial head dimensions were measured on the axial cuts (Figures 1 and 2). They were taken at the level of the trough of the radial head (level A), the level of the widest part (flare) of the head (level B), and the level of the head–
neck interface (level C). If the head–neck interface was not an abrupt step-off, the level of the change in curvature from the head to neck was used. On the axial view, orthogonal lines were drawn to find the center of the head. The maximum and minimum diameters, which intersected the center point, were found, and the diameter difference was then calculated. The medullary neck diameter was measured, in a similar fashion, at the narrowest portion of the neck (isthmus; level D). Three independent reviewers used this single method, with each reviewer measuring only 1 forearm position. A single examiner performed all the remaining measurements (Figure 3). The radial head depth of curvature was measured as the perpendicular distance from the radial head trough to a line drawn from the radial lip to the ulnar lip. The radial head height (RHH) was measured as the axial distance from the radial lip to the head–neck interface. The radial neck height (RNH) was calculated as the axial distance from the head–neck interface to the neck–bicipital tuberosity interface, again using either an abrupt step-off or the change in curvature at this transition. A midcoronal slice of the radiocapitellar joint was taken. The radiocapitellar distance was measured at the radial lip, the trough of the radial head, and the ulnar lip (Figure 4). The offset of the trough in the radial head was measured. The total distance from the radial lip to the ulnar lip was measured. The offset distance was measured as the sagittal distance from the radial lip to trough. The percentage offset was calculated as the offset distance divided by the total distance. For instance, a 50% offset would indicate a trough centered
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
349
Figure 5 Radiocapitellar measurements. RL, radial lip; TR, trough; UL, ulnar lip.
Adjusted means and standard errors were derived. Adjustment for multiple-pair comparisons was made using the Scheffe method. The software program SAS 8.2 (SAS Institute, Inc., Cary, NC) was used for all statistical analysis.
RESULTS Radial head
Figure 4 Other proximal radius measurements. DC, depth of curvature; RHH, radial head height; RNH, radial neck height.
on the radial head. A 10% offset would indicate a trough closer to the radial lip, and a 90% offset would indicate a trough closer to the ulnar lip. The proximal radioulnar joint (PRUJ) was examined (Figure 5). On the midcoronal slice, the proximal PRUJ (PPRUJ) was defined as the point where the radial head engages the PRUJ, the distal PRUJ as the point where the radial head disengages the PRUJ, and the middle PRUJ as the midpoint between the two. Axial slices were then made through these 3 points. The radioulnar distance was measured on these axial slices at the anterior lip of the ulna, the posterior lip of the ulna, and the midportion between the anterior and posterior lips. The amount of coverage at the PRUJ (Figure 6) was also determined on the same 3 axial slices. Coverage was defined as the angle between the anterior lip of the ulna to the center of the radial head to the posterior lip of the ulna (Figure 7).
Statistical analysis Two- and 3-factor nonparametric analysis of variance was used to examine the significance of difference by position, location, and area, as well as their interactions.
The measurements at level A and D were statistically similar for maximum diameter, minimum diameter, and diameter difference across the 3 forearm positions (P > .5; Table I). However, the measurements at levels B and C varied significantly across forearm position (P <.05) and were not used in the final calculations. At the level of the radial head trough (level A), the maximum diameter was 22.3 mm and the minimum diameter was 20.9 mm, a difference of 1.4 mm. This difference represented only 6.3% of the overall maximum diameter. The depth of curvature of the radial head trough was 2.3 mm, the radial head length was 9.8 mm, and the radial neck length was 10.7 mm. At the isthmus of the medullary canal (level D), the maximum diameter was 9.7 mm and the minimum diameter was 8.2 mm, a difference of 1.5 mm. This difference represented 15.6% of the overall maximum diameter. The average trough offset was 8.2 mm. Overall, the average percentage offset was 50%, ranging from 51% in supination to 49% in neutral and to 49% in pronation (P ¼ .35); however, the range was large (39%–63%). Radiocapitellar joint
The average radiocapitellar distance (Table II) at the radial lip was 4.0 mm, nearly double the distance at the trough (2.4 mm) and the ulnar lip (2.2 mm; adjusted P < .0001; Table II). The radiocapitellar distance in pronation tended to be less than in supination (adjusted P ¼ .0008).
350
Itamura et al
J Shoulder Elbow Surg March/April 2008
Figure 6 Proximal radioulnar joint (PRUJ) measurements from the (left) coronal and (right) transverse views. PPRUJ, proximal PRUJ; MPRUJ, middle PRUJ; DPRUJ, distal PRUJ; ANT, anterior; MID, midportion; POST, posterior.
justed P < .0001). The PRUJ also tended to be more congruent in the midportion or the posterior aspect rather than the anterior aspect (adjusted P < .0001). Proximal radioulnar joint coverage
The overall PRUJ coverage was consistent across forearm position (adjusted P > .05; Table IV). When looking solely at the distal PRUJ, however, the measured coverage varied significantly across forearm positions (P ¼ .0169). DISCUSSION
Figure 7 Proximal radioulnar joint coverage (degrees).
Proximal radioulnar joint
At the PRUJ, significantly different interactions were found among position, location, and area (adjusted P < .0001; Table III). The one exception was the interaction between position and area (P ¼ .11). The PRUJ tended to be most congruent at the middle PRUJ (ad-
A review of the literature yields a handful of studies about radial head morphology.1,2,4,6-8,13,14 Many different methods have been used to assess radial head morphology, including calipers, magnetic resonance imaging (MRI), and coordinate measuring machines. The literature also supports the mismatch of radial head replacements and the native anatomy.1,6,11-13 To our knowledge, no study has assessed the radiocapitellar and proximal radioulnar joint distances. Beredjiklian et al1 reviewed 56 MRI scans of patients with lateral epicondylitis. All the scans were performed in full extension and forearm supination. Using fine calipers and a metric ruler, 2 authors performed multiple measurements on the same scans.
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
351
Table I Mean and SD of maximum and minimum diameters of proximal radius by position and level (mm) Forearm position Level
Supinated (n ¼ 22)
Neutral (n ¼ 22)
Pronated (n ¼ 19)
P
Overall (n ¼ 63)
Range
22.4 6 2.4 21.0 6 2.1 1.5 6 0.6
22.6 6 2.4 21.1 6 2.1 1.5 6 0.5
21.9 6 2.4 20.5 6 2.0 1.4 6 0.5
.505 .652 .565
22.3 6 2.4 20.9 6 2.1 1.4 6 0.5
18.6-26.6 17.9-25.1 0.6-2.5
22.6 6 2.3 21.3 6 2.0 1.3 6 0.6
22.6 6 2.3 21.3 6 1.8 1.4 6 0.6
21.6 6 2.2 20.2 6 1.8 1.5 6 0.7
.006 .001 .045
22.3 6 2.3 20.9 6 1.9 1.4 6 0.6
18.5-26.4 17.6-24.9 0.4-2.9
17.9 6 3.3 15.7 6 2.7 2.2 6 1.2
16.9 6 2.1 15.1 6 2.0 1.8 6 0.7
18.6 6 2.7 16.5 6 2.2 2.1 6 1.0
.024 .022 .074
17.8 6 2.8 15.7 6 2.4 2.0 6 1.0
12.5-23.7 11.2-21.5 0.5-5.0
9.7 6 2.0 8.2 6 1.7 1.5 6 0.7
9.8 6 2.1 8.4 6 1.9 1.4 6 0.6
9.7 6 2.0 8.1 6 1.8 1.6 6 0.6
.284 .512 .787
9.7 6 2.0 8.2 6 1.8 1.5 6 0.6
6.0-14.1 5.1-12.1 0.4-3.0
Level A Max Min Diff Level B Max Min Diff Level C Max Min Diff Level D Max Min Diff
Table II Mean and SD of radiocapitellar distance by position and level (mm) Forearm position Level
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
Radial lip Trough Ulnar lip P values For 2-factor ANOVA Position Location Position vs location interaction Adjusted for multiple comparisons Radial lip vs trough Radial lip vs ulnar lip Trough vs ulnar lip Supinated vs neutral Supinated vs pronated Neutral vs pronated
4.6 6 1.3 2.7 6 0.7 2.4 6 0.6
4.2 6 0.9 2.4 6 0.5 2.1 6 0.4
3.2 6 1.0 2.1 6 0.5 2.2 6 0.4
P .0003 .0013 .0228
Overall (n ¼ 63)
Range
4.0 6 1.2 2.4 6 0.6 2.2 6 0.5
1.3 – 7.1 0.9 – 3.8 1.2 – 3.5
.0008 <.0001 .0864 <.0001 <.0001 .1777 .1307 .0008 .1944
ANOVA, Analysis of variance.
The measurements were compared with the dimensions of commercially available radial head prostheses. The prostheses were often found to overestimate the size of the native radial head. In addition, in those radial necks large enough to accommodate a prosthetic stem, there would be an average shortening of 4 mm had the implant been placed. This study highlighted the difficulty in re-creating the proximal radial anatomy using radial head replacements. Captier et al2 used a caliper to measure maximum and minimum diameters of the radial head. Of the 96 radial head specimens, 57% were elliptical, defined
as having a diameter difference of greater than 1 mm. The elliptical radial heads had a maximum diameter of 22 6 2.9 mm and a minimum diameter of 20 6 2.8 mm. The other 43% of radial heads were circular, defined as having a difference in diameters of less than 1 mm. Circular heads had a maximum diameter of 21.2 6 2.4 mm and a minimum diameter of 20.4 6 2.4 mm. These values correspond well with our own measurements of 22.3 and 20.9 mm. Cone et al3 used CT scans with 5-mm-thick slices at 5-mm intervals. The maximum and minimum diameters were measured at neutral, 45 of supination, and 45
352
Itamura et al
J Shoulder Elbow Surg March/April 2008
Table III Mean and SD of proximal radioulnar joint distances by position, location, and area (mm) Forearm position Variable PPRUJ Anterior Midpoint Posterior MPRUJ Anterior Midpoint Posterior DPRUJ Anterior Midpoint Posterior
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
4.2 6 1.9 2.6 6 0.8 1.7 6 0.6
4.5 6 2.3 2.7 6 0.7 2.3 6 0.6
4.1 6 2.4 2.6 6 0.7 2.2 6 0.7
2.5 6 0.8 2.0 6 0.4 1.3 6 0.5
2.4 6 0.9 1.9 6 0.4 1.9 6 0.6
2.0 6 0.7 1.8 6 0.4 2.6 6 1.0
3.4 6 1.4 2.5 6 1.1 2.7 6 1.1
P
Overall (n ¼ 63)
Range
.3394 .6634 .0003
4.3 6 2.2 2.6 6 0.7 2.0 6 0.7
0.7 - 9.0 1.1 - 5.2 0.8 - 3.6
2.0 6 1.0 1.6 6 0.5 1.8 6 0.5
.0238 .0019 .0002
2.3 6 0.9 1.8 6 0.4 1.7 6 0.6
0.7 - 4.5 0.7 - 2.7 0.4 - 3.5
4.3 6 1.5 3.8 6 0.9 4.2 6 1.3
<0.0001 <0.0001 <0.0001
3.2 6 1.6 2.7 6 1.2 3.1 6 1.3
1.0 - 7.9 0.8 - 5.5 0.9 - 5.9
P values for 3-factor analysis of variance Position (supinated, neutral, pronated) Location (PPRUJ, MPRUJ, DPRUJ) Position location interaction Area (anterior, midpoint, posterior) Position area interaction Location area interaction Position location area interaction Adjusted P values for multiple comparisons Position Supinated vs neutral Supinated vs pronated Neutral vs pronated Location PPRUJ vs MPRUJ PPRUJ vs DPRUJ MPRUJ vs DPRUJ Area Anterior vs midpoint Anterior vs posterior Midpoint vs posterior
<.0001 <.0001 <.0001 <.0001 0.1141 <.0001 <.0001
.0016 .0001 .087 .0001 .8998 .0001 .0001 .0001 .7433
DPRUJ, Distal proximal radioulnar joint; MPRUJ, middle proximal radioulnar joint; PPRUJ, proximal proximal radioulnar joint.
of pronation to assess rotation of the radius in relation to the radial articular fossa of the ulna. They also concluded that the radial head was ovoid, with a diameter difference of 2.5 mm at the level of the proximal radioulnar joint. They found that in supination, the axis of the maximum diameter is roughly parallel to the radial articular fossa of the ulna. King et al6 measured 28 cadaveric samples and 40 radiographs with a tactile probe coordinate measuring machine. They found that the radial head was variably offset on the axis of the neck (4.2 6 2.5 mm). They also found a poor correlation between radial head and medullary canal diameter, with a diameter difference of 1.7 to 0.7 mm. This again corresponded well with our calculated difference of 1.4 to 2.0 mm, depending on the level of the radial head measurements. Swieszkowski et al13 measured 34 cadaveric proximal radii with a coordinate measuring machine. The
mean 6 SD measurements were maximum diameter, 23.36 6 1.14 mm; radial head height, 10.14 6 1.38; and the radial head depth, 1.92 6 0.32 mm. This again correlated well with our findings of 22.3mm maximum diameter, 9.8-mm head length, and 2.3-mm depth of curvature. In our study, the maximum and minimum diameters of the radial head did not change with forearm rotation at the trough (level A) and the isthmus (level D). However, at levels B and C, the values varied significantly across forearm position. This may reflect the inconsistency and difficulty in determining the level of the flare and head–neck interface compared with the ease in identifying the trough and isthmus as bony landmarks. Therefore, our conclusions are based only on the values obtained at levels A and D. The radial head is, indeed, not perfectly circular. At the radial head trough, however, there was only a 6%
Itamura et al
J Shoulder Elbow Surg Volume 17, Number 2
353
Table IV Mean and SD of proximal radioulnar joint coverage by position and location (degrees) Forearm position Location PPRUJ MPRUJ DPRUJ
Supinated (n ¼ 22)
Neutral (n ¼ 21)
Pronated (n ¼ 20)
P
Overall (n ¼ 63)
Range
77.7 6 5.9 78.8 6 6.4 72.6 6 9.4
76.4 6 8.4 79.3 6 9.1 69.6 6 13.0
77.0 6 9.9 79.6 6 7.6 64.4 6 11.1
.7768 .148 .0169
77.76 8.0 79.2 6 7.6 69.0 6 11.6
55-99 61-100 47-97
P values for 2-factor analysis of variance Position (supinated, neutral, pronated) Location (PPRUJ, MPRUJ, DPRUJ) Position*location interaction Adjusted P values for multiple comparisons Position Supinated vs neutral Supinated vs pronated Neutral vs pronated Location PPRUJ vs MPRUJ PPRUJ vs DPRUJ MPRUJ vs DPRUJ
.6277 <.0001 .4755
.7353 .2556 .6835 .4124 <.0001 <.0001
DPRUJ, Distal proximal radioulnar joint; MPRUJ, middle proximal radioulnar joint; PPRUJ, proximal proximal radioulnar joint.
difference in the maximum and minimum diameters. To produce elliptical radial head components would also introduce rotational angulation. The data for the trough offset were inconclusive. The trough is roughly in the center of the radial head (50%), but with a large range of 39% to 63%. Again, manufacturing radial heads incorporating this wide range of trough offset would be very difficult. The medullary canal is even more elliptical, with a 16% difference between maximum and minimum diameters. Creating elliptical radial necks would introduce rotational angulation of the head on the stem and the stem into the canal. This would greatly increase the technical difficulty of performing a radial head replacement. At the radiocapitellar joint, the trough and ulnar lip are more congruent, with the joint distance increasing at the radial lip. Therefore, implants should be matched with either the trough or the ulnar lip rather than the radial lip. The joint is also more congruent in pronation than in supination; therefore, implants should fit tighter in pronation than in supination. In other words, they will naturally be looser in supination. The PRUJ is most congruent at the middle PRUJ, rather than the proximal or distal portion. It is also more congruent at the midportion or posterior aspects, with the joint distance increasing anteriorly. Implants should therefore match up with the middle PRUJ at either the middle or posterior aspect. The study had a few sources of possible bias. The cadaveric specimens were older and may have had arthritic changes. Osteophytes may have caused the measurements to overestimate the size of the true radial head. Cartilage wear may have caused the mea-
surements to underestimate the true joint distances. Being cadaveric specimens, the lack of in vivo soft tissue tensioning also may have affected the measurements. There was also investigator bias in the measurements of the radial head diameters. Three different people measured 1 forearm position each; however, this was addressed by comparing the measurements across forearm position. Overall, the measurements at level A and D were consistent across forearm positions (P > .05), as would be expected in areas with more consistent landmarks. Measurement error was also minimized by using the built-in software ruler. With the roughly 10-fold magnification of the specimens on the computer screen, one would need to make a full 1-cm error in plotting data points on the screen to create a 1-mm error in measurement. Clinical relevance
At our institutions, radial heads are implanted using a modified Boyd and Anderson posterior approach to the forearm. The proximal portion of the anconeus muscle and the elbow capsule are lifted off the ulna to allow later repair with suture anchors. This approach allows visualization of the ulnar lip and trough of the radial head as well as the posterior portion of the proximal radioulnar joint.11,12 These areas correspond to the most congruent portions of the radial head articulations. In conclusion, the elliptical shape of the native radial head is not replicated by the current cylindrical radial head prostheses. This difference is relatively small, however. The trough is roughly centered on the
354
Itamura et al
radial head, but with a large range. The elliptical medullary canal also has implications for radial head replacement. Creating elliptical radial heads and stems would introduce rotational angulation of the head on the stem and the stem into the canal. Therefore, introducing angulation, rotation, and offset would exponentially increase in technical difficulty in performing radial head replacements. Given all these factors, it is probably not cost-effective or technically feasible to manufacture elliptical radial heads and stems. Size mismatch is another common problem in radial head replacements. Radial heads sizes tend to run large and stem size choices tend to run large for the native canals. Potential problems include overstuffing the elbow or fracture of the radial neck and shaft. Shape mismatch can cause camming at the radiocapitellar and the proximal radioulnar joints. This may decrease stability, cause pain or stiffness, and lead to earlier prosthetic failure. Because it may not be feasible to create anatomic radial head replacements, it is important to maximize surgical technique. With the current cylindrical radial head prostheses, matching up only the most congruent portions of the articulations can avoid mismatch. This study points to 2 main areas: (1) the ulnar lip or trough of the radiocapitellar joint in pronation and (2) the posterior or midportion of the middle PRUJ. The clinical significance of this technique remains to be seen. We acknowledge the assistance of Chuck Nguyen, Alejandro Vasquez, Neil Parikh, Linda Chan, PhD, and Emily Ramicone, MS. REFERENCES
1. Beredjiklian PK, Nalbantoglu U, Potter HG, Hotchkiss RN. Prosthetic radial head components and proximal radial morphology: a mismatch. J Shoulder Elbow Surg 1999;8:471-5.
J Shoulder Elbow Surg March/April 2008
2. Captier G, Canovas F, Mercier N, Thomas E, Bonnel F. Biometry of radial head: biomechanical implications in pronation and supination. Surg Radiol Anat 2002;24:295-301. 3. Cone RO, Szabo R, Resnick D, Gelberman R, Taleisnik J, Gilula LA. Computed tomography of the normal radioulnar joints. Invest Radiol 1983;18:541-5. 4. Gupta GG, Lucas G, Hahn DL. Biomechanical and computer analysis of radial head prostheses. J Shoulder Elbow Surg 1997;6:37-48. 5. Itamura J, Roidis N, Mirzayan R, Vaishnav S, Learch T, Shean C. Radial head fractures: MRI evaluation of associated injuries. J Shoulder Elbow Surg 2005;14:421-4. 6. King GJ, Zarzour ZD, Patterson SD, Johnson JA. An anthropometric study of the radial head: implications in the design of a prosthesis. J Arthroplasty 2001;16:112-6. 7. Mahaisavariya B, Saekee B, Sitthiseripratip K, et al. Morphology of the radial head: a reverse engineering based evaluation using three-dimensional anatomical data of radial bone. Proc Inst Mech Eng [H] 2004;218:79-84. 8. Popovic N, Djekic J, Lemaire R, Gillet P. A comparative study between proximal radial morphology and the floating radial head prosthesis. J Shoulder Elbow Surg 2005;14:433-40. 9. Ring D. Open reduction and internal fixation of fractures of the radial head. Hand Clin 2004;20:415-27. 10. Roidis N, Stevanovic M, Martirosian A, Abbott DD, McPherson EJ, Itamura JM. A radiographic study of proximal radius anatomy with implications in radial head replacement. J Shoulder Elbow Surg 2003;12:380-4. 11. Roidis NT, Papadakis SA, Karachalios TS, Mirzayan R, Itamura JM. Radial head fractures. In: Mirzayan R, Itamura JM, editors. Shoulder and elbow trauma. New York, NY: Thieme Medical Publishers Inc; 2004. p. 22-35. 12. Roidis NT, Papadakis SA, Rigopoulos N, et al. Current concepts and controversies in the management of radial head fractures. Orthopedics 2006;29:904-16; quiz 917–8. 13. Swieszkowski W, Skalski K, Pomianowski S, Kedzior K. The anatomic features of the radial head and their implication for prosthesis design. Clin Biomech (Bristol, Avon) 2001;16:880-7. 14. van Riet RP, Van Glabbeek F, Neale PG, Bortier H, An KN, O’Driscoll SW. The noncircular shape of the radial head. J Hand Surg Am 2003;28:972-8.