- Email: [email protected]
A radiographic study of proximal radius anatomy with implications in radial head replacement
A radiographic study of proximal radius anatomy with implications in radial head replacement Nikolaos Roidis, MD, DSc, Milan Stevanovic, MD, PhD, Armen Martirosian, MD, Douglas D. Abbott, MD, Edward J. McPherson, MD, and John M. Itamura, MD,a Los Angeles, CA
Anteroposterior (AP) and lateral radiographs of 20 healthy volunteers’ forearms were taken in three views (full supination, neutral rotation, and full pronation). Radial head maximum diameter and angular measurements between the axis of forearm rotation (AFR) and the radial neck axis (RNA) were made with digital calipers. Repeated-measures analysis of variance revealed a statistically significant difference between the three AP groups, with supination having the smallest values (P ⬍ .0001), but not for the lateral groups (P ⫽ .128). Comparison of the AFR-RNA angle between the AP supinated position and the three lateral views revealed a statistically significant difference among all of the pairs, with the AP supinated position having the smallest values. The RNA most closely approximated the AFR with the forearm in the supinated position. For best approximating the native AFR during radial head replacement, the cut should be made perpendicular to the neck axis with the elbow extended and the forearm in the supinated position. (J Shoulder Elbow Surg 2003;12:380-4.)
O ne contemporary treatment for unreconstructable
radial head fractures is prosthetic radial head replacement.6,7,9 Historically, silastic implants were used, but the long-term outcomes of this prosthesis have not been favorable because of reports of implant failure and reactive synovitis, as well as its inferior mechanical characteristics.12 Metallic implants have recently become popular for prosthetic radial head replacement. Results by Knight et al9 and Harrington et al6 have shown favorable functional long-term outcome after radial head replacement using metallic prostheses.
From the Department of Orthopaedic Surgery, Keck School of Medicine, University of Southern California, University Hospital. Reprint requests: Nikolaos Roidis, MD, DSc, Consultant Orthopaedic Surgeon, University of Thessaly, 42 Pindou St, A Patissia, 11255, Athens, Greece (Hellenic Republic) (E-mail: [email protected]). Copyright © 2003 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2003/$35.00 ⫹ 0 doi:10.1016/S1058-2746(02)86881-1
380
A review of the literature shows a paucity of information about the proper position of the forearm for the radial neck osteotomy in prosthetic replacement. The axis of forearm rotation (AFR) represents the most important variable in normal forearm biomechanics and kinematics.10 The restoration of this axis is of paramount importance when a radial head implant is used. It is well known that if the prosthesis is not oriented properly to the AFR, a cam effect will occur at the radiocapitellar articulation with forearm rotation. This cam effect can lead to postoperative pain and decreased range of motion, as well as subluxation and dislocation of the prosthetic radial head implant.1,5,10 The purpose of this study was to document any elliptical shape of the radial head and determine the radiographic anatomy of the proximal radius in three different views (full supination, full pronation, and neutral rotation) to identify that position which has the smallest value for the angle between the AFR and the radial neck axis (RNA). It is our hypothesis that such a position should offer the optimal situation for the radial neck cut in radial head replacement, as it will approximate the normal biomechanical AFR. Our goal was to provide insight to optimize the radial neck osteotomy so that the normal AFR would not be substantially distorted. MATERIALS AND METHODS Twenty healthy volunteers with no history of any pathologic condition in the upper extremity were recruited for the purposes of this study. Any history of trauma in the upper extremity was considered an exclusion criterion to our study. No obvious bony deformity was noted on any of the participating subjects’ radiographs. There were 14 men and 6 women; the mean age was 32.3 years. Eighteen of the volunteers had radiographs of both forearms taken, from the elbow joint to the wrist joint, whereas two had films of their dominant forearm only. Standardized anteroposterior (AP) views in maximum active supination (palm up), neutral rotation, and maximum active pronation (palm down) were obtained of one or both sides of each subject. The forearm was positioned on the radiographic table with the elbow fully extended and the shoulder forward flexed 90°. Lateral views were obtained in the three different forearm rotational positions (thumb up, neutral, and thumb
Roidis et al
J Shoulder Elbow Surg Volume 12, Number 4
down). The lateral views were taken with the shoulder abducted 90° and the elbow flexed 90°. The same technician, using a standardized technique, obtained all radiologic views. The maximum diameter of the radial head was obtained in each of the three different views for all participants (AP and lateral). The center point of the radial head was defined. The AFR was drawn from the center of the radial head to the fovea of the distal ulna on all radiographs. Two lines spanning the outer cortices of the radial neck were drawn perpendicular to the lateral cortex of the neck at the distal margin of the radial head and at the proximal margin of the bicipital tuberosity. Next, the RNA was obtained as a line passing through the center of the radial head bisecting the previously described lines across the proximal and distal ends of the radial neck (Figure 1). The observers were blinded such that they did not know the identity of the subjects of the radiographs. In addition, the radiographs were randomized in such a way that all 6 radiographs of a single person (2 sets of 3 radiographs, AP and lateral) were never examined sequentially. All measurements were made with digital calipers (Mitutoyo Corp, Tokyo, Japan). The maximum diameter of the radial head and its center point were obtained with the digital calipers, whereas all of the angular measurements were performed with the use of a digital protractor (Mitutoyo Corp). All linear measurements of the radial head dimensions were corrected for standard radiologic magnification. No different magnification corrections for the AP and lateral views were used, as these were not considered likely to make any change in the angle measurements. The angles formed between the AFR and RNA were then tabulated in a spreadsheet for all views. Statistical analysis was performed with the Friedman test, Wilcoxon signed rank test, and Mann-Whitney U test, with P ⬍ .05 considered to be statistical significant. Original and repeat measurements of the AFR-RNA angle were made by two independent observers. The coefficient of variance was measured for intraobserver reliability, and the F test was performed to test the homogeneity of the variance for interobserver reliability.
RESULTS The mean, SD, and range of the AFR-RNA angles calculated from the radiographs are summarized for each forearm position in Tables I and II. The mean, SD, and range of the corrected values for the maximum diameter of the radial head across the three different views are summarized in Tables III and IV. There was a statistically significant difference in mean values between the three groups for the AP views (Friedman test, P ⫽ .000) (Figure 2) but not for the lateral views (Friedman test, P ⫽ .128) (Figure 3), with further pairwise comparisons also revealing a statistically significant difference in mean values between the three groups for the AP views (Wilcoxon signed ranks test: supination vs neutral, P ⫽ .000; supination vs pronation, P ⫽ .000; and neutral vs pronation, P ⫽ .002). Furthermore, comparisons of the AFR-RNA in the AP supinated position with the
381
Figure 1 AP radiograph in full supination demonstrating the AFR as drawn from the center of the radial head to the fovea of the distal ulna (solid black line). Two lines spanning the outer cortices of the radial neck are depicted perpendicular to the neck at the distal margin of the radial head and at the proximal margin of the bicipital tuberosity (proximal radius). The RNA is demonstrated as a line (dotted line, proximal radius) passing through the center of the radial head bisecting the two previous lines across the proximal and distal ends of the radial neck. As can easily be seen, in this case the AFR and RNA are almost identical.
Table I AFR-RNA findings: AP views AFR-RNA (°) Position
Mean
SD
Range
Supination Neutral Pronation
2.24 6.05 4.42
1.58 2.89 2.76
0–6.2 0–12 0–10.4
three different lateral positions (Figure 4) revealed statistically significant differences among all of the pairs (Mann-Whitney U test: AP supination vs thumb up lateral, P ⫽ .02; AP supination vs neutral lateral, P
382
Roidis et al
J Shoulder Elbow Surg July/August 2003
Table II AFR-RNA findings: Lateral views AFR-RNA (°) Position
Mean
SD
Range
Thumb up Neutral Thumb down
3.55 3.69 3.42
1.49 1.65 1.67
1.5–7 1.4–8.9 1–7
Table III Maximum diameter: AP views Maximum diameter (mm) Position
Mean
SD
Range
Supination Neutral Pronation
21.69 22.5 21.89
1.78 1.53 1.72
19.06–25.38 20–25.63 19.2–24.89
Table IV Maximum diameter: Lateral views Maximum diameter (mm) Position
Mean
SD
Range
Thumb up Neutral Thumb down
22.47 21.44 22.92
1.5 1.32 1.46
19.84–24.92 19.08–23.3 20.5–25.83
⫽ .01; and AP supination vs thumb down lateral, P ⫽ .03). The corrected values for the maximum diameter of the radial head presented with a statistically significant difference across the three different radiologic views for both AP and lateral radiographs (Friedman test, P ⫽ .000). The highest values of this variable in AP radiographs were observed in the neutral position, whereas the values in both supination and pronation were almost the same (Figures 5 and 6). Pairwise comparisons between the three different AP views revealed a statistically significant difference only for the neutral versus pronation group (Wilcoxon signed ranks test, P ⫽ .000) and neutral versus supination group (Wilcoxon signed ranks test, P ⫽ .000) but not for the supination versus pronation group (Wilcoxon signed ranks test, P ⫽ .053). All three different lateral views presented with statistically significant differences among each other: neutral versus thumb down (Wilcoxon signed ranks test, P ⫽ .000), neutral versus thumb up (Wilcoxon signed ranks test, P ⫽ .000), and thumb up versus thumb down (Wilcoxon signed ranks test, P ⫽ .001). As can be seen, there is a variability of the radial head diameter in the lateral views. The comparison between the two views reveals a wider range of values measured on the lateral views compared with the AP ones (Tables III
Figure 2 Box plot demonstrating the values of the AFR-RNA angle in AP views among the three different rotational elbow positions (full supination, neutral rotation, and full pronation). Box plots are defined as summary plots based on median values and quartiles. The box represents the interquartile range, which contains 50% of values. The whiskers are lines that extend from the box to the highest and lowest values. A line across the box indicates the median.
Figure 3 Box plot demonstrating the values of the AFR-RNA angle in lateral views among the three different rotational elbow positions (thumb up, neutral, and thumb down).
and IV). As previously stated, no different magnification corrections for the AP and lateral views were used. The magnification factor may have contributed to the variability of the results of the radial head diameter in the lateral views. The coefficient of variation was less than 5%. The F test between the two observers was not statistically significant for each radiograph set.
J Shoulder Elbow Surg Volume 12, Number 4
Figure 4 Box plot demonstrating the differences in the AFR-RNA angle between the AP supinated position and the three lateral views (thumb up, neutral, and thumb down).
Figure 5 Box plot demonstrating the values of the maximum diameter of the radial head in AP views among the three different rotational elbow positions (full supination, neutral rotation, and full pronation).
DISCUSSION Using any of the currently available radial head implants, one must strive to reproduce the normal AFR. So that anatomic radial head mechanical alignment can be restored most closely, the radial head prosthesis must be placed with the articular surface perpendicular to the AFR. This axis is not clinically apparent, however; some other reference must be sought to aid with prosthesis placement. The RNA, and especially the lateral cortex, may offer that reference point upon which the osteotomy can be easily planned.
Roidis et al
383
Figure 6 Box plot demonstrating the values of the maximum diameter of the radial head in lateral views among the three different rotational elbow positions (thumb up, neutral, and thumb down).
Our data show that in AP views (elbow extended) the RNA most closely approximates the AFR when the forearm is positioned in the supinated position. Most importantly, there is a statistically significant difference in the AFR-RNA angle between the AP supinated and the three lateral positions. In lateral views (elbow flexed 90°), there is no significant difference among the mean values for the three forearm rotational positions. Furthermore, the neutral lateral position, which corresponds to the AP pronated position that is traditionally used during the posterolateral approach to the proximal radius, presents with the highest range between minimum and maximum values. Given our findings, the best way to position the forearm for a radial neck cut during radial head replacement is with the elbow joint in extension and supination. In this position, if the neck osteotomy is made perpendicular to the lateral radial neck cortex, which is a liberally visible part after a posterolateral approach to the proximal part of the radius, it will have the greatest chance of most closely approximating the native AFR. All current commercially available metallic radial head monopolar prostheses are designed with the radial head placed perpendicular to the axis of the stem. The stem is designed to be placed down the radial shaft; the base of the prosthetic head invariably rests on the cut bony surface of the radial neck and assumes a position flush with it. Recent studies by various investigators have identified the mismatch between current prosthetic design and proximal radial morphology with respect to radial head shape and radial neck size.1,3,8 King et al8 have demonstrated that the native radial head is not perfectly
384
Roidis et al
circular or consistently elliptical. Our results with regard to radial head diameter measurements give further confirmation of the elliptical nature of the native radial head. These studies suggest that further study and refinement of prosthesis design are warranted. Traditionally, the radial neck osteotomy is done with the elbow flexed in a pronated position.4,11,13,14 This recommendation is predicated on an anatomic study by Strachan and Ellis11 in which the position of the posterior interosseous nerve in the cadaver forearm was described. They showed that pronation moved the posterior interosseous nerve more medially, by less than 1 cm, from the level of the elbow joint to the radial tubercle. They, therefore, recommended placing the forearm in pronation during exposure of the radial head to help minimize the chance of posterior interosseous nerve injury. We have found no published clinical correlation to this study to imply decreased nerve injury rates in practice with forearm pronation during the radial neck cut. Diliberti et al4 have defined a safe zone that helps to avoid injuring the posterior interosseous nerve during posterolateral approaches to the proximal part of the radius. Supination was found to decrease this zone, whereas flexion and extension of the elbow joint had no effect on the reported distances of the safe zone. This safe zone in supination was reported to have a mean of 52 ⫾ 7.8 mm of the lateral aspect of the radius.4 A typical Kocher exposure allows suboptimal exposure for a radial neck cut and offers difficult access for broaching and implant positioning. We prefer a more extensile approach such as that of Cohen and Hastings,2 which offers improved access, ligament sparing, and perhaps less chance of posterior interosseous nerve injury. Consequently, a radial neck osteotomy in supination may not be so dangerous if it contributes to the normal prosthetic component alignment. A reasonable criticism regarding the results of this study would be that the mean values of these angles are in the range of some degrees and therefore might be considered coincidental, given the manual error of a few degrees inherent in the process of making a radial neck osteotomy during surgery. On the other hand, contemporary reconstructive surgery is based on proper component alignment, and misalignment of even some degrees may lead to unfavorable results. A radial head prosthesis’ misalignment relative to the normal AFR may accentuate and accelerate any unfavorable postoperative results because of the elliptical shape and high degree of rotational motion of the radial head. The evolution of radial head replacement is still
J Shoulder Elbow Surg July/August 2003
ongoing.1,5,7 The ideal radial head prosthetic implant has yet to be designed. As current design parameters are being worked out to improve the prosthesis, we must strive to advance our surgical technique. Our radiographic study has shown a clear statistical significance with respect to the AFR-RNA angle and forearm rotational position. Although there may be some other components to radial head seating within the radial neck in addition to the relationship of the prosthesis to the RNA, our data allow us to suggest that, to best approximate the native AFR during surgery, the radial neck osteotomy should be made perpendicular to the lateral cortex with the forearm in extension and supination. In that position, the important biomechanical AFR can be most closely restored. 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. 2. Cohen MS, Hastings H II. Post-traumatic contracture of the elbow. Operative release using a lateral collateral ligament sparing approach. J Bone Joint Surg Br 1998;80:805-12. 3. Cone RO, Szabo R, Resnick D, et al. Computed tomography of the normal radioulnar joints. Invest Radiol 1983;18:541-5. 4. Diliberti T, Botte MJ, Abrams RA. Anatomical considerations regarding the posterior interosseous nerve during posterolateral approaches to the proximal part of the radius. J Bone Joint Surg Am 2000;82:809-13. 5. Gupta GG, Lucas G, Hahn DL. Biomechanical and computer analysis of radial head prostheses. J Shoulder Elbow Surg 1997; 6:37-48. 6. Harrington IJ, Sekyi-Otu A, Barrington TW, Evans DC, Tuli V. The functional outcome with metallic radial head implants in the treatment of unstable elbow fractures: a long-term review. J Trauma 2001;50:46-52. 7. Judet T, Garreau de Loubresse C, Piriou P, Charnley G. A floating prosthesis for radial-head fractures. J Bone Joint Surg Br 1996; 78:244-9. 8. King GJ, Zarzour ZD, Patterson SD, Johnson JA. An anthropometric study of the radial head. J Arthroplasty 2001;16:112-6. 9. Knight DJ, Rymaszewski LA, Amis AA, Miller JH. Primary replacement of the fractured radial head with a metal prosthesis. J Bone Joint Surg Br 1993;75:572-6. 10. Morrey BF. Radial head fractures. In: Morrey BF, editor. The elbow and its disorders. Philadelphia: Saunders; 2000. p. 34164. 11. Strachan JCH, Ellis BW. Vulnerability of the posterior interosseous nerve during radial head resection. J Bone Joint Surg Br 1981; 53:320-3. 12. Vanderwilde RS, Morrey BF, Melberg MW, Vinh TN. Inflammatory arthritis after failure of silicone rubber replacement of the radial head. J Bone Joint Surg Br 1994;76:78-81. 13. Witt J, Diliberti T, Botte MJ, Abrams RA. Toward safe exposure of the proximal part of the radius: landmarks and measurements. J Bone Joint Surg Am 2001;83:1589-90. 14. Witt JD, Kamineni S. The posterior interosseous nerve and the posterolateral approach to the proximal radius. J Bone Joint Surg Br 1998;80:240-2.
Anteroposterior (AP) and lateral radiographs of 20 healthy volunteers’ forearms were taken in three views (full supination, neutral rotation, and full pronation). Radial head maximum diameter and angular measurements between the axis of forearm rotation (AFR) and the radial neck axis (RNA) were made with digital calipers. Repeated-measures analysis of variance revealed a statistically significant difference between the three AP groups, with supination having the smallest values (P ⬍ .0001), but not for the lateral groups (P ⫽ .128). Comparison of the AFR-RNA angle between the AP supinated position and the three lateral views revealed a statistically significant difference among all of the pairs, with the AP supinated position having the smallest values. The RNA most closely approximated the AFR with the forearm in the supinated position. For best approximating the native AFR during radial head replacement, the cut should be made perpendicular to the neck axis with the elbow extended and the forearm in the supinated position. (J Shoulder Elbow Surg 2003;12:380-4.)
O ne contemporary treatment for unreconstructable
radial head fractures is prosthetic radial head replacement.6,7,9 Historically, silastic implants were used, but the long-term outcomes of this prosthesis have not been favorable because of reports of implant failure and reactive synovitis, as well as its inferior mechanical characteristics.12 Metallic implants have recently become popular for prosthetic radial head replacement. Results by Knight et al9 and Harrington et al6 have shown favorable functional long-term outcome after radial head replacement using metallic prostheses.
From the Department of Orthopaedic Surgery, Keck School of Medicine, University of Southern California, University Hospital. Reprint requests: Nikolaos Roidis, MD, DSc, Consultant Orthopaedic Surgeon, University of Thessaly, 42 Pindou St, A Patissia, 11255, Athens, Greece (Hellenic Republic) (E-mail: [email protected]). Copyright © 2003 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2003/$35.00 ⫹ 0 doi:10.1016/S1058-2746(02)86881-1
380
A review of the literature shows a paucity of information about the proper position of the forearm for the radial neck osteotomy in prosthetic replacement. The axis of forearm rotation (AFR) represents the most important variable in normal forearm biomechanics and kinematics.10 The restoration of this axis is of paramount importance when a radial head implant is used. It is well known that if the prosthesis is not oriented properly to the AFR, a cam effect will occur at the radiocapitellar articulation with forearm rotation. This cam effect can lead to postoperative pain and decreased range of motion, as well as subluxation and dislocation of the prosthetic radial head implant.1,5,10 The purpose of this study was to document any elliptical shape of the radial head and determine the radiographic anatomy of the proximal radius in three different views (full supination, full pronation, and neutral rotation) to identify that position which has the smallest value for the angle between the AFR and the radial neck axis (RNA). It is our hypothesis that such a position should offer the optimal situation for the radial neck cut in radial head replacement, as it will approximate the normal biomechanical AFR. Our goal was to provide insight to optimize the radial neck osteotomy so that the normal AFR would not be substantially distorted. MATERIALS AND METHODS Twenty healthy volunteers with no history of any pathologic condition in the upper extremity were recruited for the purposes of this study. Any history of trauma in the upper extremity was considered an exclusion criterion to our study. No obvious bony deformity was noted on any of the participating subjects’ radiographs. There were 14 men and 6 women; the mean age was 32.3 years. Eighteen of the volunteers had radiographs of both forearms taken, from the elbow joint to the wrist joint, whereas two had films of their dominant forearm only. Standardized anteroposterior (AP) views in maximum active supination (palm up), neutral rotation, and maximum active pronation (palm down) were obtained of one or both sides of each subject. The forearm was positioned on the radiographic table with the elbow fully extended and the shoulder forward flexed 90°. Lateral views were obtained in the three different forearm rotational positions (thumb up, neutral, and thumb
Roidis et al
J Shoulder Elbow Surg Volume 12, Number 4
down). The lateral views were taken with the shoulder abducted 90° and the elbow flexed 90°. The same technician, using a standardized technique, obtained all radiologic views. The maximum diameter of the radial head was obtained in each of the three different views for all participants (AP and lateral). The center point of the radial head was defined. The AFR was drawn from the center of the radial head to the fovea of the distal ulna on all radiographs. Two lines spanning the outer cortices of the radial neck were drawn perpendicular to the lateral cortex of the neck at the distal margin of the radial head and at the proximal margin of the bicipital tuberosity. Next, the RNA was obtained as a line passing through the center of the radial head bisecting the previously described lines across the proximal and distal ends of the radial neck (Figure 1). The observers were blinded such that they did not know the identity of the subjects of the radiographs. In addition, the radiographs were randomized in such a way that all 6 radiographs of a single person (2 sets of 3 radiographs, AP and lateral) were never examined sequentially. All measurements were made with digital calipers (Mitutoyo Corp, Tokyo, Japan). The maximum diameter of the radial head and its center point were obtained with the digital calipers, whereas all of the angular measurements were performed with the use of a digital protractor (Mitutoyo Corp). All linear measurements of the radial head dimensions were corrected for standard radiologic magnification. No different magnification corrections for the AP and lateral views were used, as these were not considered likely to make any change in the angle measurements. The angles formed between the AFR and RNA were then tabulated in a spreadsheet for all views. Statistical analysis was performed with the Friedman test, Wilcoxon signed rank test, and Mann-Whitney U test, with P ⬍ .05 considered to be statistical significant. Original and repeat measurements of the AFR-RNA angle were made by two independent observers. The coefficient of variance was measured for intraobserver reliability, and the F test was performed to test the homogeneity of the variance for interobserver reliability.
RESULTS The mean, SD, and range of the AFR-RNA angles calculated from the radiographs are summarized for each forearm position in Tables I and II. The mean, SD, and range of the corrected values for the maximum diameter of the radial head across the three different views are summarized in Tables III and IV. There was a statistically significant difference in mean values between the three groups for the AP views (Friedman test, P ⫽ .000) (Figure 2) but not for the lateral views (Friedman test, P ⫽ .128) (Figure 3), with further pairwise comparisons also revealing a statistically significant difference in mean values between the three groups for the AP views (Wilcoxon signed ranks test: supination vs neutral, P ⫽ .000; supination vs pronation, P ⫽ .000; and neutral vs pronation, P ⫽ .002). Furthermore, comparisons of the AFR-RNA in the AP supinated position with the
381
Figure 1 AP radiograph in full supination demonstrating the AFR as drawn from the center of the radial head to the fovea of the distal ulna (solid black line). Two lines spanning the outer cortices of the radial neck are depicted perpendicular to the neck at the distal margin of the radial head and at the proximal margin of the bicipital tuberosity (proximal radius). The RNA is demonstrated as a line (dotted line, proximal radius) passing through the center of the radial head bisecting the two previous lines across the proximal and distal ends of the radial neck. As can easily be seen, in this case the AFR and RNA are almost identical.
Table I AFR-RNA findings: AP views AFR-RNA (°) Position
Mean
SD
Range
Supination Neutral Pronation
2.24 6.05 4.42
1.58 2.89 2.76
0–6.2 0–12 0–10.4
three different lateral positions (Figure 4) revealed statistically significant differences among all of the pairs (Mann-Whitney U test: AP supination vs thumb up lateral, P ⫽ .02; AP supination vs neutral lateral, P
382
Roidis et al
J Shoulder Elbow Surg July/August 2003
Table II AFR-RNA findings: Lateral views AFR-RNA (°) Position
Mean
SD
Range
Thumb up Neutral Thumb down
3.55 3.69 3.42
1.49 1.65 1.67
1.5–7 1.4–8.9 1–7
Table III Maximum diameter: AP views Maximum diameter (mm) Position
Mean
SD
Range
Supination Neutral Pronation
21.69 22.5 21.89
1.78 1.53 1.72
19.06–25.38 20–25.63 19.2–24.89
Table IV Maximum diameter: Lateral views Maximum diameter (mm) Position
Mean
SD
Range
Thumb up Neutral Thumb down
22.47 21.44 22.92
1.5 1.32 1.46
19.84–24.92 19.08–23.3 20.5–25.83
⫽ .01; and AP supination vs thumb down lateral, P ⫽ .03). The corrected values for the maximum diameter of the radial head presented with a statistically significant difference across the three different radiologic views for both AP and lateral radiographs (Friedman test, P ⫽ .000). The highest values of this variable in AP radiographs were observed in the neutral position, whereas the values in both supination and pronation were almost the same (Figures 5 and 6). Pairwise comparisons between the three different AP views revealed a statistically significant difference only for the neutral versus pronation group (Wilcoxon signed ranks test, P ⫽ .000) and neutral versus supination group (Wilcoxon signed ranks test, P ⫽ .000) but not for the supination versus pronation group (Wilcoxon signed ranks test, P ⫽ .053). All three different lateral views presented with statistically significant differences among each other: neutral versus thumb down (Wilcoxon signed ranks test, P ⫽ .000), neutral versus thumb up (Wilcoxon signed ranks test, P ⫽ .000), and thumb up versus thumb down (Wilcoxon signed ranks test, P ⫽ .001). As can be seen, there is a variability of the radial head diameter in the lateral views. The comparison between the two views reveals a wider range of values measured on the lateral views compared with the AP ones (Tables III
Figure 2 Box plot demonstrating the values of the AFR-RNA angle in AP views among the three different rotational elbow positions (full supination, neutral rotation, and full pronation). Box plots are defined as summary plots based on median values and quartiles. The box represents the interquartile range, which contains 50% of values. The whiskers are lines that extend from the box to the highest and lowest values. A line across the box indicates the median.
Figure 3 Box plot demonstrating the values of the AFR-RNA angle in lateral views among the three different rotational elbow positions (thumb up, neutral, and thumb down).
and IV). As previously stated, no different magnification corrections for the AP and lateral views were used. The magnification factor may have contributed to the variability of the results of the radial head diameter in the lateral views. The coefficient of variation was less than 5%. The F test between the two observers was not statistically significant for each radiograph set.
J Shoulder Elbow Surg Volume 12, Number 4
Figure 4 Box plot demonstrating the differences in the AFR-RNA angle between the AP supinated position and the three lateral views (thumb up, neutral, and thumb down).
Figure 5 Box plot demonstrating the values of the maximum diameter of the radial head in AP views among the three different rotational elbow positions (full supination, neutral rotation, and full pronation).
DISCUSSION Using any of the currently available radial head implants, one must strive to reproduce the normal AFR. So that anatomic radial head mechanical alignment can be restored most closely, the radial head prosthesis must be placed with the articular surface perpendicular to the AFR. This axis is not clinically apparent, however; some other reference must be sought to aid with prosthesis placement. The RNA, and especially the lateral cortex, may offer that reference point upon which the osteotomy can be easily planned.
Roidis et al
383
Figure 6 Box plot demonstrating the values of the maximum diameter of the radial head in lateral views among the three different rotational elbow positions (thumb up, neutral, and thumb down).
Our data show that in AP views (elbow extended) the RNA most closely approximates the AFR when the forearm is positioned in the supinated position. Most importantly, there is a statistically significant difference in the AFR-RNA angle between the AP supinated and the three lateral positions. In lateral views (elbow flexed 90°), there is no significant difference among the mean values for the three forearm rotational positions. Furthermore, the neutral lateral position, which corresponds to the AP pronated position that is traditionally used during the posterolateral approach to the proximal radius, presents with the highest range between minimum and maximum values. Given our findings, the best way to position the forearm for a radial neck cut during radial head replacement is with the elbow joint in extension and supination. In this position, if the neck osteotomy is made perpendicular to the lateral radial neck cortex, which is a liberally visible part after a posterolateral approach to the proximal part of the radius, it will have the greatest chance of most closely approximating the native AFR. All current commercially available metallic radial head monopolar prostheses are designed with the radial head placed perpendicular to the axis of the stem. The stem is designed to be placed down the radial shaft; the base of the prosthetic head invariably rests on the cut bony surface of the radial neck and assumes a position flush with it. Recent studies by various investigators have identified the mismatch between current prosthetic design and proximal radial morphology with respect to radial head shape and radial neck size.1,3,8 King et al8 have demonstrated that the native radial head is not perfectly
384
Roidis et al
circular or consistently elliptical. Our results with regard to radial head diameter measurements give further confirmation of the elliptical nature of the native radial head. These studies suggest that further study and refinement of prosthesis design are warranted. Traditionally, the radial neck osteotomy is done with the elbow flexed in a pronated position.4,11,13,14 This recommendation is predicated on an anatomic study by Strachan and Ellis11 in which the position of the posterior interosseous nerve in the cadaver forearm was described. They showed that pronation moved the posterior interosseous nerve more medially, by less than 1 cm, from the level of the elbow joint to the radial tubercle. They, therefore, recommended placing the forearm in pronation during exposure of the radial head to help minimize the chance of posterior interosseous nerve injury. We have found no published clinical correlation to this study to imply decreased nerve injury rates in practice with forearm pronation during the radial neck cut. Diliberti et al4 have defined a safe zone that helps to avoid injuring the posterior interosseous nerve during posterolateral approaches to the proximal part of the radius. Supination was found to decrease this zone, whereas flexion and extension of the elbow joint had no effect on the reported distances of the safe zone. This safe zone in supination was reported to have a mean of 52 ⫾ 7.8 mm of the lateral aspect of the radius.4 A typical Kocher exposure allows suboptimal exposure for a radial neck cut and offers difficult access for broaching and implant positioning. We prefer a more extensile approach such as that of Cohen and Hastings,2 which offers improved access, ligament sparing, and perhaps less chance of posterior interosseous nerve injury. Consequently, a radial neck osteotomy in supination may not be so dangerous if it contributes to the normal prosthetic component alignment. A reasonable criticism regarding the results of this study would be that the mean values of these angles are in the range of some degrees and therefore might be considered coincidental, given the manual error of a few degrees inherent in the process of making a radial neck osteotomy during surgery. On the other hand, contemporary reconstructive surgery is based on proper component alignment, and misalignment of even some degrees may lead to unfavorable results. A radial head prosthesis’ misalignment relative to the normal AFR may accentuate and accelerate any unfavorable postoperative results because of the elliptical shape and high degree of rotational motion of the radial head. The evolution of radial head replacement is still
J Shoulder Elbow Surg July/August 2003
ongoing.1,5,7 The ideal radial head prosthetic implant has yet to be designed. As current design parameters are being worked out to improve the prosthesis, we must strive to advance our surgical technique. Our radiographic study has shown a clear statistical significance with respect to the AFR-RNA angle and forearm rotational position. Although there may be some other components to radial head seating within the radial neck in addition to the relationship of the prosthesis to the RNA, our data allow us to suggest that, to best approximate the native AFR during surgery, the radial neck osteotomy should be made perpendicular to the lateral cortex with the forearm in extension and supination. In that position, the important biomechanical AFR can be most closely restored. 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. 2. Cohen MS, Hastings H II. Post-traumatic contracture of the elbow. Operative release using a lateral collateral ligament sparing approach. J Bone Joint Surg Br 1998;80:805-12. 3. Cone RO, Szabo R, Resnick D, et al. Computed tomography of the normal radioulnar joints. Invest Radiol 1983;18:541-5. 4. Diliberti T, Botte MJ, Abrams RA. Anatomical considerations regarding the posterior interosseous nerve during posterolateral approaches to the proximal part of the radius. J Bone Joint Surg Am 2000;82:809-13. 5. Gupta GG, Lucas G, Hahn DL. Biomechanical and computer analysis of radial head prostheses. J Shoulder Elbow Surg 1997; 6:37-48. 6. Harrington IJ, Sekyi-Otu A, Barrington TW, Evans DC, Tuli V. The functional outcome with metallic radial head implants in the treatment of unstable elbow fractures: a long-term review. J Trauma 2001;50:46-52. 7. Judet T, Garreau de Loubresse C, Piriou P, Charnley G. A floating prosthesis for radial-head fractures. J Bone Joint Surg Br 1996; 78:244-9. 8. King GJ, Zarzour ZD, Patterson SD, Johnson JA. An anthropometric study of the radial head. J Arthroplasty 2001;16:112-6. 9. Knight DJ, Rymaszewski LA, Amis AA, Miller JH. Primary replacement of the fractured radial head with a metal prosthesis. J Bone Joint Surg Br 1993;75:572-6. 10. Morrey BF. Radial head fractures. In: Morrey BF, editor. The elbow and its disorders. Philadelphia: Saunders; 2000. p. 34164. 11. Strachan JCH, Ellis BW. Vulnerability of the posterior interosseous nerve during radial head resection. J Bone Joint Surg Br 1981; 53:320-3. 12. Vanderwilde RS, Morrey BF, Melberg MW, Vinh TN. Inflammatory arthritis after failure of silicone rubber replacement of the radial head. J Bone Joint Surg Br 1994;76:78-81. 13. Witt J, Diliberti T, Botte MJ, Abrams RA. Toward safe exposure of the proximal part of the radius: landmarks and measurements. J Bone Joint Surg Am 2001;83:1589-90. 14. Witt JD, Kamineni S. The posterior interosseous nerve and the posterolateral approach to the proximal radius. J Bone Joint Surg Br 1998;80:240-2.