- Email: [email protected]
Biomechanical comparison of different pin configurations during percutaneous pinning for the treatment of proximal humeral fractures
Biomechanical comparison of different pin configurations during percutaneous pinning for the treatment of proximal humeral fractures Chunyan Jiang, MD, PhD, Yiming Zhu, MD, Manyi Wang, MD, and Guowei Rong, MD, Beijing, China
We investigate the influence on fracture stability of different pin configurations during percutaneous pinning for the treatment of proximal humeral fractures. We performed a matched-paired study of 18 pairs of adult fresh-frozen humeri (36 humeri), which were divided into 4 groups. A 2-part surgical neck fracture model was used in all humeri, and 4 terminal threaded pins (2.5 mm in diameter) were used for fixation. Parallel-type pinning (box type) was carried out in 2 groups, and convergent-type pinning (fan-shaped type) was used in the other 2 groups. For each specimen, both anti-shear ultimate load and anti-torsion ultimate load were measured. There was no statistical difference between the parallel pin construct and convergent construct with regard to anti-shear resistance (P ⫽ .73). However, the parallel pin construct had a significant advantage over the convergent construct regarding anti-torsion resistance. The parallel pin construct has better torsional stability when 1 cm is used for the pin-to-pin distance. We suggest that parallel pin fixation should be applied whenever possible. (J Shoulder Elbow Surg 2007;16:235-239.)
I mproved results can be expected with surgical treat-
ment for displaced proximal humeral fractures compared with conservative treatment. Closed reduction with percutaneous pin fixation has been developed in the past decade. Because of this technique’s minimally invasive nature, further devitalization of the fracture fragments can be avoided. Patients can benefit from this technique with regard to fracture healing and early postoperative rehabilitation, but mechanical stability of pin fixation is still a concern with this type of proximal humeral fracture fixation.
From the School of Medicine, Peking University. Reprint requests: Chunyan Jiang, MD, PhD, Beijing Ji Shui Tan Hospital, 4th Medical Center, School of Medicine, Peking University, No. 31 Xinjiekoudongjie, 100035, Beijing, China (E-mail: [email protected]). Copyright © 2007 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2007/$32.00 doi:10.1016/j.jse.2006.05.011
The anatomy of the proximal humerus is similar to that of the proximal femur to a certain extent. The mechanical advantage of cannulated screws in a parallel configuration over cross-type configuration in femoral neck fracture fixation is well accepted by most orthopaedic surgeons. In reviewing related articles, however, no report on parallel pin fixation for proximal humeral fractures was found, nor has any biomechanical study been reported. We carried out a set of biomechanical studies on cadaveric models to investigate and compare mechanical stability between parallel fixation (box type) and convergent fixation (fan-shaped type). MATERIALS AND METHODS We divided 36 fresh-frozen adult humeri (18 pairs) into 2 groups, with 18 humeri (9 pairs) in each: anti-torsional stability test group (group T) and anti-shear stability test group (group S). Each group was also subdivided based on each side of a single individual into 2 subgroups, with 9 humeri in each, as follows: anti-torsional stability test group with parallel pin construct (group TB), anti-torsional stability test group with convergent pin construct (group TC), antishear stability test group with parallel pin construct (group SB), and anti-shear stability test group with convergent pin construct (group SC). The humeri were kept in saline solution at room temperature for 12 hours before testing and were soaked in saline solution throughout the process. Dual-energy bone density scanning had been used to elicit the level of osteoporosis of the cadaver. A 2-tailed paired t test was used to compare the bone density between the 2 subgroups (TB vs TC and SB vs SC). A 2-part surgical neck fracture was created in all specimens in a consistent way. An osteotomy was made by use of an oscillating saw with a 1-mm blade. Fixation via a parallel configuration (box type) with 4 AO threaded pins (2.5 mm in diameter) was used in groups TB and SB. Fixation via a convergent configuration (fan-shaped type) with 4 AO threaded pins was used in groups TC and SC. All cadaveric specimens were prepared by a single surgeon.
Determination of osteotomy line The baseline was a horizontal line made at 3 cm below the lateral cortex of the top of the greater tuberosity. The osteotomy line was angled 20° from the baseline starting at
235
236
Jiang et al
J Shoulder Elbow Surg March/April 2007
Figure 1 Two-part surgical neck fracture model and entry point of parallel pin construct. Gray line, Baseline; white line, osteotomy line. Figure 3 Anti-shear fixation jig.
Entry point of pins in convergent configuration (groups TC and SC)
Figure 2 Two-part surgical neck fracture model and entry point of convergent pin construct. Gray line, Baseline; white line, osteotomy line.
the junction between the baseline and medial margin of the humeral head (Figures 1 and 2).
Fixation of fracture models The pins were angled 45° to the humeral shaft in the coronal plane and 30° in the sagittal plane.
Entry point of pins in parallel configuration (groups TB and SB) The medial 2 entry points were made 0.5 cm proximal and distal to the points at which the baseline intersected with the lateral wall of the bicipital groove. The lateral 2 entry points were made 1 cm lateral to the medial 2 entry points (Figure 1). A specially designed parallel drilling sleeve with a 1-cm interval was used during pin fixation.
The most medial entry point was determined by the point at which the baseline intersected with the medial wall of the bicipital groove. The other 3 entry points were made consecutively with 1-cm intervals laterally on the same line, except the second and third entry points were made 0.5 cm proximal and distal to the baseline, respectively (Figure 2). Two sets of fixation jigs were designed for anti-shear stability and anti-torsion stability testing. For the anti-shear fixation jig (applied in groups TB and TC), the cadaveric humerus was clamped 30° away from the vertical line, and the compression tip was applied on the humeral head just lateral to the articular margin. The distance between the medial point of the osteotomy line and the proximal end of the clamp was 8 cm. A supporting metal plate was placed on the medial cortex of the humeral shaft (Figure 3). For the antitorsion fixation jig (applied in groups SB and SC), the humeral head was potted in bone cement proximal to the osteotomy line. A 4.0-mm Steinmann pin was drilled through the center of the humeral shaft 12 cm distal to the osteotomy line, and the compression tip (with a groove compatible with the Steinmann pin) was applied 6 cm away from the humeral shaft. By use of downward motion of the tip, torsion was produced at the fracture site (Figure 4). An Instron 5566 mechanical testing machine (Norwood, MA) was used. The force-applying probe was set in an isometric downward direction. The rate of the applied force was 1 cm/min. The machine was stopped when plastic deformation occurred, and the ultimate load was recorded. The ultimate load was defined as the point at which elastic deformation became plastic deformation. In a stress-displacement curve, ultimate load represents the point at which a straight line turns into a curved line. All data from the 4 subgroups were evaluated statistically by 2-tailed paired t tests to detect any significant difference between groups TB and TS and between groups SB and SC.
Jiang et al
J Shoulder Elbow Surg Volume 16, Number 2
237
Table II Dual-energy bone density scan values in anti-shear stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Figure 4 Anti-torsion fixation jig.
Group SB (g/cm2)
Group SC (g/cm2)
0.9623 0.7986 0.8141 0.9242 0.673 0.959 0.8465 1.039 0.7463 0.8625 0.117
0.8582 0.8138 0.9115 0.8581 0.662 1.051 0.8193 0.9878 0.7289 0.8545 0.120
P ⫽ .730.
Table I Dual-energy bone density scan values in anti-torsional stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group TB (g/cm2)
Group TC (g/cm2)
0.6456 0.8062 0.7405 1.001 0.8399 0.7991 0.8684 0.8795 1.079 0.8510 0.130
0.6398 0.955 0.7393 0.8845 0.8637 0.7902 0.857 1.026 0.7848 0.8378 0.115
P ⫽ .774.
RESULTS The bone density of all specimens was evaluated by dual-energy bone density scanning. The mean values for the 2 subgroups in the anti-torsion test group (TB and TC) were 0.851 ⫾ 0.130 g/cm2 and 0.838 ⫾ 0.115 g/cm2, respectively. No significant difference was found between groups TB and TC (P ⫽ .77) (Tables I and II). The mean values for the 2 subgroups in the anti-shear test group (SB and SC) were 0.863 ⫾ 0.117 g/cm2 and 0.855 ⫾ 0.120 g/cm2, respectively. No significant difference was found between groups SB and SC (P ⫽ .73) (Tables I and II). The ultimate load was 226.65 ⫾ 38.36 N in group TB and 186.97 ⫾ 61.25 N in group TC. A significant difference was found between groups TB and TC; the parallel configuration group had a statistically significant advantage over the convergent configuration group for anti-torsion stability (P ⫽ .04). The ultimate load was 637.84 ⫾ 174.13 N in group SB and 628.59 ⫾ 189.50 N in group SC. No
Table III Ultimate load in anti-torsional stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group TB (N)
Group TC (N)
286.12 201.63 240.5 281.26 196.65 185.62 241.15 220.34 186.61 226.653 38.363
250.15 172.74 152.81 239.07 107.99 247.27 238.52 184.16 90.04 186.972 61.248
P ⫽ .043.
significant difference was found between groups SB and SC (P ⫽ .73), suggesting that there is no statistically significant difference between the parallel configuration group and the convergent configuration group for anti-shear stability (Tables III and IV). DISCUSSION Good results have been reported with threaded pin fixation for the treatment of proximal fracture,1,3,8,11 but the mechanical stability of pin fixation is still a concern in clinical practice. Pin back-out and loss of fixation are not uncommon and, thus, influence the proper healing process of a fracture.3,11 Our study is focused on how to improve fixation stability by modifying pin configuration. It is well accepted that a parallel configuration in cannulated screw fixation is important in treating femoral neck fractures, not only because the parallel screw construct is an important prerequisite for sliding compression of the fracture line but also because of its
238
Jiang et al
J Shoulder Elbow Surg March/April 2007
Table IV Ultimate load in anti-shear stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group SB (N)
Group SC (N)
933.55 370.05 607.11 476.72 584.13 855.39 698.12 587.12 628.34 637.837 174.130
1027.46 435.49 566.57 465.85 684.56 804.83 634.28 496.64 541.62 628.589 189.049
P ⫽ .725.
better mechanical stability compared with the convergent screw construct.7,9 The anatomy of the proximal humerus is similar to that of the proximal femur to a certain extent. Theoretically, a parallel pin construct should have a better mechanical advantage than the convergent pin construct. Most authors suggest that a convergent pin configuration will achieve the best stability in proximal humeral fracture fixation.1,2,4,5,10 In our review of the relevant literature, no clinical biomechanical report or study comparing parallel and convergent pin configuration in the proximal humeral fracture fixation was found, and there is only one article referring to the biomechanical characteristics of the parallel pin construct.5 To prove this theory, we designed a cadaveric fracture model to investigate the influence of pin construct on fixation stability. We compared 2 types of configuration: parallel and convergent. Both configurations included 4 pins in retrograde fixation. Some authors suggest that, with an additional 1 or 2 antegrade pins through the greater tuberosity to the medial cortex, the whole construct will have better stability.1,2 Koval et al5 showed that the best mechanical stability is achieved in osteoporotic bone with 3 retrograde pins in a convergent construct combined with 1 antegrade pin. Naidu et al6 also suggested that an additional 2 antegrade pins stabilizing the greater tuberosity to the medial cortex will improve both axial and rotational stability. Although an antegrade pin could improve fixation stability, the end of the pin can cause impingement under the acromion during shoulder abduction and make early postoperative rehabilitation impossible. In our experience, satisfactory stability can be achieved with 3 or 4 retrograde pins with appropriate entry points and positioning without hampering postoperative rehabilitation. Therefore, the cross-type pin construct with antegrade pin fixation was not included in this study.
We compared both sides of a single individual to exclude the influence of size or morphologic differences between specimens. We also matched all specimens by bone density scan to ensure that there was no significant difference between subgroups, thus excluding the influence of osteoporosis on fixation stability. The results proved our hypothesis: although no significant difference was found with regard to anti-shear stability, the parallel configuration does have better torsional stability than the convergent configuration. The clinical application of a parallel pin construct for proximal humeral fracture fixation is not as easy as it appears in a cadaveric model. Because of its minimally invasive nature, percutaneous pinning will definitely provide limited access to the fracture site. We cannot expect to find a broad and flat lateral humeral cortex like the proximal femur. The mass of the deltoid muscle will further interfere with appropriate pin orientation, resulting in possible pin slippage on the lateral cortex of the humerus. Moreover, parallel pin fixation may be impossible in small individuals. The distance between pins plays an important role in the mechanical properties of the parallel construct. Our finding that the parallel configuration provides better torsional stability than the convergent configuration is based on a very important premise: the distance between pins is 1 cm. We do not know the limit at which the parallel configuration still has better torsional stability over the convergent configuration, but we do know that with a decrease in this distance, the parallel construct will eventually lose its mechanical advantage. Our suggestion is to use parallel pin fixation whenever possible to improve fracture stability. A specially designed parallel drill sleeve with a 1-cm pin-to-pin distance and a serrated tip enhancing cortical purchase is recommended for the clinical application of parallel pin fixation. If this is not possible in a small individual, the convergent configuration is still a good choice for percutaneous pinning. The limitation of this study is that all specimens were tested to failure in a single loading fashion. The mechanism of fixation failure is very complicated and multifactorial. Fatigue failure may also play an important role. Our study is based on a cadaveric model; in vivo biologic repair processes of fracture healing cannot be simulated, and therefore, fatigue failure analysis was not possible in this study. Furthermore, our study focused only on bony structure and did not take surrounding musculature and soft tissues into account. Actual forces around the shoulder are a combination of compression, torsion, and shear. Our specimens were devoid of potential deforming or destabilizing forces generated by these muscles, which presumably affects the results. In conclusion, on the basis of our biomechanical test results, the parallel configuration of pin fixation has better torsional stability compared with the con-
J Shoulder Elbow Surg Volume 16, Number 2
vergent configuration when 1 cm is used as the pinto-pin distance. No significant difference was found between the 2 configurations regarding anti-shear stability. We suggest that parallel pin fixation should be applied whenever possible, and a specially designed parallel drill sleeve with a 1-cm pin-to-pin distance is recommended during clinical application. REFERENCES
1. Herscovici D Jr, Saunders DT, Johnson MP, Sanders R, DiPasquale T. Percutaneous fixation of proximal humeral fractures. Clin Orthop Relat Res 2000:97-104. 2. Jaberg H, Warner JJ, Jakob RP. Percutaneous stabilization of unstable fractures of the humerus. J Bone Joint Surg Am 1992; 74:508-15. 3. Jiang CY, Huang Q, Geng XS, Wang MY, Rong GW. Percutaneous pinning for the treatment of proximal humerus fractures [in Chinese]. Zhonghua Wai Ke Za Zhi 2004;42:725-9. 4. Kocialkowski A, Wallace WA. Closed percutaneous K-wire stabilization for displaced fractures of the surgical neck of the humerus. Injury 1990;21:209-12.
Jiang et al
239
5. Koval KJ, Blair B, Takei R, Kummer FJ, Zuckerman JD. Surgical neck fractures of the proximal humerus: a laboratory evaluation of ten fixation techniques. J Trauma 1996;40:778-83. 6. Naidu SH, Bixler B, Capo JT, Moulton MJ, Radin A. Percutaneous pinning of proximal humerus fractures: a biomechanical study. Orthopedics 1997;20:1073-6. 7. Oakey JW, Stover MD, Summers HD, Sartori M, Havey RM, Patwardhan AG. Does screw configuration affect subtrochanteric fracture after femoral neck fixation? Clin Orthop Relat Res 2006: 302-6. 8. Resch H, Hubner C, Schwaiger R. Minimally invasive reduction and osteosynthesis of articular fractures of the humeral head. Injury 2001;32(Suppl 1):25-32. 9. Selvan VT, Oakley MJ, Rangan A, Al-Lami MK. Optimum configuration of cannulated hip screws for the fixation of intracapsular hip fractures: a biomechanical study. Injury 2004;35:136-41. 10. Williams GR Jr, Wong KL. Two-part and three-part fractures: open reduction and internal fixation versus closed reduction and percutaneous pinning. Orthop Clin North Am 2000;31: 1-21. 11. Zingg U, Brunnschweiler D, Keller H, Metzger U. Percutaneous minimal osteosynthesis of fractures of the proximal humerus in elderly patients. Swiss Surg 2002;8:11-4.
We investigate the influence on fracture stability of different pin configurations during percutaneous pinning for the treatment of proximal humeral fractures. We performed a matched-paired study of 18 pairs of adult fresh-frozen humeri (36 humeri), which were divided into 4 groups. A 2-part surgical neck fracture model was used in all humeri, and 4 terminal threaded pins (2.5 mm in diameter) were used for fixation. Parallel-type pinning (box type) was carried out in 2 groups, and convergent-type pinning (fan-shaped type) was used in the other 2 groups. For each specimen, both anti-shear ultimate load and anti-torsion ultimate load were measured. There was no statistical difference between the parallel pin construct and convergent construct with regard to anti-shear resistance (P ⫽ .73). However, the parallel pin construct had a significant advantage over the convergent construct regarding anti-torsion resistance. The parallel pin construct has better torsional stability when 1 cm is used for the pin-to-pin distance. We suggest that parallel pin fixation should be applied whenever possible. (J Shoulder Elbow Surg 2007;16:235-239.)
I mproved results can be expected with surgical treat-
ment for displaced proximal humeral fractures compared with conservative treatment. Closed reduction with percutaneous pin fixation has been developed in the past decade. Because of this technique’s minimally invasive nature, further devitalization of the fracture fragments can be avoided. Patients can benefit from this technique with regard to fracture healing and early postoperative rehabilitation, but mechanical stability of pin fixation is still a concern with this type of proximal humeral fracture fixation.
From the School of Medicine, Peking University. Reprint requests: Chunyan Jiang, MD, PhD, Beijing Ji Shui Tan Hospital, 4th Medical Center, School of Medicine, Peking University, No. 31 Xinjiekoudongjie, 100035, Beijing, China (E-mail: [email protected]). Copyright © 2007 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2007/$32.00 doi:10.1016/j.jse.2006.05.011
The anatomy of the proximal humerus is similar to that of the proximal femur to a certain extent. The mechanical advantage of cannulated screws in a parallel configuration over cross-type configuration in femoral neck fracture fixation is well accepted by most orthopaedic surgeons. In reviewing related articles, however, no report on parallel pin fixation for proximal humeral fractures was found, nor has any biomechanical study been reported. We carried out a set of biomechanical studies on cadaveric models to investigate and compare mechanical stability between parallel fixation (box type) and convergent fixation (fan-shaped type). MATERIALS AND METHODS We divided 36 fresh-frozen adult humeri (18 pairs) into 2 groups, with 18 humeri (9 pairs) in each: anti-torsional stability test group (group T) and anti-shear stability test group (group S). Each group was also subdivided based on each side of a single individual into 2 subgroups, with 9 humeri in each, as follows: anti-torsional stability test group with parallel pin construct (group TB), anti-torsional stability test group with convergent pin construct (group TC), antishear stability test group with parallel pin construct (group SB), and anti-shear stability test group with convergent pin construct (group SC). The humeri were kept in saline solution at room temperature for 12 hours before testing and were soaked in saline solution throughout the process. Dual-energy bone density scanning had been used to elicit the level of osteoporosis of the cadaver. A 2-tailed paired t test was used to compare the bone density between the 2 subgroups (TB vs TC and SB vs SC). A 2-part surgical neck fracture was created in all specimens in a consistent way. An osteotomy was made by use of an oscillating saw with a 1-mm blade. Fixation via a parallel configuration (box type) with 4 AO threaded pins (2.5 mm in diameter) was used in groups TB and SB. Fixation via a convergent configuration (fan-shaped type) with 4 AO threaded pins was used in groups TC and SC. All cadaveric specimens were prepared by a single surgeon.
Determination of osteotomy line The baseline was a horizontal line made at 3 cm below the lateral cortex of the top of the greater tuberosity. The osteotomy line was angled 20° from the baseline starting at
235
236
Jiang et al
J Shoulder Elbow Surg March/April 2007
Figure 1 Two-part surgical neck fracture model and entry point of parallel pin construct. Gray line, Baseline; white line, osteotomy line. Figure 3 Anti-shear fixation jig.
Entry point of pins in convergent configuration (groups TC and SC)
Figure 2 Two-part surgical neck fracture model and entry point of convergent pin construct. Gray line, Baseline; white line, osteotomy line.
the junction between the baseline and medial margin of the humeral head (Figures 1 and 2).
Fixation of fracture models The pins were angled 45° to the humeral shaft in the coronal plane and 30° in the sagittal plane.
Entry point of pins in parallel configuration (groups TB and SB) The medial 2 entry points were made 0.5 cm proximal and distal to the points at which the baseline intersected with the lateral wall of the bicipital groove. The lateral 2 entry points were made 1 cm lateral to the medial 2 entry points (Figure 1). A specially designed parallel drilling sleeve with a 1-cm interval was used during pin fixation.
The most medial entry point was determined by the point at which the baseline intersected with the medial wall of the bicipital groove. The other 3 entry points were made consecutively with 1-cm intervals laterally on the same line, except the second and third entry points were made 0.5 cm proximal and distal to the baseline, respectively (Figure 2). Two sets of fixation jigs were designed for anti-shear stability and anti-torsion stability testing. For the anti-shear fixation jig (applied in groups TB and TC), the cadaveric humerus was clamped 30° away from the vertical line, and the compression tip was applied on the humeral head just lateral to the articular margin. The distance between the medial point of the osteotomy line and the proximal end of the clamp was 8 cm. A supporting metal plate was placed on the medial cortex of the humeral shaft (Figure 3). For the antitorsion fixation jig (applied in groups SB and SC), the humeral head was potted in bone cement proximal to the osteotomy line. A 4.0-mm Steinmann pin was drilled through the center of the humeral shaft 12 cm distal to the osteotomy line, and the compression tip (with a groove compatible with the Steinmann pin) was applied 6 cm away from the humeral shaft. By use of downward motion of the tip, torsion was produced at the fracture site (Figure 4). An Instron 5566 mechanical testing machine (Norwood, MA) was used. The force-applying probe was set in an isometric downward direction. The rate of the applied force was 1 cm/min. The machine was stopped when plastic deformation occurred, and the ultimate load was recorded. The ultimate load was defined as the point at which elastic deformation became plastic deformation. In a stress-displacement curve, ultimate load represents the point at which a straight line turns into a curved line. All data from the 4 subgroups were evaluated statistically by 2-tailed paired t tests to detect any significant difference between groups TB and TS and between groups SB and SC.
Jiang et al
J Shoulder Elbow Surg Volume 16, Number 2
237
Table II Dual-energy bone density scan values in anti-shear stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Figure 4 Anti-torsion fixation jig.
Group SB (g/cm2)
Group SC (g/cm2)
0.9623 0.7986 0.8141 0.9242 0.673 0.959 0.8465 1.039 0.7463 0.8625 0.117
0.8582 0.8138 0.9115 0.8581 0.662 1.051 0.8193 0.9878 0.7289 0.8545 0.120
P ⫽ .730.
Table I Dual-energy bone density scan values in anti-torsional stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group TB (g/cm2)
Group TC (g/cm2)
0.6456 0.8062 0.7405 1.001 0.8399 0.7991 0.8684 0.8795 1.079 0.8510 0.130
0.6398 0.955 0.7393 0.8845 0.8637 0.7902 0.857 1.026 0.7848 0.8378 0.115
P ⫽ .774.
RESULTS The bone density of all specimens was evaluated by dual-energy bone density scanning. The mean values for the 2 subgroups in the anti-torsion test group (TB and TC) were 0.851 ⫾ 0.130 g/cm2 and 0.838 ⫾ 0.115 g/cm2, respectively. No significant difference was found between groups TB and TC (P ⫽ .77) (Tables I and II). The mean values for the 2 subgroups in the anti-shear test group (SB and SC) were 0.863 ⫾ 0.117 g/cm2 and 0.855 ⫾ 0.120 g/cm2, respectively. No significant difference was found between groups SB and SC (P ⫽ .73) (Tables I and II). The ultimate load was 226.65 ⫾ 38.36 N in group TB and 186.97 ⫾ 61.25 N in group TC. A significant difference was found between groups TB and TC; the parallel configuration group had a statistically significant advantage over the convergent configuration group for anti-torsion stability (P ⫽ .04). The ultimate load was 637.84 ⫾ 174.13 N in group SB and 628.59 ⫾ 189.50 N in group SC. No
Table III Ultimate load in anti-torsional stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group TB (N)
Group TC (N)
286.12 201.63 240.5 281.26 196.65 185.62 241.15 220.34 186.61 226.653 38.363
250.15 172.74 152.81 239.07 107.99 247.27 238.52 184.16 90.04 186.972 61.248
P ⫽ .043.
significant difference was found between groups SB and SC (P ⫽ .73), suggesting that there is no statistically significant difference between the parallel configuration group and the convergent configuration group for anti-shear stability (Tables III and IV). DISCUSSION Good results have been reported with threaded pin fixation for the treatment of proximal fracture,1,3,8,11 but the mechanical stability of pin fixation is still a concern in clinical practice. Pin back-out and loss of fixation are not uncommon and, thus, influence the proper healing process of a fracture.3,11 Our study is focused on how to improve fixation stability by modifying pin configuration. It is well accepted that a parallel configuration in cannulated screw fixation is important in treating femoral neck fractures, not only because the parallel screw construct is an important prerequisite for sliding compression of the fracture line but also because of its
238
Jiang et al
J Shoulder Elbow Surg March/April 2007
Table IV Ultimate load in anti-shear stability test groups
Cadaveric pair 1 2 3 4 5 6 7 8 9 Mean SD
Group SB (N)
Group SC (N)
933.55 370.05 607.11 476.72 584.13 855.39 698.12 587.12 628.34 637.837 174.130
1027.46 435.49 566.57 465.85 684.56 804.83 634.28 496.64 541.62 628.589 189.049
P ⫽ .725.
better mechanical stability compared with the convergent screw construct.7,9 The anatomy of the proximal humerus is similar to that of the proximal femur to a certain extent. Theoretically, a parallel pin construct should have a better mechanical advantage than the convergent pin construct. Most authors suggest that a convergent pin configuration will achieve the best stability in proximal humeral fracture fixation.1,2,4,5,10 In our review of the relevant literature, no clinical biomechanical report or study comparing parallel and convergent pin configuration in the proximal humeral fracture fixation was found, and there is only one article referring to the biomechanical characteristics of the parallel pin construct.5 To prove this theory, we designed a cadaveric fracture model to investigate the influence of pin construct on fixation stability. We compared 2 types of configuration: parallel and convergent. Both configurations included 4 pins in retrograde fixation. Some authors suggest that, with an additional 1 or 2 antegrade pins through the greater tuberosity to the medial cortex, the whole construct will have better stability.1,2 Koval et al5 showed that the best mechanical stability is achieved in osteoporotic bone with 3 retrograde pins in a convergent construct combined with 1 antegrade pin. Naidu et al6 also suggested that an additional 2 antegrade pins stabilizing the greater tuberosity to the medial cortex will improve both axial and rotational stability. Although an antegrade pin could improve fixation stability, the end of the pin can cause impingement under the acromion during shoulder abduction and make early postoperative rehabilitation impossible. In our experience, satisfactory stability can be achieved with 3 or 4 retrograde pins with appropriate entry points and positioning without hampering postoperative rehabilitation. Therefore, the cross-type pin construct with antegrade pin fixation was not included in this study.
We compared both sides of a single individual to exclude the influence of size or morphologic differences between specimens. We also matched all specimens by bone density scan to ensure that there was no significant difference between subgroups, thus excluding the influence of osteoporosis on fixation stability. The results proved our hypothesis: although no significant difference was found with regard to anti-shear stability, the parallel configuration does have better torsional stability than the convergent configuration. The clinical application of a parallel pin construct for proximal humeral fracture fixation is not as easy as it appears in a cadaveric model. Because of its minimally invasive nature, percutaneous pinning will definitely provide limited access to the fracture site. We cannot expect to find a broad and flat lateral humeral cortex like the proximal femur. The mass of the deltoid muscle will further interfere with appropriate pin orientation, resulting in possible pin slippage on the lateral cortex of the humerus. Moreover, parallel pin fixation may be impossible in small individuals. The distance between pins plays an important role in the mechanical properties of the parallel construct. Our finding that the parallel configuration provides better torsional stability than the convergent configuration is based on a very important premise: the distance between pins is 1 cm. We do not know the limit at which the parallel configuration still has better torsional stability over the convergent configuration, but we do know that with a decrease in this distance, the parallel construct will eventually lose its mechanical advantage. Our suggestion is to use parallel pin fixation whenever possible to improve fracture stability. A specially designed parallel drill sleeve with a 1-cm pin-to-pin distance and a serrated tip enhancing cortical purchase is recommended for the clinical application of parallel pin fixation. If this is not possible in a small individual, the convergent configuration is still a good choice for percutaneous pinning. The limitation of this study is that all specimens were tested to failure in a single loading fashion. The mechanism of fixation failure is very complicated and multifactorial. Fatigue failure may also play an important role. Our study is based on a cadaveric model; in vivo biologic repair processes of fracture healing cannot be simulated, and therefore, fatigue failure analysis was not possible in this study. Furthermore, our study focused only on bony structure and did not take surrounding musculature and soft tissues into account. Actual forces around the shoulder are a combination of compression, torsion, and shear. Our specimens were devoid of potential deforming or destabilizing forces generated by these muscles, which presumably affects the results. In conclusion, on the basis of our biomechanical test results, the parallel configuration of pin fixation has better torsional stability compared with the con-
J Shoulder Elbow Surg Volume 16, Number 2
vergent configuration when 1 cm is used as the pinto-pin distance. No significant difference was found between the 2 configurations regarding anti-shear stability. We suggest that parallel pin fixation should be applied whenever possible, and a specially designed parallel drill sleeve with a 1-cm pin-to-pin distance is recommended during clinical application. REFERENCES
1. Herscovici D Jr, Saunders DT, Johnson MP, Sanders R, DiPasquale T. Percutaneous fixation of proximal humeral fractures. Clin Orthop Relat Res 2000:97-104. 2. Jaberg H, Warner JJ, Jakob RP. Percutaneous stabilization of unstable fractures of the humerus. J Bone Joint Surg Am 1992; 74:508-15. 3. Jiang CY, Huang Q, Geng XS, Wang MY, Rong GW. Percutaneous pinning for the treatment of proximal humerus fractures [in Chinese]. Zhonghua Wai Ke Za Zhi 2004;42:725-9. 4. Kocialkowski A, Wallace WA. Closed percutaneous K-wire stabilization for displaced fractures of the surgical neck of the humerus. Injury 1990;21:209-12.
Jiang et al
239
5. Koval KJ, Blair B, Takei R, Kummer FJ, Zuckerman JD. Surgical neck fractures of the proximal humerus: a laboratory evaluation of ten fixation techniques. J Trauma 1996;40:778-83. 6. Naidu SH, Bixler B, Capo JT, Moulton MJ, Radin A. Percutaneous pinning of proximal humerus fractures: a biomechanical study. Orthopedics 1997;20:1073-6. 7. Oakey JW, Stover MD, Summers HD, Sartori M, Havey RM, Patwardhan AG. Does screw configuration affect subtrochanteric fracture after femoral neck fixation? Clin Orthop Relat Res 2006: 302-6. 8. Resch H, Hubner C, Schwaiger R. Minimally invasive reduction and osteosynthesis of articular fractures of the humeral head. Injury 2001;32(Suppl 1):25-32. 9. Selvan VT, Oakley MJ, Rangan A, Al-Lami MK. Optimum configuration of cannulated hip screws for the fixation of intracapsular hip fractures: a biomechanical study. Injury 2004;35:136-41. 10. Williams GR Jr, Wong KL. Two-part and three-part fractures: open reduction and internal fixation versus closed reduction and percutaneous pinning. Orthop Clin North Am 2000;31: 1-21. 11. Zingg U, Brunnschweiler D, Keller H, Metzger U. Percutaneous minimal osteosynthesis of fractures of the proximal humerus in elderly patients. Swiss Surg 2002;8:11-4.