Central Screw Placement in Percutaneous Screw Scaphoid Fixation: A Cadaveric Comparison of Proximal and Distal Techniques

Transcription

1 Central Screw Placement in Percutaneous Screw Scaphoid Fixation: A Cadaveric Comparison of Proximal and Distal Techniques Keith W. Chan, BS, Timothy R. McAdams, MD, Palo Alto, CA Purpose: Percutaneous screw fixation of acute minimally displaced scaphoid fractures is an attractive treatment alternative compared with cast immobilization and can be performed with either a distal/volar or proximal/dorsal approach. Central screw placement within the scaphoid appears to be an important factor for successful fixation. The purpose of this cadaveric study is to investigate whether the proximal or distal approach for percutaneous screw scaphoid fixation allows for more central placement of the screw. Methods: Twelve fresh frozen cadaveric upper limbs were studied, with 6 specimens assigned to scaphoid screw placement with a proximal approach and 6 matched specimens assigned to scaphoid screw placement with a distal approach. After screw placement, the scaphoid was sectioned evenly into quarters along the longitudinal proximal-distal axis. For each section, the distance from the center of the screw hole to the edges of the dorsal/volar/radial/ulnar axes was measured, and the means of the 2 groups were compared with a Hotelling s T 2 test to determine statistically significant central screw placement. Results: A statistically significant difference was found between the mean location of the distal fixation group and the center of the scaphoid in the midwaist and distal pole of the scaphoid (p.007 and.012, respectively) and between the mean location of the proximal and distal fixation groups in the distal pole of the scaphoid (p.045). Conclusions: We find that the proximal/dorsal approach to the percutaneous screw fixation of scaphoid waist fractures allows for a more central placement in the distal pole, but there is no significant difference when it is used in the proximal or waist region. It remains unclear whether the more central screw placement afforded by the proximal approach might translate into an improved clinical outcome. (J Hand Surg 2004;29A: Copyright 2004 by the American Society for Surgery of the Hand.) Key words: Central screw placement, dorsal, percutaneous fixation, scaphoid, volar. From the Department of Hand and Upper Limb Surgery, Palo Alto, CA. Received for publication March 17, 2003; accepted in revised form September 8, 2003 No benefits in any form have been received or will be received by a commercial party related directly or indirectly to the subject of this article. Reprint requests: Timothy R. McAdams, MD, Department of Hand and Upper Limb Surgery, 900 Welch Rd, Suite #15, Palo Alto, CA, Copyright 2004 by the American Society for Surgery of the Hand /04/29A $30.00/0 doi: /j.jhsa Percutaneous screw fixation of acute, minimally displaced scaphoid fractures is an attractive treatment alternative to cast immobilization, especially in young active patients and manual laborers. 1 This minimally invasive procedure uses a headless compression screw placed over a guidewire under fluoroscopic visualization. 2 Numerous studies show that percutaneous fixation of the scaphoid results in a faster recovery time and decreased duration of cast immobilization compared with nonsurgical management, allowing patients to regain use of the injured 74 The Journal of Hand Surgery

2 Chan and McAdams / Percutaneous Scaphoid Screw Placement 75 hand within a much shorter time course. 1,3,4 Additional advantages of percutaneous fixation are its simplicity and decreased incidence of complications associated with damage to important structures compared with open reduction and internal fixation. 5 Percutaneous screw fixation can be performed with either a distal/volar or proximal/dorsal approach. Haddad and Goddard 2 first published the technique of acute percutaneous screw scaphoid fixation using a volar approach with a headless compression screw (Acutrak; Acumed, Hillsboro, OR). The wrist is placed in ulnar deviation and extension while the guidewire is introduced into the scaphotrapezial joint from the volar surface, and the screw is directed distally to proximally. Slade et al 6 subsequently described how to accomplish percutaneous screw fixation with a proximal approach. The wrist is held in pronation and flexion, allowing more accurate guidewire positioning into the proximal pole of the scaphoid. Currently, both proximal and distal approaches are used for midwaist scaphoid fractures, with the choice largely dependent on the preference of the surgeon. Recent biomechanical studies have shown that screws placed centrally through the scaphoid provide stronger, more secure fixation and lower risk of screw migration or fracture at the screw-bone interface and, thus, have stressed the importance of careful central screw placement in the scaphoid. 7 The purpose of this cadaveric study is to investigate whether the proximal or distal approach for percutaneous screw scaphoid fixation allows for more consistently central placement of the screw throughout the scaphoid. Materials and Methods Twelve fresh-frozen cadaveric upper limbs disarticulated at the shoulder were studied. No specimens had any evidence of previous carpal injury on fluoroscopic examination. Six specimens had percutaneous scaphoid screw placement through a distal (retrograde) approach and 6 matched specimens via proximal approach, using a standard compression screw set (Acutrak). For the distal approach, the wrist was placed in ulnar deviation and extension and a guidewire was inserted through the distal pole of the scaphoid and advanced to the proximal cortex under fluoroscopic guidance. After screw length measurement and drilling, the screw was inserted and the guidepin removed. For the proximal approach, the wrist was placed in pronation and flexion and, under fluoroscopy, the guidewire was introduced into the Figure 1. The scaphoid was sectioned transversely along the proximal-distal axis in 3 planes: (I) distal, (II) mid-waist, and (III) proximal to allow measurement of screw hole displacement from the center of the scaphoid in cross-section. D, distal; P, proximal; R, radial; U, ulnar. proximal pole and advanced through the distal pole and trapezium and out through the volar skin at the base of the thumb. The guidewire was then withdrawn distally until the proximal tip of the pin was even with the proximal pole cortex. This allowed fluoroscopic visualization of pin placement without bending the guidewire across the radiocarpal joint. Once guidepin placement was deemed satisfactory, the pin was advanced proximally with the wrist flexed and pronated. After screw length measurement and drilling, the screw was inserted and the guidepin was withdrawn. After final screw placement, the scaphoid was dissected free and removed from the wrist. The screw was removed from the bone, leaving a clear screw hole. With a bandsaw, the scaphoid specimens were sectioned evenly into quarters along the longitudinal proximal-distal axis, resulting in 3 cross-sectional planes within the scaphoid, representing anatomically the proximal, middle, and distal sections through the scaphoid (Fig. 1). For each section, the distance from the center of the screw hole to the edges of the dorsal/volar/radial/ulnar axis was measured (Fig. 2). In addition, the dorsal-volar and the radial-ulnar diameters were recorded. Measure-

3 76 The Journal of Hand Surgery / Vol. 29A No. 1 January 2004 Figure 2. Schematic of scaphoid cross-section. Measurements were recorded as (a) displacement from the center of the screw hole to the dorsal edge (D), (b) displacement from the center of the screw hole to the volar edge (V), (c) displacement from the center of the screw hole to the ulnar edge (U), and (d) displacement from the center of the screw hole to the radial edge (R). ments were recorded to the nearest 0.1 mm with a digital caliper (Multitoyo America Corp, Aurora, IL). Measurement Parameters To account for the variability in size of scaphoid bones, the data were normalized using the diameter of the scaphoid at the location of the cut: the distance from the center of the screw hole to the edge of the nearest cortex was divided by the recorded diameter to obtain a percentage displacement from the edge. The percentage displacement from the center of the scaphoid, measured as one half the diameter for the given plane, was obtained by subtracting the calculated percentage from 50%. The means and standard error for the measurements of the proximal and distal fixation specimens were calculated. To compare the location of the screw in the given plane, we transformed the data to obtain 2 coordinates, 1 in the dorsal-volar axis and 1 in the radialulnar axis, ranging from 1 to 1. For the coordinate in the dorsal-volar axis, the difference between the dorsal and volar measurements was divided by the sum of the dorsal and volar measurements. For the coordinate in the radial-ulnar axis, the difference between the radial and ulnar measurements was divided by the sum of the radial and ulnar measurements. After this conversion, the center position could be denoted by the origin (0,0). The mean coordinates and sample variance for the measurements of the volar and dorsal fixation specimens were calculated. We tested the statistical significance of our data in 2 ways at each anatomic scaphoid section using a Hotelling s T 2 test. The Hotelling s T 2 test is a modification of a chi-square analysis, useful in the study of relatively small sample sizes. A 1-sample Hotelling s T 2 test compared the mean coordinates from each of the sample groups against the origin (0,0) to determine whether that fixation group was statistically different from the central position; a p value less than.05 was considered to be statistically significant. A 2-sample Hotelling s T 2 test compared the 2 sample groups against each other to determine whether the 2 fixation groups were statistically different; a p value less than.05 was considered a statistically significant difference between the 2 groups. Results For the anatomic proximal section, the distal approach screw deviated by 5.0% 5.6% in the dorsal direction along the dorsal-volar axis and by 1.7% 5.6% in the ulnar direction along the radial-ulnar axis from the center. For the same section, the proximal approach screw deviated by 6.9% 4.3% in the volar direction along the dorsal-volar axis and by 4.2% 5.0% in the radial direction along the radialulnar axis from the center. For the anatomic middle section, the distal approach screw deviated by 12.4% 2.3% in the volar direction along the dorsal-volar axis and by 3.8% 2.6% in the ulnar direction along the radial-ulnar axis, and the proximal approach screw deviated by 7.6% 5.3% in the volar direction along the dorsalvolar axis and by 8.4% 5.0% in the ulnar direction along the radial-ulnar axis from the center. For the anatomic distal section, the distal approach screw deviated by 19.6% 3.3% in the volar direction along the dorsal-volar axis and by 4.7% 3.2% in the radial direction along the radial-ulnar axis, and the proximal approach screw deviated by 4.6% 4.0% in the volar direction along the dorsal-volar axis and by 15.4% 7.4% in the radial direction along the radial-ulnar axis from the center (Fig. 3). No statistical difference from the center position was seen at the proximal, middle, or distal regions of

4 Chan and McAdams / Percutaneous Scaphoid Screw Placement 77 Figure 3. Screw location at the proximal, midwaist, and distal scaphoid sections. The data set was schematically plotted with the origin representing the center of the scaphoid. The volar specimen data are represented by the (x) symbol and the mean of the volar specimens is represented by a square. The dorsal specimen data are represented by the ( ) symbol and the mean of the dorsal specimens is represented by a triangle. (A) Proximal section: No statistical difference was found between the means of the volar and dorsal specimens and the center. (B) Midwaist section: The means of both volar and dorsal specimens deviated volarly and ulnarly but were not statistically different from the center. (C) Distal section: Although both mean screw placements of the volar and dorsal specimens deviated in the volar and radial direction, the large volar displacement of the volar approach specimens resulted in statistically significant noncentral screw placement. the scaphoid in the proximal approach group (p , respectively). No statistical difference from the center position was seen in the proximal section of the scaphoid in the distal approach group (p.768), but in the middle and distal sections of the scaphoid, the means of the distal fixation group were significantly different from the center (p.007,.012, respectively). Statistical analysis comparing the 2 fixation groups directly showed no statistical difference in the means of the proximal and distal groups at the proximal and middle sections of the scaphoid (p.09,.435, respectively) but did show a significant difference at the distal section of the scaphoid (p.045). Discussion To our knowledge, this is the first anatomic study to directly compare the central screw placement between the proximal and distal approaches of percutaneous scaphoid fixation. In this study, we find that the proximal approach for percutaneous screw scaphoid fixation results in a statistically significantly more central screw placement in the distal scaphoid pole compared with the distal approach. No statistically significant difference between central screw placement in the 2 approach groups was found in the proximal pole or midwaist regions. Anatomically, the proximal approach allows for a more accurate placement of the guidewire because the proximal pole of the scaphoid is visualized as a circle on fluoroscopy, increasing the ease of aiming the guidewire into the center of the circle. 6 In addition, the proximal approach is less likely to damage the superficial branch of the radial artery. 5 There is, however, a theoretical risk of damaging important structures when the guidewire is advanced blindly through the scaphoid into the trapezium and out the radial base of the thumb, even though Slade et al declare the region to be free from tendons or neurovascular structures. 8 The procedure for the distal approach is technically less demanding and allows greater ease of guidewire insertion and placement because the guidewire does not cross the radiocarpal joint and so it is at less risk of being bent during fluoroscopic evaluation. The distal approach, however, requires the guidewire to be inserted at the level of the scaphotrapezial joint, which can lead to volar placement of the screw in the distal pole because the volar surface of the trapezium limits the ability to enter the scaphoid in a central position. Although an improved angle of insertion can be attained by removing several millimeters of the volar trapezium, this requires a small incision and is therefore no longer truly a percutaneous approach. In addition, the distal approach is not as effective as the proximal approach in securely fixing and compressing the more proximal fractures of the scaphoid. 9 Fracture of the carpal scaphoid is common and accounts for 70% of carpal fractures, with an annual estimated incidence of 38 fractures per population of 100, Traditionally, management for an acute nondisplaced scaphoid fracture involves conservative treatment with immobilization in an above- or belowelbow thumb spica cast for an average of 10 to 12

5 78 The Journal of Hand Surgery / Vol. 29A No. 1 January 2004 weeks or longer if the scaphoid fails to show evidence of clinical and radiographic healing. 4 These fractures typically occur in young adults and prolonged immobilization and activity restriction can be particularly problematic in this patient population. The morbidity of such immobilization is even greater for those who require full use of their hands in their occupation, such as physicians, athletes, and manual laborers. Frequent radiographic re-evaluation during recovery is needed to monitor for nonunion or displacement of the scaphoid fragments. 5 Scaphoid nonunion or malunion can result in pain and loss of function, motion, and strength for the affected hand. 6,7 Percutaneous screw fixation of acute, nondisplaced or minimally displaced scaphoid waist fractures is an attractive treatment option in this often young, active patient population. For athletes or workers who require manual dexterity, regaining use of the affected hand as early as possible is of utmost importance. 10 Significant morbidity is seen with the traditional cast immobilization treatment for scaphoid waist fractures, including joint stiffness, muscle atrophy, and loss of dexterity secondary to the lengthy immobilization. 11 Furthermore, many researchers recommend acute above-elbow immobilization for these fractures, which may risk elbow stiffness. There is evidence that central placement of the percutaneous screw is important for successful percutaneous screw fixation of scaphoid fractures. Trumble et al 15 showed decreased time to fracture union in open scaphoid repair with more centrally located screws. A biomechanical study showed central screw placement to be stronger, with greater stiffness and increased load to failure than eccentrically placed screws, thus reducing the likelihood of screw failure. 7 These studies, however, did not evaluate the conical headless screw (Acutrak) used in our study. It is possible that the increased compression afforded by this screw, as well as percutaneous technique, may decrease the importance of central screw placement in terms of scaphoid fracture healing rates seen clinically. Trumble et al hypothesized that central placement of the screw is most important in the proximal pole of the scaphoid, because they inserted screws in the distal pole of the scaphoid toward the proximal pole, similar to the percutaneous distal technique. 7 We find in our study that both proximal and distal approaches resulted in central placement of the screw in the proximal pole; however, only the proximal approach allowed for central placement of the screw distally. No studies have been published examining the importance of central screw placement in the distal pole of the scaphoid. Nevertheless, the proximal approach allows the screw to be closer to center throughout the scaphoid compared with the distal approach, which tends to deviate in the volar direction in the distal pole. A scaphoid fracture was not simulated in this study because of the difficulty in reproducing the osteotomy and was not deemed necessary to assess screw placement. We therefore cannot comment on whether one approach provides better compression of the scaphoid fragments. A limitation of this study may be the sectioning method. Only 1 section from each anatomic area (proximal, middle, and distal) was analyzed. It is possible that a continuum of measurements throughout the scaphoid may provide a more precise result. In addition, the cross-sections were fitted into a circle for purposes of analysis based on a coordinate system. This may lead to approximations rather than precise data points when analyzing a biological structure that is not circular. We believe, however, that this does not change our overall findings in terms of where the scaphoid screw deviated from a central position. Based on this study, the proximal approach to the percutaneous screw fixation of scaphoid waist fractures allows for a more central placement in the distal pole but no significant difference in the proximal or waist region. It remains unclear whether the more central screw placement afforded by the proximal approach might translate into improved clinical outcome. A prospective clinical trial is underway at our institution to compare the potential benefits and complications of these 2 techniques. We would like to thank Dr Jon Kosek (Palo Alto VA Hospital, Department of Pathology) for help with cutting the scaphoid specimens. We also thank John Dolph (Stanford University, Division of Anatomy) for helping to procure cadaver specimens. We would like to thank Eric Stone (Stanford University, Department of Statistics) for his statistical consultation and analysis. We would like to thank the Stanford Medical Scholars Program for their funding and support for this project. Finally, we thank Acumed for their generous loan of equipment used in this study. References 1. Inoue G, Shionoya K. Herbert screw fixation by limited access for acute fractures of the scaphoid. J Bone Joint Surg Br 1997;79: Haddad FS, Goddard NJ. Acute percutaneous scaphoid fixation: A pilot study. J Bone Joint Surg Br 1998;80B:95 99.

6 Chan and McAdams / Percutaneous Scaphoid Screw Placement Bond CD, Shin AY, McBride MT, Dao KD. Percutaneous screw fixation or cast immobilization for nondisplaced scaphoid fractures. J Bone Joint Surg Am 2001;83A: Taras JS, Sweet S, Shum W, Weiss LE, Bartolozzi A. Percutaneous and arthroscopic screw fixation of scaphoid fractures in the athlete. Hand Clin 1999;15: Kamineni S, Lavy CB. Percutaneous fixation of scaphoid fractures. An anatomical study. J Hand Surg 1999;24B: Slade JF, Grauer JN, Mahoney JD. Arthroscopic reduction and percutaneous fixation of scaphoid fractures with a novel dorsal technique. Orthop Clin North Am 2001;32: McCallister WV, Knight J, Kaliappan R, Trumble TE. Central placement of the screw in simulated fractures of the scaphoid waist: a biomechanical study. J Bone Joint Surg Am 2003;85: Slade JF, Gutow AP, Geissler WB. Percutaneous internal fixation of scaphoid fractures via an arthroscopically assisted dorsal approach. J Bone Joint Surg Am 2002;84(suppl 2): Faran KJ, Ichioka N, Trzeciak MA, Han S, Medige J, Moy OJ. Effect of bone quality on the forces generated by compression screws. J Biomech 1999;32: Cooney WP, Dobyns JH, Linscheid RL. Fractures of the scaphoid: a rational approach to management. Clin Orthop 1980;149: Skirven T, Trope J. Complications of immobilization. Hand Clin 1994;10: Gelberman RH, Wolock BS, Siegel DB. Fractures and nonunion of the carpal scaphoid. J Bone Joint Surg Am 1989; 71: Gellman H, Caputo RJ, Carter V, Aboulafia A, McKay M. Comparison of short and long thumb-spica casts for nondisplaced fractures of the carpal scaphoid. J Bone Joint Surg Am 1989;71: Verdan C. Fractures of the scaphoid. Surg Clin North Am 1960;40: Trumble TE, Clarke T, Kreder HJ. Non-union of the scaphoid. Treatment with cannulated screws compared with treatment with Herbert screws. J Bone Joint Surg Am 1996;78: