External fixation of an unstable fracture of the distal end of the radius provides a simple way to maintain axial alignment through ligamentotaxis and to neutralize external forces during healing. However, external fixation is associated with complications, including infection, loosening of the pins, loss of reduction, and permanent stiffness15,16,21,24. There have been many attempts to change the design of external fixators of the distal end of the radius in order to facilitate their use and to reduce the rate of complications7,30. Early motion of the wrist has been advocated to decrease the stiffness associated with prolonged immobilization7,21. Dynamic external fixators have been used in clinical trials in an attempt to decrease stiffness, but pin-track infections and settling of the fracture have still occurred7,30.
Kirschner wires have been used to augment fixation in order to help to prevent settling of the fracture and loss of reduction23,28. With use of a model of an unstable distal radial fracture, we demonstrated that supplemental Kirschner-wire transfixion or outrigger4 fixation significantly increases the biomechanical stability of external fixation (p < 0.05)33. If supplementary fixation of the fracture provides sufficient stability, early removal of all or part of the external fixator may be possible in order to allow motion of the wrist during healing. To the best of our knowledge, no biomechanical studies have been performed to investigate early removal of an external fixator in a distal radial fracture model with simulated partial healing.
The purpose of the present study was to observe the load-displacement behavior of an in vitro partially healed distal radial fracture model with supplemental Kirschner-wire transfixion or outrigger fixation. We hypothesized that a simulated partially healed fracture supplemented with Kirschner-wire fixation would be sufficiently stable to allow early loading and motion of the wrist.
Materials and methods
Preparation of the Specimens
Sixteen fresh-frozen cadaveric hand-wrist-forearm specimens were mounted in mid-rotation in Bondo resin pots (Dynatron/Bondo, Atlanta, Georgia). Except for the tendons of the five wrist motors (the extensor carpi radialis longus, extensor carpi radialis brevis, extensor carpi ulnaris, flexor carpi ulnaris, and flexor carpi radialis), the pronator quadratus, the wrist capsule, the ligaments of the wrist, the dorsal retinaculum, and the interosseous membrane, which were left intact, all skin and soft tissue were dissected off of the specimens. The tendons of the five wrist motors were isolated and were secured with Bunnell sutures, with use of number-1 braided polyester, for the application of load. A single syndesmotic screw was inserted eleven centimeters proximal to the tip of the ulnar styloid process with the forearm in neutral rotation. An external fixator (Orthoframe Mayo fixator; Orthologic, Phoenix, Arizona) was applied, with a standard mid-axial method, to the radial diaphysis and the index metacarpal. Each pin cluster consisted of two 2.0-millimeter threaded pins. An unstable metaphyseal fracture was simulated by removal of a one-centimeter-wide dorsal wedge of corticocancellous bone centered two centimeters from the articular margin. The volar aspect of the osteotomy site was divided, but cortical contact remained.
The partially healed model was simulated by injection of a butyl-rubber caulking compound (Red Devil Caulk; Red Devil, Union, New Jersey) into the site of the osteotomy. After the external fixator had been applied and the osteotomy site had been prepared, the ends of the bone were dried with ethanol. Caulk was injected into the osteotomy site with a caulk gun and was pressurized to integrate the material into the cancellous surface. The caulk was allowed to dry for forty-eight hours before testing. The indentation stiffness of the butyl-rubber compound was determined with a method described by Markel et al. for the testing of callus and tissue from the gap of a canine tibial fracture model22. Fully cured butyl-rubber compound was formed into 2.0-millimeter-thick sections and was indented with a 1.6-millimeter-diameter ball at a rate of 2.75 millimeters per minute in a specially designed uniaxial testing machine to a maximum depth of 0.37 millimeter. The stepper-motor-controlled machine has a resolution of 0.01 millimeter.
Each potted specimen was mounted vertically in a loading jig to facilitate tendon-loading. Three monitoring flags were used for motion analysis. Each flag contained three nonlinear infrared emitting diodes designed for the Optotrak motion analysis system (Northern Digital, Waterloo, Canada). One flag was mounted 0.5 centimeter proximal to the articular margin of the distal end of the radius (within the osteotomy fragment). A second flag was mounted in the proximal end of the radial diaphysis. A third flag was mounted in the metacarpal of the long finger. Anteroposterior radiographs were made of each specimen, and the proximal and distal edges of the osteotomy site were digitized for later motion analysis.
Group I (Controls)
Three control groups were tested to determine the effect of fixation with Kirschner wires alone, outrigger fixation alone, and simulated partial healing alone on relative motion between the bone fragments during active motion of the wrist. The distal pin cluster of the external fixator was removed, and the three different fixation constructs were tested (Fig. 1).
The first construct (Group IA) consisted of two 0.062-inch (0.16-centimeter) diameter Kirschner wires (transfixion wires) inserted into the distal fragment and across the site of the osteotomy. The wires crossed the osteotomy site at an angle of approximately 45 degrees in order to engage the far cortex of the proximal fragment. One of the wires entered the tip of the radial styloid process and crossed the site in the coronal plane (the styloid transfixion wire), and the other entered the dorsal lip of the radius and crossed the site in the sagittal plane (the dorsal transfixion wire). Four specimens were tested in this manner.
The second construct (Group IB) consisted of two 0.062-inch (0.16-centimeter) diameter Kirschner wires attached through an outrigger device to the remaining proximal arm of the fixator. These wires were placed across both cortices of the distal fragment but did not cross the site of the osteotomy and thus were not transfixion Kirschner wires. One wire entered the distal fragment from the radial styloid process (the styloid outrigger wire), and the other entered the fragment from a straight dorsal direction (the dorsal outrigger wire). The same four specimens that were tested in Group IA were tested in Group IB; the sequence of testing of the transfixion and outrigger wires was alternated so as not to introduce artifact related to repeated testing.
The third construct (Group IC) consisted of simulated partial healing with caulk and without augmentation with Kirschner wires. Four specimens were tested in this manner.
Group II (Standard Augmented External Fixation)
In Group II, eight specimens were tested, before and after the osteotomy, with standard external fixation and six Kirschner-wire constructs.
Each specimen was tested before the osteotomy was performed but after the external fixator was applied, to control for any imprecision of the testing apparatus and the three-dimensional kinematic measurements. After the osteotomy, six Kirschner-wire constructs were tested: a single styloid transfixion wire, a single dorsal transfixion wire, styloid and dorsal transfixion wires, a single styloid outrigger wire, a single dorsal outrigger wire, and styloid and dorsal outrigger wires (Fig. 2). The order of testing of the configurations was randomized for each specimen to control for any variability due to insertion and removal of the wires. No caulk was used in this group.
Group III (Experimental Group)
The same eight specimens that were tested in Group II were tested in Group III. However, in Group III, caulk was injected into the osteotomy site and was allowed to cure for forty-eight hours before testing (Fig. 3). The six Kirschner-wire constructs that were tested in Group II were tested after removal of the distal pin cluster to allow an unrestricted range of motion of the wrist. The constructs were tested in randomized order to minimize the effect of removal and replacement of the wires.
Testing of the Specimens
The specimen and the loading jig were fixed to a pneumatic loading device. Preload was applied to the wrist motors to approximate physiological muscle tone1,19,20,32. A load of 9.8 newtons was applied to the flexor carpi radialis, flexor carpi ulnaris, and extensor carpi ulnaris tendons, and 9.8 newtons was distributed equally between the extensor carpi radialis brevis and longus tendons, for a total of 39.2 newtons across the wrist. Incremental load increases of 19.6 newtons were applied in either extension or flexion to a maximum of 98.0 newtons across the wrist. In Groups I and III, in which the wrist was free to move, the flexion or extension force caused the hand to flex or extend within the constraints of the wrist capsule and ligaments. Two loading cycles were performed before actual testing to control for viscoelastic deformation of the ligaments and soft tissues of the wrist2 Motion data were recorded with use of the flags and infrared markers during a third loading cycle for each experimental configuration.
The specimens were kept moist with isotonic saline solution during testing. In addition, the wrists were loosely wrapped with saline solution-soaked gauze during storage to minimize evaporative losses.
Analysis of the Data
The rotational and translational motions of the distal fragment relative to the proximal fragment were calculated by translating the global coordinates of the Optotrak system into a local reference system with use of the digitized radiographs of each specimen. The local coordinate system referenced the distal edge of the osteotomy site with respect to the proximal edge. The range of motion of the distal fragment was measured in all three planes. The results were compared among the three groups with use of a derived analysis-of-variance model with an F of 218.1. In addition, the stability of individual Kirschner-wire constructs was compared within each group with a one-tailed paired Student t test. A p value of 0.05 or less was considered significant. All calculations were performed with Excel software (Microsoft, Redmond, Washington).
In Group I (controls), the mean range of motion (and standard deviation) of the distal fragment was 19.4 ± 7.2 degrees in the sagittal plane (flexion-extension) and 12.0 ± 2.9 degrees in the coronal plane (radioulnar deviation), which represents a highly unstable fracture. The transfixion Kirschner wires (Group IA) allowed a mean range of motion of 15.1 ± 2.7 degrees in the sagittal plane and 10.9 ± 1.3 degrees in the coronal plane. The Kirschner wires attached to an outrigger (Group IB) allowed a mean range of range of motion of 14.7 ± 2.3 degrees in the sagittal plane and 9.6 ± 1.4 degrees in the coronal plane. Finally, the use of caulk to simulate a partially healed fracture (Group IC) allowed a mean range of motion of 28.4 ± 4.0 degrees in the sagittal plane and 15.5 ± 1.8 degrees in the coronal plane (Table I). The indentation stiffness of the butyl-rubber caulking material was 0.6 ± 0.1 newton per millimeter.
In Group II (standard augmented external fixation), there was minimum motion of the distal fragment in the sagittal plane (mean, 0.04 degree) or the coronal plane (mean, 0.02 degree) before the osteotomy. After the osteotomy and application of the Kirschner-wire constructs, the mean overall range of motion of the distal fragment was 3.2 ± 1.2 degrees in the sagittal plane and 1.6 ± 0.4 degrees in the coronal plane (Table I). The post hoc Scheffé test revealed that the mean range of motion in the sagittal and coronal planes in Group II was significantly different (p < 0.05) than that for all constructs in Group I. Of note, no significant differences were detected, with the numbers available, between individual Kirschner-wire constructs in Group II (p > 0.05).
In Group III (the experimental group), the mean range of motion of the distal fragment was 3.8 ± 0.7 degrees in the sagittal plane and 1.9 ± 0.4 degrees in the coronal plane. The post hoc Scheffé test revealed that the mean range of motion in the sagittal and coronal planes in Group III was significantly less than that for all constructs in Group I (p < 0.05). However, the mean range of motion in the sagittal and coronal planes in Group III was not found to be significantly different than that in Group II (p > 0.05). In addition, no significant differences with regard to motion were detected between any of the Kirschner-wire constructs in Group III.
With use of the t test, we could not detect an increase in the stability provided by outrigger fixation compared with that provided by Kirschner-wire transfixion (p = 0.62). We also found that all of the augmented constructs were significantly more stable than all of the constructs in Group I (p < 0.001).
Although Abraham Colles was evidently satisfied with the results of his treatment of distal radial fractures in 18149, more recent authors have drawn attention to the high prevalence of unsatisfactory results. In 1952, DePalma hypothesized that a residual dorsal tilt of the distal end of the radius of more than 5 degrees led to a poor result13. Gartland and Werley found that immobilization of a distal radial fracture in a cast resulted in a 60 percent loss of reduction and an unsatisfactory result with regard to pain and loss of function in nineteen (32 per cent) of sixty patients17. Cole and Obletz documented radial shortening of three millimeters or more in twenty-two (67 percent) of the thirty-three patients and radial shortening of six millimeters or more in eleven patients (33 percent) after fixation with pins and plaster8. Chapman et al. reported radial shortening of five millimeters or more in twenty (25 percent) of eighty patients who had been managed with the same technique; a complication led to a reoperation thirteen patients (16 percent)6. Short et al. found that loss of volar tilt after a distal radial fracture led to progressive load on the ulnocarpal and radioscaphoid articulations, which caused pain and early degenerative disease29. Taleisnik and Watson reported an association between malunion of the distal end of the radius and dynamic midcarpal instability31.
Unsatisfied with the available methods of treatment, Cooney et al., in 1979, critically reviewed external fixation for the treatment of distal radial fractures and reported a good result for fifty-one (85 percent) of sixty patients, with decreased radial shortening and improved volar tilt11. Since then, external fixation has become a popular and reliable method for the treatment of these frequently seen fractures. A common algorithm for unstable distal radial fractures is external fixation, supplemental fixation with Kirschner wires, and, frequently, the use of a bone graft or bone substitute2,3,14,18,26. Although early motion is thought to be important after any articular fracture, concern over loss of stability of the fracture has made early motion difficult to achieve after external fixation of the distal end of the radius.
In clinical practice, early removal of a standard external fixator is desirable, in order to prevent stiffness and to lessen the chance of a deep pin-track infection, provided that the removal does not decrease the stability of the fracture. Increased distraction across the wrist and an increased duration of external fixation are associated with adverse effects. With use of the carpal height index to quantify distraction, Kaempffe et al. discovered that an increase in the duration and amount of distraction of distal radial fractures treated with external fixation adversely affected the final range of motion of the wrist, function, grip strength, and level of pain21. The scores for motion were affected most strongly.
In response to this problem, some authors have recommended the use of a dynamic external fixator that allows a range of motion of the wrist during healing7. However, Sommerkamp et al., in a prospective, randomized study, showed that the use of an articulated external fixator yielded inferior subjective and objective outcomes compared with those obtained with a standard external fixator30. A fundamental concept of external fixation of the distal end of the radius is that ligamentotaxis helps to stabilize the fracture; the dynamic external fixator does not do this consistently.
In the present study, we created an in vitro osteotomy model to test the stability of a partially healed distal radial fracture after removal of the distal pin cluster of a standard external fixator. We demonstrated that, in such a model, the addition of Kirschner wires allowed the distal pin cluster of the external fixator to be removed. This was possible because the Kirschner wires provided stability that was comparable with that in the model of an acute fracture with standard augmented external fixation. We also demonstrated that a variety of Kirschner-wire constructs could be used to achieve similar stability.
There are several limitations to the present study. The results in our extra-articular fracture model cannot be applied to more comminuted intra-articular fractures. In addition, although the loads across the wrist during activities of daily living have been estimated, the values have been based on theoretical calculations1,5,10. We used a maximum load of approximately 100 newtons on the basis of both these theoretical calculations and the results of similar biomechanical investigations of instability of the wrist, disorders of the distal radioulnar joint, and fractures of the wrist11,25,27,32. Finally, we are not sure if caulk is the most appropriate material with which to simulate partial healing. The indentation stiffness of the caulking material was 0.55 newton per millimeter, which is approximately 5 percent of the stiffness that Markel et al. reported in their analysis of two-week-old tissue from the tibial fracture gap in a canine model22. Thus, although caulk is considerably less stiff than the gap tissue in canines, we believed that the material would acceptably simulate partial healing because it would not overestimate the stiffness of tissue in a fracture gap. On the basis of our data, we were unable to define an exact state of fracture-healing at which it is safe to dismantle the external fixator, but we were able to conclude that, at some point before complete healing occurs, an augmented partially healed fracture is strong enough to tolerate moderate loads without a change in the alignment of the fracture and it is as stable as an acute fracture treated with augmented external fixation.
In conclusion, our findings support the concept of modular disassembly of an external fixator to allow an early range of motion of a wrist with a partially healed distal radial fracture treated with supplemental Kirschner-wire fixation. In our in vitro fracture model, we found that a combination of simulated partial healing and supplemental fixation with Kirschner wires provided stability that was comparable with that of an acute simulated fracture treated with standard augmented external fixation.
*Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received, but are directed solely to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Orthologic, Incorporated.
Investigation performed at the Yale Hand Biomechanics Laboratory, Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, New Haven
1. An, K. N.; Chao, E. Y.; Cooney, W. P.; and and Linscheid, R. L.: Forces in the normal and abnormal hand. J. Orthop. Res.
, 3: 202-211, 1985.
2. Axelrod, T. S., and and McMurtry, R. Y.: Open reduction and internal fixation of comminuted, intraarticular fractures of the distal radius. J. Hand Surg.
, 15A: 1-11, 1990.
3. Bass, R. L.; Blair, W. F.; and and Hubbard, P. P.: Results of combined internal and external fixation for the treatment of severe AO-C3 fractures of the distal radius. J. Hand Surg.
, 20A: 373-381, 1995.
4. Braun, R. M., and and Gellman, H.: Dorsal pin placement and external fixation for correction of dorsal tilt in fractures of the distal radius. J. Hand Surg.
, 19A: 653-655, 1994.
5. Chao, E. Y.; Opgrande, J. D.; and and Axmear, F. E.: Three-dimensional force analysis of finger joints in selected isometric hand functions. J. Biomech.
, 9: 387-396, 1976.
6. Chapman, D. R.; Bennett, J. B.; Bryan, W. J.; and and Tullos, H. S.: Complications of distal radial fractures: pins and plaster treatment. J. Hand Surg.
, 7: 509-512, 1982.
7. Clyburn, T. A.: Dynamic external fixation for comminuted intra-articular fractures of the distal end of the radius. J. Bone and Joint Surg.
, 69-A: 248-254, Feb. 1987.
8. Cole, J. M., and and Obletz, B. E.: Comminuted fractures of the distal end of the radius treated by skeletal transfixion in plaster cast. An end-result study of thirty-three cases. J. Bone and Joint Surg.
, 48-A: 931-945, July 1966.
9. Colles, A.: On the fracture of the carpal extremity of the radius. Edinburgh Med. and Surg. J.
, 10: 182-186, 1814.
10. Cooney, W. P., III, and and Chao, E. Y. S.: Biomechanical analysis of static forces in the thumb during hand function. J. Bone and Joint Surg.
, 59-A: 27-36, Jan. 1977.
11. Cooney, W. P., III; Linscheid, R. L.; and and Dobyns, J. H.: External pin fixation for unstable Colles' fractures. J. Bone and Joint Surg.
, 61-A: 840-845, Sept. 1979.
12. Crisco, J. J., and Wolfe, S. W.: In vivo load-displacement behavior of the carpal scaphoid ligament complex: initial measurements, time-dependence and repeatability. In Advances in the Biomechanics of the Hand and Wrist, pp. 457-463. Edited by F. Schuind. New York, Plenum Press, 1994.
13. DePalma, A. F.: Comminuted fractures of the distal end of the radius treated by ulnar pinning. J. Bone and Joint Surg.
, 34-A: 651-662, July 1952.
14. Fitoussi, F.; Ip, W. Y.; and and Chow, S. P.: Treatment of displaced intra-articular fractures of the distal end of the radius with plates. J. Bone and Joint Surg.
, 79-A: 1303-1312, Sept. 1997.
15. Frykman, G. K.; Tooma, G. S.; Boyko, K.; and and Henderson, R.: Comparison of eleven external fixators for treatment of unstable wrist fractures. J. Hand Surg.
, 14A: 247-254, 1989.
16. Frykman, G. K.; Peckham, R. H.; Willard, K.; and and Saha, S.: External fixators for treatment of unstable wrist fractures. A biomechanical, design feature, and cost comparison. Hand Clin.
, 9: 555-565, 1993.
17. Gartland, J. J., and and Werley, C. W.: Evaluation of healed Colles' fractures. J. Bone and Joint Surg.
, 33-A: 895-907, Oct. 1951.
18. Geissler, W. B., and and Fernandez, D. L.: Percutaneous and limited open reduction of the articular surface of the distal radius. J. Orthop. Trauma
, 5: 255-264, 1991.
19. Hara, T.; Horii, E.; An, K. N.; Cooney, W. P.; Linscheid, R. L.; and and Chao, E. Y.: Force distribution across wrist joint: application of pressure-sensitive conductive rubber. J. Hand Surg.
, 17A: 339-347, 1992.
20. Horii, E.; Garcia-Elias, M.; An, K. N.; Bishop, A. T.; Cooney, W. P.; Linscheid, R. L.; and and Chao, E. Y. S.: Effect on force transmission across the carpus in procedures used to treat Kienböck's disease. J. Hand Surg.
, 15A: 393-400, 1990.
21. Kaempffe, F. A.; Wheeler, D. R.; Peimer, C. A.; Hvisdak, K. S.; Ceravolo, J.; and and Senall, J.: Severe fractures of the distal radius: effect of amount and duration of external fixator distraction on outcome. J. Hand Surg.
, 18A: 33-41, 1993.
22. Markel, M. D.; Wikenheiser, M. A.; and and Chao, E. Y.: A study of fracture callus material properties: relationship to the torsional strength of bone. J. Orthop. Res.
, 8: 843-850, 1990.
23. Missakian, M. L.; Cooney, W. P.; Amadio, P. C.; and and Glidewell, H. L.: Open reduction and internal fixation for distal radius fractures. J. Hand Surg.
, 17A: 745-755, 1992.
24. Nakata, R. Y.; Chand, Y.; Matiko, J. D.; Frykman, G. K.; and and Wood, V. E.: External fixators for wrist fractures; a biomechanical and clinical study. J. Hand Surg.
, 10A: 845-851, 1985.
25. Palmer, A. K., and and Werner, F. W.: Biomechanics of the distal radioulnar joint. Clin. Orthop.
, 187: 26-35, 1984.
26. Pike, L. M., and and Wolfe, S. W.: Alternatives to bone graft in the treatment of distal radius fractures. Atlas Hand Clin.
, 2: 125-150, 1997.
27. Ruby, L. K.; An, K. N.; Linscheid, R. L.; Cooney, W. P., III; and and Chao, E. Y. S.: The effect of scapholunate ligament section on scapholunate motion. J. Hand Surg.
, 12A: 767-771, 1987.
28. Seitz, W. H., Jr.; Froimson, A. I.; Leb, R.; and and Shapiro, J. D.: Augmented external fixation of unstable distal radius fractures. J. Hand Surg.
, 16A: 1010-1016, 1991.
29. Short, W. H.; Palmer, A. K.; Werner, F. W.; and and Murphy, D. J.: A biomechanical study of distal radial fractures. J. Hand Surg.
, 12A: 529-534, 1987.
30. Sommerkamp, T. G.; Seeman, M.; Silliman, J.; Jones, A.; Patterson, S.; Walker, J.; Semmler, M.; Browne, R.; and and Ezaki, M.: Dynamic external fixation of unstable fractures of the distal part of the radius. A prospective, randomized comparison with static external fixation. J. Bone and Joint Surg.
, 76-A: 1149-1161, Aug. 1994.
31. Taleisnik, J., and and Watson, H. K.: Midcarpal instability caused by malunited fractures of the distal radius. J. Hand Surg.
, 9A: 350-357, 1984.
32. Trumble, T.; Glisson, R. R.; Seaber, A. V.; and and Urbaniak, J. R.: Forearm force transmission after surgical treatment of distal radioulnar joint disorders. J. Hand Surg.
, 12A: 196-202, 1987.
33. Wolfe, S. W.; Swigart, C. R.; Grauer, J.; Slade, J. F., III; and and Panjabi, M. M.: Augmented external fixation of distal radius fractures: a biomechanical analysis. J. Hand Surg.
, 23A: 127-134, 1998.