- Email: [email protected]
Surgical management of delayed union and nonunion of distal radius fractures
Original Communications Surgical Management of Delayed Union and Nonunion of Distal Radius Fractures Diego L. Fernandez, MD, Berne, Switzerland, David Ring, MD, Jesse B. Jupiter, MD, Boston, MA Ten patients with malaligned fractures of the distal radius that demonstrated either delayed healing or the development of an atrophic or synovial nonunion on standard radiographs were treated with surgical realignment, stable internal fixation, and autogenous iliac crest bone grafting. All 10 fractures healed with acceptable radiologic alignment within 3 months of the index procedure. After an average follow-up period of 3 years 6 months (range, 2 years to 8 years 6 months) patients had an average of 105° wrist flexion and extension, 145° forearm rotation, and 73% grip strength compared with the opposite limb. In the treatment of malaligned, ununited fractures of the distal radius, specific techniques and implants must be tailored to the deformity of the distal radius and the shape of the distal fragment. A stable, well-aligned wrist with preservation of at least 50° mobility in flexion and extension was achieved in every patient, but the final result was compromised by associated problems in 3 patients. (J Hand Surg 2001;26A:201–209. Copyright © 2001 by the American Society for Surgery of the Hand.) Key words: Distal radius, wrist, nonunion, bone graft, internal fixation.
The management of fractures of the distal end of the radius has focused on the maintenance of skeletal alignment and restoration of function of the hand and upper limb. Fracture union, in contrast, has less often
From the Department of Orthopaedic Surgery, University of Berne, Switzerland, and Lindenhof Hospital, Berne, Switzerland; and the Department of Orthopaedic Surgery, Harvard Medical School Hand and Upper Extremity Service, Massachusetts General Hospital, Boston, MA. Received for publication May 5, 2000; accepted in revised form December 1, 2000. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Jesse B. Jupiter, MD, Department of Orthopaedics, Harvard Medical School Hand and Upper Extremity Service, Massachusetts General Hospital, ACC 527, 15 Parkman St, Boston, MA 02114. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A02-0009$35.00/0 doi:10.1053/jhsu.2001.22917
been a source of concern. With increased interest in surgical intervention in the treatment of some distal radius fractures has come recognition that arrest in bony healing can occur. When associated with deformity, surgical correction of an ununited fracture of the distal radius can be problematic (due to a small, osteopenic distal fragment, associated soft tissue contracture, and atrophic status at the site of delayed healing) and has led some to recommend total wrist arthrodesis.1,2 Several series describe surgical attempts to gain union.3–5 Segalman and Clark4 have suggested that surgical attempts to gain union are worthwhile when at least 5 mm subchondral bone beneath the lunate facet of the distal radius is available for the application of implants.4 Given the fact that the radiocarpal and midcarpal articulations are often uninvolved, an attempt to maintain functional mobility of the wrist by obtaining both anatomic realignment of the distal fragment The Journal of Hand Surgery 201
202 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
and union of the fracture seems warranted. Improvements in the implants and surgical techniques for the fixation of small, periarticular fractures have made these goals more realistic. Failure to achieve union can be salvaged with total wrist arthrodesis.5 We review our experience with delayed union or nonunion and malalignment of fractures of the distal radius treated by debridement, bony alignment, internal fixation, and autogenous bone grafting.
Materials and Methods Ten patients with fractures of the distal radius with delayed union or nonunion were referred to 2 surgeons with an interest in surgery of the wrist. One surgeon saw 6 patients over a 15-year period and the other saw 4 patients over a 5-year period. Surgical treatment was recommended based on the combination of delayed union or nonunion and/or malalignment. Variations in treatment reflect both the evolution in implants and techniques that occurred over the study period and the need for distinct approaches to correct various deformities. Fractures that lack radiographic signs of union (ie, bony trabeculae crossing the fracture site) are commonly referred to as delayed unions at 4 months and nonunions at 6 months after the injury. Seven patients had painful, unstable atrophic nonunions of ⱖ6 months’ duration. Three patients had delayed union with an atrophic fracture site between 4 and 6 months after the time of the injury: 1 patient had an atrophic nonunion with loose internal fixation 4 months after plate and screw fixation (patient 5) and 2 had more stable fibrous unions of between 4 and 6 months’ duration (patients 1 and 10). In 8 patients the indication for surgical treatment was based on malalignment in addition to delayed union or nonunion. The most common deformity was radial deviation of the distal fragment or loss of ulnar inclination: 7 patients had ⱕ10° ulnar inclination. Ulnar inclination of the distal radial articular surface averaged 9° (range, -5° to 25°). One patient had more than 20° dorsal tilt of the distal radial articular surface and another had excessive palmar tilt (⬎25°). Palmar tilt averaged 7° (range, -10° to 45°). In 3 patients the ulna had ⱖ3 mm positive variance; the ulnar variance averaged 2 mm (range, -2 to 5 mm). Information was obtained from chart review and telephone interview in all cases. Telephone interviews were used to verify functional status and demographic data, but the final follow-up report reflects the latest clinical and radiographic follow-up
examination. Six of 10 patients returned for a final interview and examination at the time the data were compiled. For all patients a complete set of standard posteroanterior and lateral radiographs were available from the time of injury, the time of presentation, after the initial procedure, and at the final follow-up examination. The patient characteristics are given in Table 1. There were 4 men and 6 women with an average age of 46 years (range, 19 –78 years). All the patients were right-handed. The dominant limb was involved in 5 patients and the nondominant limb in 5. The original fracture was the result of a high-energy mechanism in 6 patients and a fall from a standing height in 4. The original fracture was classified according to the system of Fernandez and Jupiter6 as a bending fracture (type 1) in 5 patients (3 dorsal and 2 volar), a compression-type fracture (type 2) in 2 patients, and a complex high-energy injury (type 5) in 3 patients. According to the comprehensive classification of fractures,7 6 fractures were classified as group A2 and 1 each as group A3, C1, C2, and C3. Associated injuries included a fracture of the scaphoid in 1 patient and a carpal ligament injury (initially unrecognized) in 1 patient. In 2 patients the initial fracture was treated with cast immobilization; the other 8 were treated surgically. Surgical treatment consisted of external fixation in 1 patient, external fixation and percutaneous insertion of K-wires in 5 patients, and open reduction and internal fixation in 2 patients (one with K-wires and staples and the others with a dorsal T-shaped plate). One of the patients who had external fixation and percutaneous K-wire fixation also had plate and screw fixation of a fracture of the ulna. One patient had morselized allograft cancellous bone placed in a metaphyseal bony defect, but no patient received an autogenous bone graft. None of the patients had further surgical attempts to gain union of the fracture before presentation. All 10 patients had wrist pain. One patient, a 78-year-old woman, had a synovial nonunion with a 50° radial deviation deformity and marked instability at the fracture site (Fig. 1). It was not possible to accurately measure wrist motion in this patient. The average preoperative motion in the remaining 9 patients was 35° extension (range, 0° to 50°), 45° flexion (range, 20° to 70°), 5° radial deviation (range, 0° to 10°), 15° ulnar deviation (range, 0° to 30°), 60° pronation (range, 0° to 80°), and 30° supination (range, 0° to 45°). Grip strength, measured using a Jamar dynamometer (Asimov Engineering; Los An-
39/F
78/F 42/F
43/F 48/M
60/F
28/M
41/M 66/F
2
3 4
5 6
7
8
9 10
R L
L
R
R L
L R
L
R
Bending
High energy High energy
Bending (volar) Compression
MVA Fall (height)
Ice hockey Bending (volar) High energy
Bending (⫹scaphoid fracture)
Fall (standing) Compression
Fall (standing) MVA (intraarticular fracture) MVA Fall (height)
A2 A3
A2
C1
A2 C3
A2 C2
A2
A2
Comprehensive Fernandez Classification6 Classification7
Fall (standing) Bending
Skateboarding
Mechanism of Injury
MVA, motor vehicle accident; ORIF, open reduction and internal fixation. * All the patients were right-handed.
19/M
1
Case Age (yr)/ Affected No. Gender Limb* Initial Treatment External fixation, K-wires External fixation, K-wires Cast External fixation, K-wires, allograft bone Dorsal T-plate Ulna plate, external fixation, and K-wires External fixation and K-wires ORIF K-wires and staples, Herbert screw fixation of scaphoid Cast External fixation 7 5
8
6.5
4 9
25 8
10
4
50 35
20
45
30 35
— 50
0
40
40 20
35
50
45 30
— 70
30
70
60 40
75
65
80 70
— 70
0
70
45 30
40
35
40 15
— 25
0
20
10 0
10
0
10 5
— 5
0
10
15 10
5
20
20 12
— 30
0
20
45 31
31
38
57 38
5 67
7
42
Time From Preoperative Motion (°) Injury to Initial Grip Strength Procedure Radial Ulnar (% Opposite (mo) Extension Flexion Pronation Supination Deviation Deviation Limb)
Table 1. Preoperative Data for 10 Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 203
204 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
Figure 1. A 78-year-old woman injured her left (nondominant) wrist in a fall. A volarly displaced metaphyseal fracture associated with a fracture of the distal ulna was treated with cast immobilization. (A) Twenty-five months after the injury she presented with a painful, unstable, and severely deformed wrist. (B) Posteroanterior and (C) lateral radiographs demonstrate a synovial nonunion of the distal radius. (Figure continues)
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 205
Figure 1. (Continued) Two years after internal fixation and autogenous cancellous bone grafting of the radius and resection of the deformed distal ulna, (D) posteroanterior and (E) lateral radiographs demonstrate solid healing in good alignment. Stable fixation was achieved with 2 plates incorporating fixed-angled bolts distally and oriented orthogonal to one another.
geles, CA), averaged 36 of the opposite uninvolved limb (range, 7% to 67%). The average interval from the time of injury to the index procedure for open reduction and internal fixation of an ununited fracture of the distal radius was 9 months (range, 4 –25 months). Only 3 patients had surgery less than 6 months after the original injury. The procedure included debridement of the nonunion site, including any synovial membrane and avascular bone; realignment with a small skeletal distractor; plate and screw fixation; and autogenous iliac crest cancellous bone grafting (Table 2). The surgical approach and plate fixation were dependent on the direction of the deformity and the architecture of the distal fragment. Six patients were treated through a volar8 and 4 through a dorsal9 exposure. Because the patients in the study were treated over a 15-year period a number of different types of implants were used, reflecting the evolution of implants designed specifically for the distal end of the radius. Plates used on the volar surface of the distal radius included
a 3.5-mm T-shaped plate in 4 patients (supplemented by a minifragment blade plate applied to the radial surface of the distal radius in 1 patient and an interfragmentary compression screw in another), a 3.5-mm dynamic compression plate in 1 patient, and a 2.7-mm T-shaped plate supplemented with 2 Kwires in 1 patient. Plates used on the dorsal surface included a -plate in 1 patient,9 a 3.5-mm dynamic compression plate in 1 patient, a semitubular plate in 1 patient, and a 2.0-mm condylar blade plate supplemented with a K-wire in 1 patient. Arthrolysis of the radiocarpal and distal radioulnar articulations were performed in 2 patients in an attempt to help restore functional motion. Z-lengthening of the brachioradialis was required in 1 patient to achieve restoration of radial length. Four patients with incongruity or arthrosis of the distal radioulnar joint had either a hemiresection interposition (Bowers’) arthroplasty10,11 or a Darrach resection of the distal ulna (Fig. 1). The standardized rating scale applied by Fernan-
1 2
45
28 40 102
24 36 30
38
36 48
40
70 65 70
60 65 60
40
70 25
Extension
DRUJ, distal radioulnar joint; RC, radiocarpal; DCP, dynamic compression plate.
10
7 8 9
4 5 6
3
Initial Procedure
Volar T-plate Dorsal 2.7-mm blade plate, K-wire, Bower’s DRUJ arthroplasty, arthrolysis RC joint Volar T-plate, radial 2.7-mm blade plate; Z-lengthening of brachioradialis, Darrach resection of distal ulna Dorsal -plate; DRUJ release Volar 3.5-mm DCP Dorsal 3.5-mm DCP; Bower’s DRUJ arthroplasty Volar T-plate Volar 2.7-mm T-plate, K-wires Volar plate T-plate, interfragmentary compression screw Dorsal semitubular plate
Case No.
Length of Follow-up (mo)
45
55 55 65
30 55 45
40
70 30
Flexion
60
85 90 90
60 80 75
70
80 80
Pronation
70
75 85 90
60 85 70
50
60 40
Supination
Postoperative Motion (°)
15
10 10 20
20 15 10
10
10 10
Radial Deviation
15
25 20 30
30 20 15
10
20 15
Ulnar Deviation
58
63 90 120
51 73 67
67
71 70
Grip Strength (% of Opposite Side)
Fair
Good Excellent Excellent
Poor Good Good
Good
Excellent Fair
Overall Rating According to Fernandez
Complications
—
— — —
— — —
— Superficial wound infection at iliac crest incision —
Table 2. Operative and Post-operative Data for Ten Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius
206 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 207
Table 3. Rating of the Functional Result According to Fernandez11 Radiocarpal pain No pain Mild Moderate Severe Distal radioulnar pain Severe Moderate Mild No pain Wrist flexion and extension Arc of 130° to 140° Arc of 100° to 129° Arc of 80° to 99° ⬍80° arc Forearm rotation Arc of 160° to 180° Arc of 140° to 159° Arc of 120° to 139° ⬍120° arc Grip strength (compared with opposite limb) ⱖ80% 65% to 79% 40% to 64% ⬍40%
4 points 2 points 1 point 0 points 0 points 1 point 2 points 4 points 4 points 3 points 2 points 1 point 4 points 3 points 2 points 1 point 4 points 3 points 2 points 1 point
Overall rating: 18 to 20 points, excellent; 15 to 17 points, good; 12 to 14 points, fair; ⬍11 points, poor.
dez11 in the analysis of the treatment of malunited fractures of the distal radius was used to assess the overall functional result (Table 3). This is one of the most commonly used rating systems used to evaluate reconstructive surgery of the distal radius. The final follow-up radiographs were evaluated for the presence of posttraumatic arthrosis according to the criteria of Knirk and Jupiter.12 Grade 0 is defined
as no arthrosis, grade 1 is slight joint space narrowing, grade 2 is marked joint space narrowing and osteophyte formation, and grade 3 is complete absence of a joint space with osteophytes and cysts.
Results The patients were monitored for an average of 3 years 6 months (range, 2 years to 8 years 6 months). Fracture union was noted in each patient within 3 months of the initial procedure. According to the rating system of Fernandez11 there were 3 excellent, 4 good, 2 fair, and 1 poor result. An average of 55° wrist extension (range, 25° to 70°) and 50° flexion (range, 30° to 70°) were restored. Radial and ulnar deviation averaged 15° (range, 10° to 20°) and 20° (range, 15° to 30°), respectively. The average pronation and supination was 75° (range, 60° to 90°) and 70° (range, 40° to 90°), respectively. Postoperative grip strength averaged 73% percent of the opposite limb (range, 51% to 120%). The postoperative alignment of the distal radial articular surface was measured on posteroanterior and lateral radiographs (Table 4). Ulnar inclination of the distal radial articular surface averaged 22° (range, 15° to 25°). Palmar tilt averaged 5° (range, 0° to 20°). Ulnar variance could not be measured in the 4 patients who had partial or complete resection of the distal ulna. Ulnar variance in the remaining 6 patients averaged 0 mm (range, -2–2 mm). Radiographic evidence of arthrosis was graded according to the scale of Knirk and Jupiter12 as follows: grade 0 in 4 patients, grade 1 in 5 patients, and grade 2 in 1 patient. All 3 of the unsatisfactory results were related to problems associated with the initial injury and subsequent treatment course. One patient had severe
Table 4. Radiographic Data From 10 Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius Case No.
Pre/Postoperative Ulnar Inclination (°)
Pre/Postoperative Palmar Tilt (°)
1 2 3 4 5 6 7 8 9 10
25/25 20/25 ⫺5/15 10/20 10/18 ⫺5/18 5/20 0/25 0/24 25/25
⫺25/0 ⫺10/5 45/20 ⫺5/10 5/0 0/0 ⫺5/0 ⫺10/0 20/0 ⫺10/10
Pre/Postoperative Ulnar Variance (mm) 2/2 1/Bowers 5/Darrach 1/Bowers 3/0 2/Bowers 0/1 ⫺2/⫺2 4/0 ⫺1/0
Arthrosis: Knirk/ Jupiter Grade 0 0 1 1 0 2 1 1 1 0
208 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
stiffness after prolonged immobilization in an external fixator and then a cast. She regained 55° flexion and extension through the radiocarpal articulation and 120° forearm rotation, but still had mild pain. Another patient had an initially unrecognized carpal ligament injury contributing to a carpal instability that limited motion and grip strength and caused pain. The third unsatisfactory result was due to persistent discomfort in a patient whose initial fracture was the result of high-energy trauma. There were few complications due to the surgical treatment in this series. One obese patient had a superficial infection in the wound used to access the iliac crest. This was treated with debridement and parenteral antibiotics followed by oral antibiotics. Three patients requested plate removal (2 volar plates and 1 dorsal plate), but none had symptoms related to the plate.
Discussion Although healing problems at the distal radius have historically been considered rare,13–16 they have been discussed more frequently during the last decade.2–5,17 Some investigators have speculated that delay in healing of a fracture of the distal radius may have become more common since the advent of surgical techniques that restore the length of the fractured distal radius and thus create a bony gap in the metaphysis that may delay healing.1,4 Healing problems in the distal radius also seem to be related to unstable situations, such as concomitant fracture of the distal radius and ulna, as well as to an inadequate period of immobilization.1–5 Patients with delayed union or nonunion of the distal radius may have dysfunction related to pain and instability, but many problems arise from the associated malalignment. The treatment therefore resembles the treatment for malunion of the distal radius except that the surgeon must also address the delayed or arrested attempts at healing. The definition of delayed union or nonunion is straightforward in the presence of an atrophic nonunion, particularly a synovial pseudoarthrosis. Patients with these types of nonunions typically present with severe dysfunction related to deformity, instability, and pain. Nonoperative treatment is appropriate only in patients with limited functional demands (such as the invalid elderly) in whom symptoms can sometimes be managed in a splint. In the case of a stable fibrous nonunion, the indication for surgery relates more to dysfunction associated with malalign-
ment of the fracture fragments in an active person. Further immobilization to encourage fracture healing would only prolong and exacerbate this dysfunction and would not address the malalignment; such care would be appropriate only in patients with limited functional demands. In active patients, intervention is indicated as soon as dysfunction and limited healing become apparent. A metaphyseal fracture in a healthy patient should demonstrate signs of progressive healing within 3 months on standard radiographs. Patients with limited radiographic evidence of healing 4 months after the injury should be considered to have delayed healing The variety of techniques used in this series reflects the evolution of surgical techniques and implants during the study period, which was long by virtue of the relative rarity of this problem. This variation also reflects the need to address specific characteristics of the deformity with specific techniques. For instance, a dorsal approach and implant are used when there is dorsal displacement or angulation of the distal fragment or when a dorsal exposure is needed to remove loose implants. A volar exposure is used for volarly directed displacement or loose volar implants. If the alignment is relatively neutral, then a volar exposure is usually preferred because volar implants are less likely in our experience to irritate or damage the overlying tendons. Due to the rarity of this problem it is difficult to make treatment recommendations based on data analysis. It appears that a variety of implants and implant combinations can be successful provided that stability is achieved. Our current preferred technique, as reflected in the treatment of 1 of the more recent patients in the series (Fig. 1), reflects our interest in newer implants providing fixed-angle fixation and fixation in orthogonal planes, both of which can be more secure in poor-quality bone. Internal fixation should focus on the radial styloid as this portion of the distal fragment usually offers a greater amount of bone for the application of fixation devices. One plate can be applied directly on the radial aspect of the distal radius and a second on the volar or dorsal aspect. We offer additional recommendations based on our experience with these patients, although they cannot be supported statistically due to the rarity of this problem. In the presence of severe malalignment of the fracture it may be necessary to lengthen or divide the brachioradialis tendon to facilitate realignment. When the wrist is stiff, particularly in the presence of a synovial nonunion, incision of the
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 209
dorsal radiocarpal and/or volar distal radioulnar joint capsule should be considered. Arthrosis, deformity, or incongruity of the distal radioulnar joint must be addressed. Based on our experience we favor resection of the distal ulna (Darrach procedure), particularly when there is severe shortening of the distal radius that cannot be overcome. While it can be argued that some of the fractures in our series might have healed with additional immobilization, surgery performed as soon as delayed union or nonunion and malalignment are recognized addresses the radiocarpal and radioulnar joint dysfunction caused by the deformity and limits the need for prolonged external immobilization, which might exacerbate stiffness. The dysfunction associated with malalignment is the primary indication for surgical reconstruction while the delayed healing is the factor that increases the complexity of the problem and the difficulty of the procedure. The results of attempts to gain union of ununited fractures of the distal radius are not entirely predictable, as demonstrated by the fact that 3 of our patients had unsatisfactory results. Patients with such complex problems often present with associated concerns that can degrade the outcome. On the other hand, in each of our patients, surgical realignment of the distal fragment substantially enhanced radiocarpal and radioulnar mobility as well as carpal kinematics, as reflected by an increase in grip strength. Although the results in 3 patients were rated unsatisfactory using the strict scale of Fernandez,11 the goals of surgery (ie, a stable, healed, well-aligned wrist with preservation of some radiocarpal and midcarpal motion) were met after a single procedure in each patient and substantial functional improvement was observed. In other words, although patients and surgeons should expect some limitation of motion and incomplete relief of pain (occasionally substantial), our experience supports performing surgery in an attempt to gain union of fractures of the distal radius associated with delayed union or nonunion rather than proceeding directly to wrist arthrodesis in all cases. Wrist motion facilitates positioning of the hand in space, enhances strength and precision of the digits, and provides important proprioceptive feedback. It is therefore preferable to preserve even a small amount
of wrist motion. Our experience suggests that most ununited fractures of the distal radius are amenable to attempts to realign and heal the fracture.
References 1. Weber SC, Szabo RM. Severely comminuted distal radial fractures as an unsolved problem: complications associated with external fixation and pins and plaster techniques. J Hand Surg 1986;11A:157–165. 2. McKee MD, Waddell JP, Yoo D, Richards RR. Nonunion of distal radial fractures associated with distal ulnar shaft fractures: a report of four cases. J Orthop Trauma 1997; 11:49 –53. 3. Harper WM, Jones JM. Non-union of Colles’ fracture: report of two cases. J Hand Surg 1990;15B:121–123. 4. Segalman KA, Clark GL. Un-united fractures of the distal radius: a report of 12 cases. J Hand Surg 1998;23A:914 – 919. 5. Smith VA, Wright TW. Nonunion of the distal radius. J Hand Surg 1999;24B:601– 603. 6. Fernandez DL, Jupiter JB. Fractures of the distal radius: a practical approach to management. New York: SpringerVerlag, 1996. 7. Mu¨ller ME, Nazarian S, Koch P, Schatzker J. The comprehensive classification of fractures of long bones. Berlin: Springer-Verlag, 1990. 8. Henry AK. The upper limb and neck. In: Extensile exposure. 2nd ed. Edinburgh: Churchill Livingstone, 1973; 100 –107. 9. Ring D, Jupiter JB, Brennwald J, Bu¨chler U, Hastings H II. Prospective multicenter trial of a plate for dorsal fixation of distal radius fractures. J Hand Surg 1997;22A:777–784. 10. Bowers WH. Distal radioulnar joint arthroplasty: the hemiresection-interposition technique. J Hand Surg 1985; 10A:169 –178. 11. Fernandez DL. Radial osteotomy and Bowers arthroplasty for malunited fractures of the distal end of the radius. J Bone Joint Surg 1988;70A:1538 –1551. 12. Knirk JL, Jupiter JB. Intra-articular fractures of the distal end of the radius in young adults. J Bone Joint Surg 1986;68A:647– 659. 13. Bacorn RW, Kurtzke JF. A study of two thousand cases from the New York State Workmen’s Compensation Board. J Bone Joint Surg 1953;38A:643– 658. 14. Hamada G. Extra-articular graft for non-union in Colles’ fracture. J Bone Joint Surg 1944;26:833– 835. 15. Lachman JW. Report of the Meeting of the Orthopaedic Associations of the English-speaking world. J Bone Joint Surg 1958;40B:595–599. 16. Watson-Jones R. Injuries of the wrist. In: Fractures and other bone and joint injuries. 3rd ed. Baltimore: Williams & Wilkins, 1943;538 –545. 17. Szabo RM, Weber SC. Comminuted intraarticular fractures of the distal radius. Clin Orthop 1988;230:39 – 48.
The management of fractures of the distal end of the radius has focused on the maintenance of skeletal alignment and restoration of function of the hand and upper limb. Fracture union, in contrast, has less often
From the Department of Orthopaedic Surgery, University of Berne, Switzerland, and Lindenhof Hospital, Berne, Switzerland; and the Department of Orthopaedic Surgery, Harvard Medical School Hand and Upper Extremity Service, Massachusetts General Hospital, Boston, MA. Received for publication May 5, 2000; accepted in revised form December 1, 2000. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Jesse B. Jupiter, MD, Department of Orthopaedics, Harvard Medical School Hand and Upper Extremity Service, Massachusetts General Hospital, ACC 527, 15 Parkman St, Boston, MA 02114. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A02-0009$35.00/0 doi:10.1053/jhsu.2001.22917
been a source of concern. With increased interest in surgical intervention in the treatment of some distal radius fractures has come recognition that arrest in bony healing can occur. When associated with deformity, surgical correction of an ununited fracture of the distal radius can be problematic (due to a small, osteopenic distal fragment, associated soft tissue contracture, and atrophic status at the site of delayed healing) and has led some to recommend total wrist arthrodesis.1,2 Several series describe surgical attempts to gain union.3–5 Segalman and Clark4 have suggested that surgical attempts to gain union are worthwhile when at least 5 mm subchondral bone beneath the lunate facet of the distal radius is available for the application of implants.4 Given the fact that the radiocarpal and midcarpal articulations are often uninvolved, an attempt to maintain functional mobility of the wrist by obtaining both anatomic realignment of the distal fragment The Journal of Hand Surgery 201
202 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
and union of the fracture seems warranted. Improvements in the implants and surgical techniques for the fixation of small, periarticular fractures have made these goals more realistic. Failure to achieve union can be salvaged with total wrist arthrodesis.5 We review our experience with delayed union or nonunion and malalignment of fractures of the distal radius treated by debridement, bony alignment, internal fixation, and autogenous bone grafting.
Materials and Methods Ten patients with fractures of the distal radius with delayed union or nonunion were referred to 2 surgeons with an interest in surgery of the wrist. One surgeon saw 6 patients over a 15-year period and the other saw 4 patients over a 5-year period. Surgical treatment was recommended based on the combination of delayed union or nonunion and/or malalignment. Variations in treatment reflect both the evolution in implants and techniques that occurred over the study period and the need for distinct approaches to correct various deformities. Fractures that lack radiographic signs of union (ie, bony trabeculae crossing the fracture site) are commonly referred to as delayed unions at 4 months and nonunions at 6 months after the injury. Seven patients had painful, unstable atrophic nonunions of ⱖ6 months’ duration. Three patients had delayed union with an atrophic fracture site between 4 and 6 months after the time of the injury: 1 patient had an atrophic nonunion with loose internal fixation 4 months after plate and screw fixation (patient 5) and 2 had more stable fibrous unions of between 4 and 6 months’ duration (patients 1 and 10). In 8 patients the indication for surgical treatment was based on malalignment in addition to delayed union or nonunion. The most common deformity was radial deviation of the distal fragment or loss of ulnar inclination: 7 patients had ⱕ10° ulnar inclination. Ulnar inclination of the distal radial articular surface averaged 9° (range, -5° to 25°). One patient had more than 20° dorsal tilt of the distal radial articular surface and another had excessive palmar tilt (⬎25°). Palmar tilt averaged 7° (range, -10° to 45°). In 3 patients the ulna had ⱖ3 mm positive variance; the ulnar variance averaged 2 mm (range, -2 to 5 mm). Information was obtained from chart review and telephone interview in all cases. Telephone interviews were used to verify functional status and demographic data, but the final follow-up report reflects the latest clinical and radiographic follow-up
examination. Six of 10 patients returned for a final interview and examination at the time the data were compiled. For all patients a complete set of standard posteroanterior and lateral radiographs were available from the time of injury, the time of presentation, after the initial procedure, and at the final follow-up examination. The patient characteristics are given in Table 1. There were 4 men and 6 women with an average age of 46 years (range, 19 –78 years). All the patients were right-handed. The dominant limb was involved in 5 patients and the nondominant limb in 5. The original fracture was the result of a high-energy mechanism in 6 patients and a fall from a standing height in 4. The original fracture was classified according to the system of Fernandez and Jupiter6 as a bending fracture (type 1) in 5 patients (3 dorsal and 2 volar), a compression-type fracture (type 2) in 2 patients, and a complex high-energy injury (type 5) in 3 patients. According to the comprehensive classification of fractures,7 6 fractures were classified as group A2 and 1 each as group A3, C1, C2, and C3. Associated injuries included a fracture of the scaphoid in 1 patient and a carpal ligament injury (initially unrecognized) in 1 patient. In 2 patients the initial fracture was treated with cast immobilization; the other 8 were treated surgically. Surgical treatment consisted of external fixation in 1 patient, external fixation and percutaneous insertion of K-wires in 5 patients, and open reduction and internal fixation in 2 patients (one with K-wires and staples and the others with a dorsal T-shaped plate). One of the patients who had external fixation and percutaneous K-wire fixation also had plate and screw fixation of a fracture of the ulna. One patient had morselized allograft cancellous bone placed in a metaphyseal bony defect, but no patient received an autogenous bone graft. None of the patients had further surgical attempts to gain union of the fracture before presentation. All 10 patients had wrist pain. One patient, a 78-year-old woman, had a synovial nonunion with a 50° radial deviation deformity and marked instability at the fracture site (Fig. 1). It was not possible to accurately measure wrist motion in this patient. The average preoperative motion in the remaining 9 patients was 35° extension (range, 0° to 50°), 45° flexion (range, 20° to 70°), 5° radial deviation (range, 0° to 10°), 15° ulnar deviation (range, 0° to 30°), 60° pronation (range, 0° to 80°), and 30° supination (range, 0° to 45°). Grip strength, measured using a Jamar dynamometer (Asimov Engineering; Los An-
39/F
78/F 42/F
43/F 48/M
60/F
28/M
41/M 66/F
2
3 4
5 6
7
8
9 10
R L
L
R
R L
L R
L
R
Bending
High energy High energy
Bending (volar) Compression
MVA Fall (height)
Ice hockey Bending (volar) High energy
Bending (⫹scaphoid fracture)
Fall (standing) Compression
Fall (standing) MVA (intraarticular fracture) MVA Fall (height)
A2 A3
A2
C1
A2 C3
A2 C2
A2
A2
Comprehensive Fernandez Classification6 Classification7
Fall (standing) Bending
Skateboarding
Mechanism of Injury
MVA, motor vehicle accident; ORIF, open reduction and internal fixation. * All the patients were right-handed.
19/M
1
Case Age (yr)/ Affected No. Gender Limb* Initial Treatment External fixation, K-wires External fixation, K-wires Cast External fixation, K-wires, allograft bone Dorsal T-plate Ulna plate, external fixation, and K-wires External fixation and K-wires ORIF K-wires and staples, Herbert screw fixation of scaphoid Cast External fixation 7 5
8
6.5
4 9
25 8
10
4
50 35
20
45
30 35
— 50
0
40
40 20
35
50
45 30
— 70
30
70
60 40
75
65
80 70
— 70
0
70
45 30
40
35
40 15
— 25
0
20
10 0
10
0
10 5
— 5
0
10
15 10
5
20
20 12
— 30
0
20
45 31
31
38
57 38
5 67
7
42
Time From Preoperative Motion (°) Injury to Initial Grip Strength Procedure Radial Ulnar (% Opposite (mo) Extension Flexion Pronation Supination Deviation Deviation Limb)
Table 1. Preoperative Data for 10 Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 203
204 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
Figure 1. A 78-year-old woman injured her left (nondominant) wrist in a fall. A volarly displaced metaphyseal fracture associated with a fracture of the distal ulna was treated with cast immobilization. (A) Twenty-five months after the injury she presented with a painful, unstable, and severely deformed wrist. (B) Posteroanterior and (C) lateral radiographs demonstrate a synovial nonunion of the distal radius. (Figure continues)
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 205
Figure 1. (Continued) Two years after internal fixation and autogenous cancellous bone grafting of the radius and resection of the deformed distal ulna, (D) posteroanterior and (E) lateral radiographs demonstrate solid healing in good alignment. Stable fixation was achieved with 2 plates incorporating fixed-angled bolts distally and oriented orthogonal to one another.
geles, CA), averaged 36 of the opposite uninvolved limb (range, 7% to 67%). The average interval from the time of injury to the index procedure for open reduction and internal fixation of an ununited fracture of the distal radius was 9 months (range, 4 –25 months). Only 3 patients had surgery less than 6 months after the original injury. The procedure included debridement of the nonunion site, including any synovial membrane and avascular bone; realignment with a small skeletal distractor; plate and screw fixation; and autogenous iliac crest cancellous bone grafting (Table 2). The surgical approach and plate fixation were dependent on the direction of the deformity and the architecture of the distal fragment. Six patients were treated through a volar8 and 4 through a dorsal9 exposure. Because the patients in the study were treated over a 15-year period a number of different types of implants were used, reflecting the evolution of implants designed specifically for the distal end of the radius. Plates used on the volar surface of the distal radius included
a 3.5-mm T-shaped plate in 4 patients (supplemented by a minifragment blade plate applied to the radial surface of the distal radius in 1 patient and an interfragmentary compression screw in another), a 3.5-mm dynamic compression plate in 1 patient, and a 2.7-mm T-shaped plate supplemented with 2 Kwires in 1 patient. Plates used on the dorsal surface included a -plate in 1 patient,9 a 3.5-mm dynamic compression plate in 1 patient, a semitubular plate in 1 patient, and a 2.0-mm condylar blade plate supplemented with a K-wire in 1 patient. Arthrolysis of the radiocarpal and distal radioulnar articulations were performed in 2 patients in an attempt to help restore functional motion. Z-lengthening of the brachioradialis was required in 1 patient to achieve restoration of radial length. Four patients with incongruity or arthrosis of the distal radioulnar joint had either a hemiresection interposition (Bowers’) arthroplasty10,11 or a Darrach resection of the distal ulna (Fig. 1). The standardized rating scale applied by Fernan-
1 2
45
28 40 102
24 36 30
38
36 48
40
70 65 70
60 65 60
40
70 25
Extension
DRUJ, distal radioulnar joint; RC, radiocarpal; DCP, dynamic compression plate.
10
7 8 9
4 5 6
3
Initial Procedure
Volar T-plate Dorsal 2.7-mm blade plate, K-wire, Bower’s DRUJ arthroplasty, arthrolysis RC joint Volar T-plate, radial 2.7-mm blade plate; Z-lengthening of brachioradialis, Darrach resection of distal ulna Dorsal -plate; DRUJ release Volar 3.5-mm DCP Dorsal 3.5-mm DCP; Bower’s DRUJ arthroplasty Volar T-plate Volar 2.7-mm T-plate, K-wires Volar plate T-plate, interfragmentary compression screw Dorsal semitubular plate
Case No.
Length of Follow-up (mo)
45
55 55 65
30 55 45
40
70 30
Flexion
60
85 90 90
60 80 75
70
80 80
Pronation
70
75 85 90
60 85 70
50
60 40
Supination
Postoperative Motion (°)
15
10 10 20
20 15 10
10
10 10
Radial Deviation
15
25 20 30
30 20 15
10
20 15
Ulnar Deviation
58
63 90 120
51 73 67
67
71 70
Grip Strength (% of Opposite Side)
Fair
Good Excellent Excellent
Poor Good Good
Good
Excellent Fair
Overall Rating According to Fernandez
Complications
—
— — —
— — —
— Superficial wound infection at iliac crest incision —
Table 2. Operative and Post-operative Data for Ten Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius
206 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 207
Table 3. Rating of the Functional Result According to Fernandez11 Radiocarpal pain No pain Mild Moderate Severe Distal radioulnar pain Severe Moderate Mild No pain Wrist flexion and extension Arc of 130° to 140° Arc of 100° to 129° Arc of 80° to 99° ⬍80° arc Forearm rotation Arc of 160° to 180° Arc of 140° to 159° Arc of 120° to 139° ⬍120° arc Grip strength (compared with opposite limb) ⱖ80% 65% to 79% 40% to 64% ⬍40%
4 points 2 points 1 point 0 points 0 points 1 point 2 points 4 points 4 points 3 points 2 points 1 point 4 points 3 points 2 points 1 point 4 points 3 points 2 points 1 point
Overall rating: 18 to 20 points, excellent; 15 to 17 points, good; 12 to 14 points, fair; ⬍11 points, poor.
dez11 in the analysis of the treatment of malunited fractures of the distal radius was used to assess the overall functional result (Table 3). This is one of the most commonly used rating systems used to evaluate reconstructive surgery of the distal radius. The final follow-up radiographs were evaluated for the presence of posttraumatic arthrosis according to the criteria of Knirk and Jupiter.12 Grade 0 is defined
as no arthrosis, grade 1 is slight joint space narrowing, grade 2 is marked joint space narrowing and osteophyte formation, and grade 3 is complete absence of a joint space with osteophytes and cysts.
Results The patients were monitored for an average of 3 years 6 months (range, 2 years to 8 years 6 months). Fracture union was noted in each patient within 3 months of the initial procedure. According to the rating system of Fernandez11 there were 3 excellent, 4 good, 2 fair, and 1 poor result. An average of 55° wrist extension (range, 25° to 70°) and 50° flexion (range, 30° to 70°) were restored. Radial and ulnar deviation averaged 15° (range, 10° to 20°) and 20° (range, 15° to 30°), respectively. The average pronation and supination was 75° (range, 60° to 90°) and 70° (range, 40° to 90°), respectively. Postoperative grip strength averaged 73% percent of the opposite limb (range, 51% to 120%). The postoperative alignment of the distal radial articular surface was measured on posteroanterior and lateral radiographs (Table 4). Ulnar inclination of the distal radial articular surface averaged 22° (range, 15° to 25°). Palmar tilt averaged 5° (range, 0° to 20°). Ulnar variance could not be measured in the 4 patients who had partial or complete resection of the distal ulna. Ulnar variance in the remaining 6 patients averaged 0 mm (range, -2–2 mm). Radiographic evidence of arthrosis was graded according to the scale of Knirk and Jupiter12 as follows: grade 0 in 4 patients, grade 1 in 5 patients, and grade 2 in 1 patient. All 3 of the unsatisfactory results were related to problems associated with the initial injury and subsequent treatment course. One patient had severe
Table 4. Radiographic Data From 10 Patients With Delayed Union or Nonunion of a Fracture of the Distal Radius Case No.
Pre/Postoperative Ulnar Inclination (°)
Pre/Postoperative Palmar Tilt (°)
1 2 3 4 5 6 7 8 9 10
25/25 20/25 ⫺5/15 10/20 10/18 ⫺5/18 5/20 0/25 0/24 25/25
⫺25/0 ⫺10/5 45/20 ⫺5/10 5/0 0/0 ⫺5/0 ⫺10/0 20/0 ⫺10/10
Pre/Postoperative Ulnar Variance (mm) 2/2 1/Bowers 5/Darrach 1/Bowers 3/0 2/Bowers 0/1 ⫺2/⫺2 4/0 ⫺1/0
Arthrosis: Knirk/ Jupiter Grade 0 0 1 1 0 2 1 1 1 0
208 Fernandez, Ring, and Jupiter / Nonunion of the Distal Radius
stiffness after prolonged immobilization in an external fixator and then a cast. She regained 55° flexion and extension through the radiocarpal articulation and 120° forearm rotation, but still had mild pain. Another patient had an initially unrecognized carpal ligament injury contributing to a carpal instability that limited motion and grip strength and caused pain. The third unsatisfactory result was due to persistent discomfort in a patient whose initial fracture was the result of high-energy trauma. There were few complications due to the surgical treatment in this series. One obese patient had a superficial infection in the wound used to access the iliac crest. This was treated with debridement and parenteral antibiotics followed by oral antibiotics. Three patients requested plate removal (2 volar plates and 1 dorsal plate), but none had symptoms related to the plate.
Discussion Although healing problems at the distal radius have historically been considered rare,13–16 they have been discussed more frequently during the last decade.2–5,17 Some investigators have speculated that delay in healing of a fracture of the distal radius may have become more common since the advent of surgical techniques that restore the length of the fractured distal radius and thus create a bony gap in the metaphysis that may delay healing.1,4 Healing problems in the distal radius also seem to be related to unstable situations, such as concomitant fracture of the distal radius and ulna, as well as to an inadequate period of immobilization.1–5 Patients with delayed union or nonunion of the distal radius may have dysfunction related to pain and instability, but many problems arise from the associated malalignment. The treatment therefore resembles the treatment for malunion of the distal radius except that the surgeon must also address the delayed or arrested attempts at healing. The definition of delayed union or nonunion is straightforward in the presence of an atrophic nonunion, particularly a synovial pseudoarthrosis. Patients with these types of nonunions typically present with severe dysfunction related to deformity, instability, and pain. Nonoperative treatment is appropriate only in patients with limited functional demands (such as the invalid elderly) in whom symptoms can sometimes be managed in a splint. In the case of a stable fibrous nonunion, the indication for surgery relates more to dysfunction associated with malalign-
ment of the fracture fragments in an active person. Further immobilization to encourage fracture healing would only prolong and exacerbate this dysfunction and would not address the malalignment; such care would be appropriate only in patients with limited functional demands. In active patients, intervention is indicated as soon as dysfunction and limited healing become apparent. A metaphyseal fracture in a healthy patient should demonstrate signs of progressive healing within 3 months on standard radiographs. Patients with limited radiographic evidence of healing 4 months after the injury should be considered to have delayed healing The variety of techniques used in this series reflects the evolution of surgical techniques and implants during the study period, which was long by virtue of the relative rarity of this problem. This variation also reflects the need to address specific characteristics of the deformity with specific techniques. For instance, a dorsal approach and implant are used when there is dorsal displacement or angulation of the distal fragment or when a dorsal exposure is needed to remove loose implants. A volar exposure is used for volarly directed displacement or loose volar implants. If the alignment is relatively neutral, then a volar exposure is usually preferred because volar implants are less likely in our experience to irritate or damage the overlying tendons. Due to the rarity of this problem it is difficult to make treatment recommendations based on data analysis. It appears that a variety of implants and implant combinations can be successful provided that stability is achieved. Our current preferred technique, as reflected in the treatment of 1 of the more recent patients in the series (Fig. 1), reflects our interest in newer implants providing fixed-angle fixation and fixation in orthogonal planes, both of which can be more secure in poor-quality bone. Internal fixation should focus on the radial styloid as this portion of the distal fragment usually offers a greater amount of bone for the application of fixation devices. One plate can be applied directly on the radial aspect of the distal radius and a second on the volar or dorsal aspect. We offer additional recommendations based on our experience with these patients, although they cannot be supported statistically due to the rarity of this problem. In the presence of severe malalignment of the fracture it may be necessary to lengthen or divide the brachioradialis tendon to facilitate realignment. When the wrist is stiff, particularly in the presence of a synovial nonunion, incision of the
The Journal of Hand Surgery / Vol. 26A No. 2 March 2001 209
dorsal radiocarpal and/or volar distal radioulnar joint capsule should be considered. Arthrosis, deformity, or incongruity of the distal radioulnar joint must be addressed. Based on our experience we favor resection of the distal ulna (Darrach procedure), particularly when there is severe shortening of the distal radius that cannot be overcome. While it can be argued that some of the fractures in our series might have healed with additional immobilization, surgery performed as soon as delayed union or nonunion and malalignment are recognized addresses the radiocarpal and radioulnar joint dysfunction caused by the deformity and limits the need for prolonged external immobilization, which might exacerbate stiffness. The dysfunction associated with malalignment is the primary indication for surgical reconstruction while the delayed healing is the factor that increases the complexity of the problem and the difficulty of the procedure. The results of attempts to gain union of ununited fractures of the distal radius are not entirely predictable, as demonstrated by the fact that 3 of our patients had unsatisfactory results. Patients with such complex problems often present with associated concerns that can degrade the outcome. On the other hand, in each of our patients, surgical realignment of the distal fragment substantially enhanced radiocarpal and radioulnar mobility as well as carpal kinematics, as reflected by an increase in grip strength. Although the results in 3 patients were rated unsatisfactory using the strict scale of Fernandez,11 the goals of surgery (ie, a stable, healed, well-aligned wrist with preservation of some radiocarpal and midcarpal motion) were met after a single procedure in each patient and substantial functional improvement was observed. In other words, although patients and surgeons should expect some limitation of motion and incomplete relief of pain (occasionally substantial), our experience supports performing surgery in an attempt to gain union of fractures of the distal radius associated with delayed union or nonunion rather than proceeding directly to wrist arthrodesis in all cases. Wrist motion facilitates positioning of the hand in space, enhances strength and precision of the digits, and provides important proprioceptive feedback. It is therefore preferable to preserve even a small amount
of wrist motion. Our experience suggests that most ununited fractures of the distal radius are amenable to attempts to realign and heal the fracture.
References 1. Weber SC, Szabo RM. Severely comminuted distal radial fractures as an unsolved problem: complications associated with external fixation and pins and plaster techniques. J Hand Surg 1986;11A:157–165. 2. McKee MD, Waddell JP, Yoo D, Richards RR. Nonunion of distal radial fractures associated with distal ulnar shaft fractures: a report of four cases. J Orthop Trauma 1997; 11:49 –53. 3. Harper WM, Jones JM. Non-union of Colles’ fracture: report of two cases. J Hand Surg 1990;15B:121–123. 4. Segalman KA, Clark GL. Un-united fractures of the distal radius: a report of 12 cases. J Hand Surg 1998;23A:914 – 919. 5. Smith VA, Wright TW. Nonunion of the distal radius. J Hand Surg 1999;24B:601– 603. 6. Fernandez DL, Jupiter JB. Fractures of the distal radius: a practical approach to management. New York: SpringerVerlag, 1996. 7. Mu¨ller ME, Nazarian S, Koch P, Schatzker J. The comprehensive classification of fractures of long bones. Berlin: Springer-Verlag, 1990. 8. Henry AK. The upper limb and neck. In: Extensile exposure. 2nd ed. Edinburgh: Churchill Livingstone, 1973; 100 –107. 9. Ring D, Jupiter JB, Brennwald J, Bu¨chler U, Hastings H II. Prospective multicenter trial of a plate for dorsal fixation of distal radius fractures. J Hand Surg 1997;22A:777–784. 10. Bowers WH. Distal radioulnar joint arthroplasty: the hemiresection-interposition technique. J Hand Surg 1985; 10A:169 –178. 11. Fernandez DL. Radial osteotomy and Bowers arthroplasty for malunited fractures of the distal end of the radius. J Bone Joint Surg 1988;70A:1538 –1551. 12. Knirk JL, Jupiter JB. Intra-articular fractures of the distal end of the radius in young adults. J Bone Joint Surg 1986;68A:647– 659. 13. Bacorn RW, Kurtzke JF. A study of two thousand cases from the New York State Workmen’s Compensation Board. J Bone Joint Surg 1953;38A:643– 658. 14. Hamada G. Extra-articular graft for non-union in Colles’ fracture. J Bone Joint Surg 1944;26:833– 835. 15. Lachman JW. Report of the Meeting of the Orthopaedic Associations of the English-speaking world. J Bone Joint Surg 1958;40B:595–599. 16. Watson-Jones R. Injuries of the wrist. In: Fractures and other bone and joint injuries. 3rd ed. Baltimore: Williams & Wilkins, 1943;538 –545. 17. Szabo RM, Weber SC. Comminuted intraarticular fractures of the distal radius. Clin Orthop 1988;230:39 – 48.