- Email: [email protected]
Recurrent dorsal angulation of the distal radius fracture during dynamic external fixation
Recurrent Dorsal Angulation of the Distal Radius Fracture During Dynamic External Fixation Satoshi Kawaguchi,MD, PhD, Kazuzi Sawada, MD, Yuki Nabeta, MD, Mitsuru Hayakawa, MD, PhD, Muroran, Japan, Mitsuhiro Aoki, MD, PhD, Sapporo,Japan Thirty-three fractures of the distal radius treated with a dynamic external fixator (that allowed for wrist motion between 2 to 4 weeks after surgery) were analyzed, focusing on loss of fracture reduction during external fixation. Fractures with preoperative dorsal angulation greater than 20 ~ and those involving the distal radioulnar joint had a significantly larger loss of reduction of dorsal angulation (8.9 ~ and 6.9 ~ respectively) than fractures with less severe preoperative dorsal angulation or those with an intact distal radioulnar joint (3.0 ~ and 2.6 ~ respectively). In contrast, preoperative radial shortening (>2 mm) and involvement of the radiocarpal joint did not significantly increase the loss of dorsal angulation. Neither of the 2 dynamic external fixation systems studied consistently stabilized Colles' fractures with preoperative dorsal angulation of greater than 20 ~ or involvement of the distal radioulnar joint. (J Hand Surg 1998;23A:920-925. Copyright 9 1998 by the American Society for Surgery of the Hand.)
External fixators with articulation have gained some popularity for the treatment of intra-articular fractures, such as those in the distal radius ~-3 or the distal tibia. 4"5 These devices (dynamic external fixators) have been proposed to allow for early motion of the affected joint, leading to early return of joint mobility as well as to improved articular nutrition. However, results of recent studies have argued against such beneficial effects of mobilization with dynamic external fixators, especially in fractures of the distal end of the radius. Sommerkamp et al. 6 and
From the Departmentof OrthopedicSurgery,Nikko MemorialHospital, Muroran, Japan; and the Department of Orthopedic Surgery, Sapporo Medical UniversitySchool of Medicine, Sapporo,Japan. Received for publication May 7, 1997; accepted in revised form March 5, 1998-. No benefitsin any formhave been receivedor will be receivedfrom a commercialparty related directlyor indirectlyto the subject of this article. Reprint requests: Satoshi Kawaguchi, MD, PhD, Department of Orthopedic Surgery,Nikko Memorial Hospital, 1-5-13,Shintomi-cho, Muroran, 051, Japan. Copyright9 1998by the AmericanSocietyfor Surgeryof the Hand 0363-5023/98/23A05-002753.00/0 920
The Journal of Hand Surgery
McQueen et al. 7 in a randomized study found no improvement in mobility and function with early mobilization of the wrist using the Clyburn model or the Pennig model of dynamic fixator, respectively. Although Lennox et al. 8 emphasized the acceptable results of external fixation with the Clyburn device, they mobilized the wrist joint in only 4 of 20 patients during external fixation. Therefore, the validity and usefulness of such dynamic external fixation in the treatment of fractures of the distal radius remains controversial. An important role of external fixation is to keep the fracture stable in a reduced position until bone union. Thus, the loss of fracture reduction in the radius might be a serious problem for the dynamic external fixation. In this study, we reviewed patients with distal radius fracture who had been treated with the Clyburn or the Pennig dynamic external fixator followed by early postoperative mobilization of the wrist joint. By comparing the preoperative radiographic features of fracture and the extent of recurrence of fracture displacement after external fixation, we evaluated the consistency of the stabilizing effect
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998
of the dynamic external fixation for distal radius fractures.
Materials and Methods Patients From April 1991 to February 1995, 46 fractures of the distal radius in 45 patients were treated using the Clyburn model (Zimmer, Inc, Warsaw, IN) or the Pennig model (Orthofix, Srl, Italy) of articulated external fixator. Four fractures treated in conjunction with internal fixation and/or bone grafting and 7 fractures with improper postoperative reduction, as judged by either the residual dorsal angulation of greater than 10 ~ or the residual radial shortening of greater than 2 mm on postoperative radiographs, were excluded from the analysis. In addition, 1 fracture with the device removed at 2 weeks due to pin infection and 1 volar Barton type fracture were excluded. The remaining 33 Colles' fractures that had been treated with dynamic external fixation alone with acceptable initial fracture reduction (dorsal angulation of -< 10 ~ and radial shortening of -<2 mm) were included in this study. There were 3 men and 29 women with an average age of 62 years (range, 40-77 years). The patients included a woman who first sustained a fracture on 1 side and then the other side 5 months later. None of these patients had associated upper extremity injuries. Thirty fractures were caused by a simple fall, 2 fractures by a fall down stairs, and 1 by motor vehicle accident. The fractures as classified by Universal (modified Gartland) criteria, 9 Frykman's criteria, m and AO criteria I ~ are depicted in Table 1. A die-punch fragment was present in 2 fractures. The average length of time from injury to operation was 7.4 days (range, 3-16 days). The average follow-up period was 24 months (range, 12-41 months). The Clyburn device was applied in 10 fractures and the Pennig device in 23 fractures, depending on the availability of the devices and surgeon's preference.
Surgical Procedures and Mobilization The surgical procedures were performed by following the instruction manuals for the external fixator devices and the original scientific papersJ -3 Briefly, under axillary block or general anesthesia, 2 longitudinal skin incisions, 2 to 3 cm in length, were made over the radial aspect of the second metacarpal and the radius. Following adequate exposure of the
921
Table 1. Classification of the Fractures According to the Universal, 9 Frykman,l~ and AO 11 Criteria Classification~Fracture Type Universal I II III IV Frykman I II III IV V VI VII VIII AO A2 A3 B1
No. 6 4 15 8
1 7 3 5 1 1 8 7 1 9 1
c1
8
C2 C3
4 10
cortical surfaces, 2 half-pins were inserted in the second metacarpal bone and the shaft of the radius, respectively. Closed reduction was then made by traction and careful manipulation of the wrist under fluoroscopic control. The device was placed with the mobile joint aligned with the proximal end of the capitate, ie, the center of wrist rotation. Traction force was then applied to distract the radiocarpal joint at the extent adequately retaining reduced position of the fracture. Patients were allowed to institute both flexion and extension of the wrist by mobilizing the single ball joint of the Clyburn fixator or 1 of the double ball joints of the Penning fixator 2 to 4 weeks after surgery (2 weeks in 14 fractures, 3 weeks in 11 fractures, and 4 weeks in 8 fractures; average, 2.8 weeks) and the device was left in place from 5 to 8 weeks (average, 6.0 weeks). According to the learning curve of surgeons for both application of an external fixator and wrist motion exercise, the joint mobilization was started late for the initial fracture cases, then became earlier as experience increased.
Radiographic Evaluations Dorsal angulation and shortening of the radial articular surface were measured on lateral and anteroposterior radiographs by one of the authors
922
Kawaguchi et al. / Dynamic External Fixation for Colles' Fractures
(S.K.) using a protractor at the time of injury, after application of the external fixator, and after removal of the external fixator. The difference in the value of dorsal angulation and shortening of the radial articular surface from application to removal of the external fixator was regarded as recurrent dorsal angulation and recurrent radial shortening during external fixation, respectively. Since the application of an external fixator occasionally deteriorated on the lateral radiographs, we asked a hand surgeon (M.H.) to independently measure the radiographic parameters on the same lateral views. We then evaluated the consistency and accuracy of the measurements between that surgeon and ourselves. The recurrence of dorsal angulation and radial shortening was analyzed with respect to (1) timing of mobilization, (2) preoperative radiologic deformity (dorsal angulation and radial shortening), (3) articular involvement (radiocarpai joint and distal radioulnar joint), (4) presence of a distal ulnar fracture and/or die-punch fragment, and (5) type of fixator. Preoperative dorsal angulation was graded into 2 groups; -<20 ~ and greater than 20 ~. Preoperative radial shortening was also graded into 2 groups: -<2 m m and greater than 2 mm. Articular involvement and the presence of a distal ulna fracture were chosen as parameters based on Frykman's classification. 1~ Data were analyzed statistically by Student's t-test and p < .05 was considered significant. Data with the number of smaller than 5 cases were not subjected to statistical analysis.
Results Interobserver consistency (between S.K. and M.H.) of the measurements taken from the lateral views was the average difference in the value of dorsal angulation for 1.1 o (range, 1o to 7~ The data from both examiners showed consistent significance
in the extent of recurrence with respect to dorsal angulation of the radius and distal radioulnar involvement and the presence of a die-punch fragment, but not to other parameters examined. Therefore, the data obtained by the primary author (S.K.) was used in the following discussion. All but 1 of the fractures had positive dorsal angulation at the time of injury (average, 17.2 ~ __+ 14.4~ the average preoperative radial shortening was 2.7 _+ 2.3 mm. Dorsal angulation and radial shortening immediately after application of the external fixator was on average - 0 . 2 ~ __ 7.5 ~ and 0.6 - 1.3 mm, respectively. At this stage, reduction and stabilization of the fracture was acceptable in all cases. The average recurrence in dorsal angulation at the time of removal of the external fixator was 4.8 ~ _+ 5.7 ~ The average recurrence of radial shortening was 0.94 _ 0.8 mm. No fractures had recurrent radial shorting o f greater than 2 mm during extemal fixation. Thus, radial shortening was not included in the following analysis. Table 2 shows the relationship between the extent of recurrent dorsal angulation and the timing of joint mobilization. As shown, the value of recurrent dorsal angulation increased in parallel with early joint mobilization. Notably, fractures with the joint mobilized 2 or 3 weeks after surgery had an average recurrent dorsal angulation of greater than 5 ~ In contrast, all but 1 of the fractures with the joint mobilized at 4 weeks had recurrent dorsal angulation of -<2 ~ We then assessed several radiologic factors that potentially influence collapse of the fractures into recurrent dorsal angulation. As shown in Table 3, fractures with preoperative dorsal angulation -<20 ~ had an average recurrent dorsal angulation of 3.0 ~ whereas fractures with preoperative dorsal angulation greater than 20 ~ had a signifcantly larger recurrence (8.9~ p < .006). On the contrary, in fractures
Table 2. Relationship Between Recurrent Dorsal Angulation and Institution of Wrist Mobilization Dorsal Angulation, o ( +_SD)
Institution of Postoperative Joint Mobilization
No.
Preoperative
Postapplication *
Postremovalt
Recurrent$
At2wk At3wk At2and3wk At4wk
14 11 25 8
19 4- 15 16--- 17 18_+ 15 16--- 11
-3-+9 1_+7 -1_+8 3_+4
2-9 5--- 11 4_+ 10 4-4-4
6-6 5---6 6-+6 2_+3
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation.
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998
923
Table 3. Relationship Between Recurrent Dorsal Angulation and Preoperative Radiologic Deformity Dorsal Angulation, ~ ( -+SD) Preoperative Deformity Dorsal angulation --<20~ >20 ~ Radial shortening <--2mm >2mm
No.
Preoperative
Postapplication *
Postremovalt
Recurrent$
23 10
10--_8 34 _ 11
0-+7 - 1 -+ 8
2-+7 8 +_ 10
3-+4 w 9 -4- 7
17 16
16_+ 11 19_+ 18
-0-+8 -0_+7
5-+7 3-+ 10
5-+6 5-+5
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. w p < .006.
with radial shortening -<2 mm and those with radial shortening greater than 2 mm, there was no significant correlation between preoperative radial shortening and recurrence of dorsal angulation. The involvement o f the radiocarpal joint had no obvious effect on the recurrence of dorsal angulation compared with fractures with an intact radiocarpal joint (Table 4). In contrast, fractures that involved the distal radioulnar joint had a significantly larger recurrent dorsal angulation (6.9~ p < .03) compared with those with no distal radioulnar joint involvement (2.6~ Table 5 shows the effect of the presence of distal ulna fractures as well as die-punch fragments on the recurrence of dorsal angulation. The presence of a distal ulna fracture had an equivalent degree of recurrent dorsal angulation compared with fractures without a distal ulna fracture. Two fractures with die-punch fragments both showed recurrent dorsal
angulation of as large as 16 ~ and 22 ~ respectively. The recurrence of these 2 fractures occurred before institution of wrist mobilization. The average loss of reduction of dorsal angulation in 10 fractures treated with the Clyburn device was 5.4 ~ _+ 6.2 ~ and that in 23 fractures with the Pennig device was 4.6 ~ _+ 5.6 ~ an insignificant difference (Table 6). However, dorsal angulation at the time of removal of the Clyburn external fixator was 8.7 ~ -+ 8.7 ~ and at the time of removal of the Pennig device, 1.7 ~ _+ 7.9 ~ The fractures treated with the Clyburn device tended to show greater dorsal angulation.
Discussion The validity of early mobilization of the wrist joint after surgery using articulated external fixators has been controversial in the treatment of distal radius fractures. 1-3'6-s This controversy is due in part to the
Table 4. Relationship Between Recurrent Dorsal Angulation and Articular Involvement Dorsal Angulation, o ( -+SD) Articular Involvement Radiocarpal joint involvement No Yes Distal radioulnar joint involvement No Yes
No.
Preoperative
Postapplication*
Postremovalt
Recurrent5;
10 23
21_+ 16 16-+ 14
-2-+8 0-+7
1-+9 5-+8
4-+4 5-+6
16 17
18-+ 14 16-+15
-1_+7 0-+8
1 +8 6-+9
3-+3 7-+7
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. :~ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. w p < .03.
w
924
Kawaguchi et al. / Dynamic External Fixation for Colles' Fractures
Table 5. Relationship Between Recurrent Dorsal Angulation and Presence of Distal Ulnar Fracture and Die-Punch Fragment Dorsal Angulation, ~ ( + SD)
Distal ulnar fracture No Yes Die-punch fragment Yes
No.
Preoperative
Postapplication*
13 20
18• 17• 13
2---6 -2---8
2
23 • 11
-10 • 4
Postremoval?
7 • 10 3-7 10 • 1
Recurrent$
6---6 4• 19 • 4w
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. wRecurrence of dorsal angulation bad occurred before wrist mobilization was instituted.
complexity of multifocal analysis in those studies. In this study, we conducted a simple analysis, focusing on the relationship between recurrence of the fracture displacement during external fixation and preoperative radiographic features of the fracture as well as institution of the wrist mobilization. We found that neither the extent of preoperative radial shortening, involvement of the radiocarpal joint, nor type of fixator significantly influenced the recurrent dorsal angulation. In contrast, fractures with preoperative dorsal angulation greater than 20 ~ and those involving the distal radioulnar joint had significantly larger recurrence (8.9 ~ and 6.9 ~, respectively) than fractures with less severe initial angulation or those with intact distal radioulnar joint (3.0 ~ and 2.6 ~, respectively). These findings suggest that the "dynamic external fixation" system does not consistently stabilize all types of Colles' fracture, but rather that it has a limited stabilizing effect on certain fracture types, such as those with severe initial dorsal angulation and distal radioulnar joint involvement. In our present study, the value of recurrent dorsal angulation tended to increase in parallel with early
wrist mobilization (Table 2). Our data also suggest that institution of wrist mobilization should be delayed 4 weeks after surgery if the recurrence of dorsal angulation is to be diminished -<2 ~ Because of a lack of serial postoperative x-ray films, we could not determine the accurate time of the recurrence during application of the external fixation devices. Nevertheless, delay of stable fibrous union at the thin and comminuted dorsal radial cortex compared with the thick volar cortex may contribute to the progress of the recurrent dorsal angulation in early wrist mobilization. The average recurrence in dorsal angulation in this series of fractures was 4.8 ~ In this respect, Jenkins et a1.12 noted an average of only 0.1 ~ of recurrent dorsal angulation in 32 fractures treated by a static external fixator. In addition, in the study by Hutchinson et al., 13 the average recurrence was less than 3.0 ~ in 44 fractures with static external fixation. No studies with "dynamic external fixation" have described the degree of recurrent dorsal angulation, with the exception of a recent study by Dienst et a l i a showing 1~ of recurrent dorsal angulation in an average of 30
Table 6. Relationship Between Recurrent Dorsal Angulation and the External Fixator Used Dorsal Angulation, ~ (+_SD) Fixator
No.
Clyburn Pennig
10 23
Preoperative
24 • 15 14 • 13
Postapplication*
Postremovalt
Recurrent);
3+6 - 2 --- 8
8 -+ 9 2- 8
5+ 6 5• 6
Negative value indicates angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postrernoval dorsal angulation and postapplication dorsal angulation.
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998 925
fractures, although they allowed only flexion and used supplementary Kirschner wires in some cases. As Clyburn 1 stopped allowing extension of the wrist because of frequent recurrent collapse, it seems likely that early mobilization of the wrist with a dynamic external fixator becomes less beneficial for stabilization of the fracture compared with the application of static external fixator. In fact, variable motion was evident at the fracture site in cadaveric wrists by mobilization when the Pennig external fixator had been applied. 6 It should be noted that none of the fractures exhibited significant recurrence in radial shortening, suggesting that the skeletal traction yielded by an external fixator can reduce the compressive force generated by active wrist mobilization and proves to be effective against radial shortening. For 2 cases with a die-punch fragment, recurrence of dorsal angulation occurred before institution of the wrist mobilization. This indicates the necessity of supplemental fixation, such as insertion of Kirschner wires or plating. Aoki et al. 15 previously reported the results of a treatment protocol composed of open reduction, bone grafting, internal fixation, and external fixation followed by early postoperative mobilization of the wrist. In that study, only 1 of 10 fractures exhibited a loss of reduction despite significant comminution and extensive articular involvement in those fractures. We believe that dorsal bone grafting is an important factor in preventing the dorsal radial cortex from collapsing when the dynamic external fixator system is used. The authors thank Professor Sei-ichi Ishii, Chairmanof Orthopedic Surgery, Sapporo Medical University,for careful readingof this manuscript and useful comments and suggestions.
References 1. Clyburn TA. Dynamic external fixation for comminuted intra-articular fractures of the distal end of the radius. J Bone Joint Surg 1987;69A:248-254. 2. Pennig D, Gausepohl T. Extraarticular and transarticular external fixation with early motion in distal radius fractures and malunions. Orthop Surg Techniques 1995;9:51-65.
3. Pennig D, Gausepohl T. External fixation of the wrist. Injury 1996;27:1-15. 4. Saleh M, Shanahan MDG, Fern ED. Intra-articular fractures of the distal tibia: surgical management by limited internal fixation and articulated distraction. Injury 1993; 24:37-40. 5. Marsh JL, Bonar S, Nepota JV, Decoster TA, Hurwitz SR. Use of an articulated extemal fixator for fractures of the tibial plafond. J Bone Joint Surg 1995;77A: 1498-1509. 6. Sommerkamp TG, Seeman M, Silliman J, et al. Dynamic external fixation of unstable fractures of the distal part of the radius. J Bone Joint Surg 1994;76A:1149-1161. 7. McQueen MM, Hajducka C, Court-Brown CM. Redisplaced unstable fractures of the distal radius: a prospective randomised comparison of four methods of treatment. J Bone Joint Surg 1996;78B:404-409. 8. Lennox JD, Page BJ II, Mandell RM. Use of the Clyburn external fixator in fractures of the distal radius. J Trauma 1989;29:326-331. 9. Cooney WP III, Linscheid RL, Dobyns JH. Fractures and dislocations of the wrist. In: Rockwood CA Jr, Green DP, Bucholz RW, eds. Rockwood and Green's fractures in adults. 3rd ed. Philadelphia: JB Lippincott, 1991:563-678. t0. Frykman G. Fracture of the distal radius including sequelae-shoulder-hand-finger syndrome, disturbance in the distal radio-ulnar joint and impairment of nerve function: a clinical and experimental study. Acta Orthop Scand Suppl 1967;108:27-31. 11. MiJller ME, Nazarian S, Koch P, Schatzker J. Radius/ulna, distal segment. In: Miiller ME, Nazarian S, Koch P, Schatzker J, eds. The comprehensive classification of fractures of long bones. Berlin: Springer-Verlag, 1990:106115. 12. Jenkins NH, Jones DG, Johnson SR, Mintowt-Czyz WJ. External fixation of Colles' fractures: an anatomical study. J Hand Surg 1987;69B:207-211. 13. Hutchinson DT, Strenz GO, Cautilli RA. Pins and plaster vs external fixation in the treatment of unstable distal radial fractures: a randomized prospective study. J Hand Surg 1995 ;20B :365-372. 14. Dienst M, Wozasek GE, Seligson D. Dynamic external fixation for distal radius fractures. Clin Orthop 1997;338: 160-171. 15. Aoki M, Ishii S, Usui M, Chiba H, Ooyama N. Open reduction, bone grafting, and external fixation for comminuted intra-articular fractures of the distal radius: early postoperative motion of the wrist using the Clyburn dynamic external fixator. Orthop Int 1993; 1:136-141.
External fixators with articulation have gained some popularity for the treatment of intra-articular fractures, such as those in the distal radius ~-3 or the distal tibia. 4"5 These devices (dynamic external fixators) have been proposed to allow for early motion of the affected joint, leading to early return of joint mobility as well as to improved articular nutrition. However, results of recent studies have argued against such beneficial effects of mobilization with dynamic external fixators, especially in fractures of the distal end of the radius. Sommerkamp et al. 6 and
From the Departmentof OrthopedicSurgery,Nikko MemorialHospital, Muroran, Japan; and the Department of Orthopedic Surgery, Sapporo Medical UniversitySchool of Medicine, Sapporo,Japan. Received for publication May 7, 1997; accepted in revised form March 5, 1998-. No benefitsin any formhave been receivedor will be receivedfrom a commercialparty related directlyor indirectlyto the subject of this article. Reprint requests: Satoshi Kawaguchi, MD, PhD, Department of Orthopedic Surgery,Nikko Memorial Hospital, 1-5-13,Shintomi-cho, Muroran, 051, Japan. Copyright9 1998by the AmericanSocietyfor Surgeryof the Hand 0363-5023/98/23A05-002753.00/0 920
The Journal of Hand Surgery
McQueen et al. 7 in a randomized study found no improvement in mobility and function with early mobilization of the wrist using the Clyburn model or the Pennig model of dynamic fixator, respectively. Although Lennox et al. 8 emphasized the acceptable results of external fixation with the Clyburn device, they mobilized the wrist joint in only 4 of 20 patients during external fixation. Therefore, the validity and usefulness of such dynamic external fixation in the treatment of fractures of the distal radius remains controversial. An important role of external fixation is to keep the fracture stable in a reduced position until bone union. Thus, the loss of fracture reduction in the radius might be a serious problem for the dynamic external fixation. In this study, we reviewed patients with distal radius fracture who had been treated with the Clyburn or the Pennig dynamic external fixator followed by early postoperative mobilization of the wrist joint. By comparing the preoperative radiographic features of fracture and the extent of recurrence of fracture displacement after external fixation, we evaluated the consistency of the stabilizing effect
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998
of the dynamic external fixation for distal radius fractures.
Materials and Methods Patients From April 1991 to February 1995, 46 fractures of the distal radius in 45 patients were treated using the Clyburn model (Zimmer, Inc, Warsaw, IN) or the Pennig model (Orthofix, Srl, Italy) of articulated external fixator. Four fractures treated in conjunction with internal fixation and/or bone grafting and 7 fractures with improper postoperative reduction, as judged by either the residual dorsal angulation of greater than 10 ~ or the residual radial shortening of greater than 2 mm on postoperative radiographs, were excluded from the analysis. In addition, 1 fracture with the device removed at 2 weeks due to pin infection and 1 volar Barton type fracture were excluded. The remaining 33 Colles' fractures that had been treated with dynamic external fixation alone with acceptable initial fracture reduction (dorsal angulation of -< 10 ~ and radial shortening of -<2 mm) were included in this study. There were 3 men and 29 women with an average age of 62 years (range, 40-77 years). The patients included a woman who first sustained a fracture on 1 side and then the other side 5 months later. None of these patients had associated upper extremity injuries. Thirty fractures were caused by a simple fall, 2 fractures by a fall down stairs, and 1 by motor vehicle accident. The fractures as classified by Universal (modified Gartland) criteria, 9 Frykman's criteria, m and AO criteria I ~ are depicted in Table 1. A die-punch fragment was present in 2 fractures. The average length of time from injury to operation was 7.4 days (range, 3-16 days). The average follow-up period was 24 months (range, 12-41 months). The Clyburn device was applied in 10 fractures and the Pennig device in 23 fractures, depending on the availability of the devices and surgeon's preference.
Surgical Procedures and Mobilization The surgical procedures were performed by following the instruction manuals for the external fixator devices and the original scientific papersJ -3 Briefly, under axillary block or general anesthesia, 2 longitudinal skin incisions, 2 to 3 cm in length, were made over the radial aspect of the second metacarpal and the radius. Following adequate exposure of the
921
Table 1. Classification of the Fractures According to the Universal, 9 Frykman,l~ and AO 11 Criteria Classification~Fracture Type Universal I II III IV Frykman I II III IV V VI VII VIII AO A2 A3 B1
No. 6 4 15 8
1 7 3 5 1 1 8 7 1 9 1
c1
8
C2 C3
4 10
cortical surfaces, 2 half-pins were inserted in the second metacarpal bone and the shaft of the radius, respectively. Closed reduction was then made by traction and careful manipulation of the wrist under fluoroscopic control. The device was placed with the mobile joint aligned with the proximal end of the capitate, ie, the center of wrist rotation. Traction force was then applied to distract the radiocarpal joint at the extent adequately retaining reduced position of the fracture. Patients were allowed to institute both flexion and extension of the wrist by mobilizing the single ball joint of the Clyburn fixator or 1 of the double ball joints of the Penning fixator 2 to 4 weeks after surgery (2 weeks in 14 fractures, 3 weeks in 11 fractures, and 4 weeks in 8 fractures; average, 2.8 weeks) and the device was left in place from 5 to 8 weeks (average, 6.0 weeks). According to the learning curve of surgeons for both application of an external fixator and wrist motion exercise, the joint mobilization was started late for the initial fracture cases, then became earlier as experience increased.
Radiographic Evaluations Dorsal angulation and shortening of the radial articular surface were measured on lateral and anteroposterior radiographs by one of the authors
922
Kawaguchi et al. / Dynamic External Fixation for Colles' Fractures
(S.K.) using a protractor at the time of injury, after application of the external fixator, and after removal of the external fixator. The difference in the value of dorsal angulation and shortening of the radial articular surface from application to removal of the external fixator was regarded as recurrent dorsal angulation and recurrent radial shortening during external fixation, respectively. Since the application of an external fixator occasionally deteriorated on the lateral radiographs, we asked a hand surgeon (M.H.) to independently measure the radiographic parameters on the same lateral views. We then evaluated the consistency and accuracy of the measurements between that surgeon and ourselves. The recurrence of dorsal angulation and radial shortening was analyzed with respect to (1) timing of mobilization, (2) preoperative radiologic deformity (dorsal angulation and radial shortening), (3) articular involvement (radiocarpai joint and distal radioulnar joint), (4) presence of a distal ulnar fracture and/or die-punch fragment, and (5) type of fixator. Preoperative dorsal angulation was graded into 2 groups; -<20 ~ and greater than 20 ~. Preoperative radial shortening was also graded into 2 groups: -<2 m m and greater than 2 mm. Articular involvement and the presence of a distal ulna fracture were chosen as parameters based on Frykman's classification. 1~ Data were analyzed statistically by Student's t-test and p < .05 was considered significant. Data with the number of smaller than 5 cases were not subjected to statistical analysis.
Results Interobserver consistency (between S.K. and M.H.) of the measurements taken from the lateral views was the average difference in the value of dorsal angulation for 1.1 o (range, 1o to 7~ The data from both examiners showed consistent significance
in the extent of recurrence with respect to dorsal angulation of the radius and distal radioulnar involvement and the presence of a die-punch fragment, but not to other parameters examined. Therefore, the data obtained by the primary author (S.K.) was used in the following discussion. All but 1 of the fractures had positive dorsal angulation at the time of injury (average, 17.2 ~ __+ 14.4~ the average preoperative radial shortening was 2.7 _+ 2.3 mm. Dorsal angulation and radial shortening immediately after application of the external fixator was on average - 0 . 2 ~ __ 7.5 ~ and 0.6 - 1.3 mm, respectively. At this stage, reduction and stabilization of the fracture was acceptable in all cases. The average recurrence in dorsal angulation at the time of removal of the external fixator was 4.8 ~ _+ 5.7 ~ The average recurrence of radial shortening was 0.94 _ 0.8 mm. No fractures had recurrent radial shorting o f greater than 2 mm during extemal fixation. Thus, radial shortening was not included in the following analysis. Table 2 shows the relationship between the extent of recurrent dorsal angulation and the timing of joint mobilization. As shown, the value of recurrent dorsal angulation increased in parallel with early joint mobilization. Notably, fractures with the joint mobilized 2 or 3 weeks after surgery had an average recurrent dorsal angulation of greater than 5 ~ In contrast, all but 1 of the fractures with the joint mobilized at 4 weeks had recurrent dorsal angulation of -<2 ~ We then assessed several radiologic factors that potentially influence collapse of the fractures into recurrent dorsal angulation. As shown in Table 3, fractures with preoperative dorsal angulation -<20 ~ had an average recurrent dorsal angulation of 3.0 ~ whereas fractures with preoperative dorsal angulation greater than 20 ~ had a signifcantly larger recurrence (8.9~ p < .006). On the contrary, in fractures
Table 2. Relationship Between Recurrent Dorsal Angulation and Institution of Wrist Mobilization Dorsal Angulation, o ( +_SD)
Institution of Postoperative Joint Mobilization
No.
Preoperative
Postapplication *
Postremovalt
Recurrent$
At2wk At3wk At2and3wk At4wk
14 11 25 8
19 4- 15 16--- 17 18_+ 15 16--- 11
-3-+9 1_+7 -1_+8 3_+4
2-9 5--- 11 4_+ 10 4-4-4
6-6 5---6 6-+6 2_+3
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation.
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998
923
Table 3. Relationship Between Recurrent Dorsal Angulation and Preoperative Radiologic Deformity Dorsal Angulation, ~ ( -+SD) Preoperative Deformity Dorsal angulation --<20~ >20 ~ Radial shortening <--2mm >2mm
No.
Preoperative
Postapplication *
Postremovalt
Recurrent$
23 10
10--_8 34 _ 11
0-+7 - 1 -+ 8
2-+7 8 +_ 10
3-+4 w 9 -4- 7
17 16
16_+ 11 19_+ 18
-0-+8 -0_+7
5-+7 3-+ 10
5-+6 5-+5
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. w p < .006.
with radial shortening -<2 mm and those with radial shortening greater than 2 mm, there was no significant correlation between preoperative radial shortening and recurrence of dorsal angulation. The involvement o f the radiocarpal joint had no obvious effect on the recurrence of dorsal angulation compared with fractures with an intact radiocarpal joint (Table 4). In contrast, fractures that involved the distal radioulnar joint had a significantly larger recurrent dorsal angulation (6.9~ p < .03) compared with those with no distal radioulnar joint involvement (2.6~ Table 5 shows the effect of the presence of distal ulna fractures as well as die-punch fragments on the recurrence of dorsal angulation. The presence of a distal ulna fracture had an equivalent degree of recurrent dorsal angulation compared with fractures without a distal ulna fracture. Two fractures with die-punch fragments both showed recurrent dorsal
angulation of as large as 16 ~ and 22 ~ respectively. The recurrence of these 2 fractures occurred before institution of wrist mobilization. The average loss of reduction of dorsal angulation in 10 fractures treated with the Clyburn device was 5.4 ~ _+ 6.2 ~ and that in 23 fractures with the Pennig device was 4.6 ~ _+ 5.6 ~ an insignificant difference (Table 6). However, dorsal angulation at the time of removal of the Clyburn external fixator was 8.7 ~ -+ 8.7 ~ and at the time of removal of the Pennig device, 1.7 ~ _+ 7.9 ~ The fractures treated with the Clyburn device tended to show greater dorsal angulation.
Discussion The validity of early mobilization of the wrist joint after surgery using articulated external fixators has been controversial in the treatment of distal radius fractures. 1-3'6-s This controversy is due in part to the
Table 4. Relationship Between Recurrent Dorsal Angulation and Articular Involvement Dorsal Angulation, o ( -+SD) Articular Involvement Radiocarpal joint involvement No Yes Distal radioulnar joint involvement No Yes
No.
Preoperative
Postapplication*
Postremovalt
Recurrent5;
10 23
21_+ 16 16-+ 14
-2-+8 0-+7
1-+9 5-+8
4-+4 5-+6
16 17
18-+ 14 16-+15
-1_+7 0-+8
1 +8 6-+9
3-+3 7-+7
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. :~ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. w p < .03.
w
924
Kawaguchi et al. / Dynamic External Fixation for Colles' Fractures
Table 5. Relationship Between Recurrent Dorsal Angulation and Presence of Distal Ulnar Fracture and Die-Punch Fragment Dorsal Angulation, ~ ( + SD)
Distal ulnar fracture No Yes Die-punch fragment Yes
No.
Preoperative
Postapplication*
13 20
18• 17• 13
2---6 -2---8
2
23 • 11
-10 • 4
Postremoval?
7 • 10 3-7 10 • 1
Recurrent$
6---6 4• 19 • 4w
Negative values indicate angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postremoval dorsal angulation and postapplication dorsal angulation. wRecurrence of dorsal angulation bad occurred before wrist mobilization was instituted.
complexity of multifocal analysis in those studies. In this study, we conducted a simple analysis, focusing on the relationship between recurrence of the fracture displacement during external fixation and preoperative radiographic features of the fracture as well as institution of the wrist mobilization. We found that neither the extent of preoperative radial shortening, involvement of the radiocarpal joint, nor type of fixator significantly influenced the recurrent dorsal angulation. In contrast, fractures with preoperative dorsal angulation greater than 20 ~ and those involving the distal radioulnar joint had significantly larger recurrence (8.9 ~ and 6.9 ~, respectively) than fractures with less severe initial angulation or those with intact distal radioulnar joint (3.0 ~ and 2.6 ~, respectively). These findings suggest that the "dynamic external fixation" system does not consistently stabilize all types of Colles' fracture, but rather that it has a limited stabilizing effect on certain fracture types, such as those with severe initial dorsal angulation and distal radioulnar joint involvement. In our present study, the value of recurrent dorsal angulation tended to increase in parallel with early
wrist mobilization (Table 2). Our data also suggest that institution of wrist mobilization should be delayed 4 weeks after surgery if the recurrence of dorsal angulation is to be diminished -<2 ~ Because of a lack of serial postoperative x-ray films, we could not determine the accurate time of the recurrence during application of the external fixation devices. Nevertheless, delay of stable fibrous union at the thin and comminuted dorsal radial cortex compared with the thick volar cortex may contribute to the progress of the recurrent dorsal angulation in early wrist mobilization. The average recurrence in dorsal angulation in this series of fractures was 4.8 ~ In this respect, Jenkins et a1.12 noted an average of only 0.1 ~ of recurrent dorsal angulation in 32 fractures treated by a static external fixator. In addition, in the study by Hutchinson et al., 13 the average recurrence was less than 3.0 ~ in 44 fractures with static external fixation. No studies with "dynamic external fixation" have described the degree of recurrent dorsal angulation, with the exception of a recent study by Dienst et a l i a showing 1~ of recurrent dorsal angulation in an average of 30
Table 6. Relationship Between Recurrent Dorsal Angulation and the External Fixator Used Dorsal Angulation, ~ (+_SD) Fixator
No.
Clyburn Pennig
10 23
Preoperative
24 • 15 14 • 13
Postapplication*
Postremovalt
Recurrent);
3+6 - 2 --- 8
8 -+ 9 2- 8
5+ 6 5• 6
Negative value indicates angulation over the volar side. * Immediately after application of the external fixator. t Immediately after removal of the external fixator. $ The difference in the degree of dorsal angulation between postrernoval dorsal angulation and postapplication dorsal angulation.
The Journal of Hand Surgery / Vol. 23A No. 5 September 1998 925
fractures, although they allowed only flexion and used supplementary Kirschner wires in some cases. As Clyburn 1 stopped allowing extension of the wrist because of frequent recurrent collapse, it seems likely that early mobilization of the wrist with a dynamic external fixator becomes less beneficial for stabilization of the fracture compared with the application of static external fixator. In fact, variable motion was evident at the fracture site in cadaveric wrists by mobilization when the Pennig external fixator had been applied. 6 It should be noted that none of the fractures exhibited significant recurrence in radial shortening, suggesting that the skeletal traction yielded by an external fixator can reduce the compressive force generated by active wrist mobilization and proves to be effective against radial shortening. For 2 cases with a die-punch fragment, recurrence of dorsal angulation occurred before institution of the wrist mobilization. This indicates the necessity of supplemental fixation, such as insertion of Kirschner wires or plating. Aoki et al. 15 previously reported the results of a treatment protocol composed of open reduction, bone grafting, internal fixation, and external fixation followed by early postoperative mobilization of the wrist. In that study, only 1 of 10 fractures exhibited a loss of reduction despite significant comminution and extensive articular involvement in those fractures. We believe that dorsal bone grafting is an important factor in preventing the dorsal radial cortex from collapsing when the dynamic external fixator system is used. The authors thank Professor Sei-ichi Ishii, Chairmanof Orthopedic Surgery, Sapporo Medical University,for careful readingof this manuscript and useful comments and suggestions.
References 1. Clyburn TA. Dynamic external fixation for comminuted intra-articular fractures of the distal end of the radius. J Bone Joint Surg 1987;69A:248-254. 2. Pennig D, Gausepohl T. Extraarticular and transarticular external fixation with early motion in distal radius fractures and malunions. Orthop Surg Techniques 1995;9:51-65.
3. Pennig D, Gausepohl T. External fixation of the wrist. Injury 1996;27:1-15. 4. Saleh M, Shanahan MDG, Fern ED. Intra-articular fractures of the distal tibia: surgical management by limited internal fixation and articulated distraction. Injury 1993; 24:37-40. 5. Marsh JL, Bonar S, Nepota JV, Decoster TA, Hurwitz SR. Use of an articulated extemal fixator for fractures of the tibial plafond. J Bone Joint Surg 1995;77A: 1498-1509. 6. Sommerkamp TG, Seeman M, Silliman J, et al. Dynamic external fixation of unstable fractures of the distal part of the radius. J Bone Joint Surg 1994;76A:1149-1161. 7. McQueen MM, Hajducka C, Court-Brown CM. Redisplaced unstable fractures of the distal radius: a prospective randomised comparison of four methods of treatment. J Bone Joint Surg 1996;78B:404-409. 8. Lennox JD, Page BJ II, Mandell RM. Use of the Clyburn external fixator in fractures of the distal radius. J Trauma 1989;29:326-331. 9. Cooney WP III, Linscheid RL, Dobyns JH. Fractures and dislocations of the wrist. In: Rockwood CA Jr, Green DP, Bucholz RW, eds. Rockwood and Green's fractures in adults. 3rd ed. Philadelphia: JB Lippincott, 1991:563-678. t0. Frykman G. Fracture of the distal radius including sequelae-shoulder-hand-finger syndrome, disturbance in the distal radio-ulnar joint and impairment of nerve function: a clinical and experimental study. Acta Orthop Scand Suppl 1967;108:27-31. 11. MiJller ME, Nazarian S, Koch P, Schatzker J. Radius/ulna, distal segment. In: Miiller ME, Nazarian S, Koch P, Schatzker J, eds. The comprehensive classification of fractures of long bones. Berlin: Springer-Verlag, 1990:106115. 12. Jenkins NH, Jones DG, Johnson SR, Mintowt-Czyz WJ. External fixation of Colles' fractures: an anatomical study. J Hand Surg 1987;69B:207-211. 13. Hutchinson DT, Strenz GO, Cautilli RA. Pins and plaster vs external fixation in the treatment of unstable distal radial fractures: a randomized prospective study. J Hand Surg 1995 ;20B :365-372. 14. Dienst M, Wozasek GE, Seligson D. Dynamic external fixation for distal radius fractures. Clin Orthop 1997;338: 160-171. 15. Aoki M, Ishii S, Usui M, Chiba H, Ooyama N. Open reduction, bone grafting, and external fixation for comminuted intra-articular fractures of the distal radius: early postoperative motion of the wrist using the Clyburn dynamic external fixator. Orthop Int 1993; 1:136-141.