Intra-Articular Fractures of the Distal Radius: A Prospective Randomised Controlled Trial Comparing Static Bridging and Dynamic Non-Bridging External Fixation

INTRODUCTION

The management of unstable intra-articular fractures of the distal radius remains controversial. One method of treatment is external fixation, most commonly with static fixators that bridge the wrist and immobilize this joint as well as the fracture. Immobilization of the wrist has been associated with significant morbidity (Agee, 1993; Hertel and Ballmer, 1994; Kaempffe et al., 1993), whereas early mobilization leads to earlier restoration of movement, earlier return of function and fewer complications (Dias et al., 1987; Jenkins et al., 1987; McAuliffe et al., 1987). As a consequence, dynamic external fixators were developed in the hope that the benefits of early mobilization could be achieved whilst using a fixator. These fixators bridge the wrist but have a hinge or other device at the level of the wrist joint that allows movement. Recently, non-bridging or radial–radial fixators have gained popularity. Several authors have reported their experience using this type of fixator (Bishay et al., 1994; Fischer et al., 1999; Forgon and Mammel, 1981; Jenkins et al., 1987; Kohut, 1995; Krishnan et al., 1998; McQueen, 1999; McQueen et al., 1999; Melendez et al., 1989). However, few studies have compared the outcomes of static and dynamic external fixators in the management of distal radial fractures (McQueen et al., 1996; McQueen 1998; Sommerkamp et al., 1994). The aims of this study were to compare the clinical, functional and radiological outcomes of a static bridging external fixator (Hoffmann II Compact frame) with a dynamic non-bridging external fixator that allowed early mobilization (Delta frame) in the treatment of intra-articular fractures of the distal radius. The Delta frame is a new AO configuration, which has produced favourable outcomes for Frykmann VII and VIII distal radial fractures (Krishnan et al., 1998).

METHODS

Between March 1997 and April 2000, 60 patients with intra-articular fractures of the distal radius, including complex comminuted fractures, were enrolled in the study. Patients were excluded if they were unable to give consent, suffered from any musculoskeletal or neurological disease, were unable to follow routine pin track care, had any other associated fractures of the hand, wrist or forearm or had a previous fracture of the same wrist. The study was approved by the Flinders Medical Centre Ethics Committee.

After obtaining consent, patients were randomly allocated by a closed envelope method into one of two groups of 30 patients. Patients in one group were treated with an AO Delta frame (Fig 1) and patients in the other were treated with a Hoffmann II Compact frame (Fig 2).

An orthopaedic surgeon performed all the operations. All the fractures were reduced by closed manipulation aided by 5 kg of horizontal finger-trap traction. Fluoroscopy was used in all cases. The Delta frame was constructed by placing four 2.5 mm self-tapping pins into the distal radial fragments in two horizontal planes. Two pins were inserted in the dorso-radial aspect of the distal radius and two in its dorso-ulnar aspect. These pins transfixed the fracture fragments and supported the articular surface. The pins mainly functioned as subarticular supports in the many cases with severe comminution and osteoporosis. A 4 mm threaded pin was then inserted into the shaft of the radius approximately 6 cm proximal to the fracture and the frame was assembled using three connecting rods to produce a triangular-shaped construct which did not cross the wrist joint. The Hoffmann frame consisted of two self-tapping pins in the distal radial shaft proximal to the fracture, and two similar pins in the second metacarpal with one or two connecting rods between them.

All pins were inserted after incision and open dissection of the soft tissues down to bone to protect neurovascular structures and tendons. Pin sites were dressed with betadine-soaked swabs. Intravenous cephalosporin (cephazolin, 1 g) was given intra-operatively and two further doses were given post-operatively, followed by a week of oral antibiotics (cephadine, 250 mg q.i.d.). A palmar plaster of Paris slab was applied for the first week and all external fixators were removed at 6 weeks in the outpatient department.

Instructions were given to all patients regarding finger, elbow and shoulder mobilization exercises. Those with a Delta frame were also instructed in wrist mobilization exercises commencing 2 weeks post-operatively. Patients with a Hoffmann frame did not commence wrist mobilization exercises until 6 weeks, when the frame was removed. Physiotherapy was prescribed for both groups.

An orthopaedic surgeon reviewed all patients post-operatively at 1, 6, 12, 26 and 52 weeks and any complications were recorded. Wrist and hand function was assessed by a research physiotherapist weekly for the first 6 weeks, then at 12, 26 and 52 weeks following surgery. Assessments included measurements of pain using a 0–10 visual analogue scale, range of movement using a goniometer and grip strength using a Jamar dynamometer. Grip strength was expressed as a percentage of the uninjured side. An activity of daily living questionnaire was also used to assess the performance of 17 different unilateral and bilateral functional activities. These activities are listed in Table 1.

Standard antero-posterior and lateral radiographs were taken of the wrist pre-operatively and at 1, 6, 26 and 52 weeks post-operatively. An independent radiologist evaluated all pre- and post-operative radiographs. Pre-operative fracture classifications and post-operative measurements of palmar tilt, radial inclination, radial length and radial step were recorded. The results were compared with population-based reference ranges (Krishnan, 2002), as no radiographs of the unaffected wrist were taken for comparison.

Differences between the two groups for pain, range of movement, grip strength and ability to perform activities of daily living were assessed using the Mann–Whitney U test at each post-operative time frame. Median values for each of these measures were calculated for each group.

Radiological differences between the two groups were assessed from the last set of radiographs of each patient using the Mann–Whitney U test. Median values for measurements of palmar tilt, radial inclination, radial length and radial step at 12 months were calculated for each group.

RESULTS

Forty-one women and 19 men were included in the study. The average age was 56 (range of 18–83) years. Patient details and fracture classification are summarized in Table 2. The two groups were similar for age and fracture type but there were more women in the Hoffmann group and more dominant arms in the Delta group. Forty-eight of the 50 fractures were intra-articular (AO type B2-C3), with all except one being complex intra-articular types (AO type C). The majority of patients had fractures with two or three main intra-articular fragments but ones with multifragmented intra-articular fragments were not excluded: in these cases the pins did not fix the fragments but acted as subarticular support for the fragments.

DISCUSSION

Results from previous studies comparing static and dynamic external fixators for the treatment of distal radial fractures have been conflicting. In the study by Sommerkamp et al. (1994), dynamic fixation using a bridging fixator (Clyburn, 1987) was compared with static fixation using a bridging AO fixator in 50 unstable distal radial fractures. The dynamic fixator group showed no greater range of movement either at the time the fixator was removed, or any other time up until the final 1-year assessment. These findings are similar to those of the current study. In the same study (Sommerkamp et al., 1994), the dynamic fixator group showed a significantly greater loss of radial length and more complications when compared with the static group. Using a modified Gartland and Werley Score (Gartland and Werley, 1951), only 76% of the dynamic fixator group achieved an excellent or good result compared to 92% of the static fixator group. These findings differ from those of the current study in which no radiological differences were found between the two groups.

In the study by McQueen et al. (1996), four different methods of treating re-displaced unstable distal radial fractures were compared. These were closed reduction and a forearm cast; open reduction with bone grafting, a Kirschner wire and a forearm cast; closed reduction and a bridging Pennig external fixator with the ball joint locked; and closed reduction and a bridging Pennig fixator with the ball joint released. Displaced intra-articular fractures were specifically excluded from the study. There was no difference in the radiological results between the static and dynamic Pennig bridging external fixator groups. Functional results at 6 weeks, 3 and 6 months and 1 year post-operatively also showed no difference between any of the groups. These findings are similar to those of the current study.

In a subsequent study, McQueen (1998) compared a static bridging Pennig fixator with a dynamic non-bridging Pennig fixator for the treatment of redisplaced unstable fractures of the distal radius. Again, displaced intra-articular fractures were excluded. There were no significant differences in outcomes at the later assessments, but the radiological results were significantly better in the dynamic non-bridging group. McQueen (1998) suggested that “non-bridging external fixation was the treatment of choice for unstable fractures of the distal radius which have sufficient space for the placement of pins in the distal fragment.” The difference in the results of McQueen’s (1998) and the current study may be due to the inclusion of comminuted fractures in the current study. The presence of patients with complex comminuted fractures in this study is also reflected in the wide ranges seen for palmar tilt and radial length. Although the degree of reduction obtained for some of these patients did not fall within the limits defined as acceptable by Earnshaw et al. (2002), the reductions were considered satisfactory as function was restored.

The rate of pin site infection in the current study was similar for both the Delta and Hoffmann frame groups. Almost one-third of patients developed an infection despite the use of prophylactic antibiotics. This rate of infection is high but reflects our low threshold for treating pin site problems. Patients with pin site erythema, with or without positive bacterial culture, were considered to have an infection and were treated with antibiotics. In all but two cases, the infections were successfully treated with oral antibiotics. There were three cases of extensor pollicis longus rupture, all of which occurred in the Delta frame group. In the study by McQueen (1998) the pattern of rupture was similar, with two extensor pollicis longus ruptures in 60 cases, both occurring in the dynamic non-bridging fixator group. Although rupture of this tendon is well documented following distal radius fracture, especially undisplaced or minimally displaced ones (Helal et al., 1982; Hirasawa et al., 1990; Hove, 1994), this problem appears to be a specific complication of non-bridging fixation.

In conclusion, this study demonstrated that the outcomes of patients with complex unstable intra-articular fractures of the distal radius are similar, regardless of whether they are treated with a static bridging external fixator or a dynamic non-bridging external fixator.