Abstract
We prospectively measured hand and wrist function in rheumatoid patients undergoing excision of the distal ulna. Range of motion, visual analogue pain scores and grip strength were measured in 22 wrists, and the Jebsen hand function test was administered to seven patients, preoperatively and at 3 and 12 months. At 1 year there were improvements in forearm pronation (P = 0.04), supination (P = 0.03) and wrist extension (P = 0.02), but a reduction in flexion (P = 0.009). Active radial deviation was reduced and ulnar deviation increased. There was a significant improvement in grip strength (P = 0.05) and reduction in wrist pain (P = < 0.0001). At 1 year the Jebsen hand function test showed improvements in simulated feeding, stacking checkers, and lifting large empty cans. Excision of the distal ulna in rheumatoid patients results in an improvement in some aspects of hand function.
INTRODUCTION
Resection of the distal ulna has been used to treat disorders of the distal radio-ulnar joint in patients with rheumatoid arthritis (Fraser et al., 1999; Gainor and Schaberg, 1985; Jackson et al., 1974; Leslie et al., 1990; Nanchahal et al., 1996; Newman, 1987; Rana et al., 1973; Rasker et al., 1980; Thirupathi et al., 1983). Whilst most previous studies have commented on improvements in pain, power and range of movement, only a few have attempted to measure functional outcome following surgery (Jackson et al., 1974; Newman, 1987; Rana et al., 1973; Rasker et al., 1980). Jackson et al. (1974), who graded pre-operative function on a scale of I (best) to IV (worst) and graded postoperative function as “much better”, “same” or “much worse”, found that 75% of patients were functionally “much better” following surgery. Rana et al. (1973) assessed function using tests that required rotation of the forearm and grip, for example turning a door knob and using a knife and fork. They demonstrated that 93% of wrists showed a functional improvement. Rasker et al. (1980) classified function on a 4-point scale ranging from “undisturbed” to “no use” and they too found a significant improvement in function following surgery. In contrast Newman (1987) found that only 25% of patients in his series had an improvement in upper limb function, with 60% remaining unchanged. His assessment of function involved five tests: turning a tap and doorknob, opening a jar lid, pouring a kettle and carrying a shopping bag.
The Jebsen hand function test (Jebsen et al., 1969) has been shown to be a useful and reliable measure for assessing hand function in patients with rheumatoid arthritis (Jones et al., 1991; Sharma et al., 1994; Vliet Vlieland et al., 1996). We have undertaken a prospective study to objectively measure functional outcome using the Jebsen hand function test, visual analogue pain scale, grip strength and range of motion as outcome measures.
PATIENTS AND METHODS
We prospectively studied 20 patients with rheumatoid arthritis who underwent excision of the distal ulna on 23 wrists, between November 1997 and January 2002. Range of motion data were collected in 22 of these wrists, and Jebsen hand function was measured in both hands of seven patients (December 2000 onwards). All patients had been optimized on medication prior to surgery and had well-controlled disease. The mean age of the patients was 58 years (range 28–80 years) with a male to female ratio of 1:4. The mean time from disease onset to time of surgery was 16 years (range 4–48 years).
Operative procedure
All operations were carried out by the senior author (J.N.) under general anaesthesia, using a tourniquet. A straight dorsal skin incision was made and the extensor retinaculum was reflected along the sixth dorsal compartment. An extensor tenosynovectomy was undertaken in all patients and a segment of the terminal branch of the posterior interosseous nerve was always resected. The distal ulna was excised just proximal to the distal radioulnar joint. Synovectomy of the radiocarpal joint was performed and Lister’s tubercle excised in all cases before the capsule was closed. The extensor carpi ulnaris tendon was relocated to its dorsal position with a radially based sling of extensor retinaculum and the remaining retinaculum was sutured deep to the tendons, with an overlying strip to prevent bowstringing. Patients with a tendency to radial rotation of the wrist on active extension preoperatively also underwent an extensor carpi radialis longus (ECRL) to extensor carpi ulnaris (ECU) tendon transfer (six wrists). A closed suction drain was inserted if there was excessive bleeding from bone ends and a palmar plaster of Paris splint and padded bandage was applied. If present, the drain was removed at 24 h and patients commenced active pronation and supination exercises on the next day. The plaster splint was replaced with a thermoplastic splint at 3 weeks, when active wrist flexion and extension were commenced.
Jebsen hand function
Hand function was measured preoperatively in both hands, using the Jebsen hand function test (Jebsen et al., 1969), by the same hand therapist (C.B.) in a group of 24 rheumatoid patients who were scheduled for excision of the distal ulna. Complete data were available at 1 year for seven of these patients. The remaining 17 patients were excluded because they had undergone further surgery on the operated hand, for example metacarpophalangeal joint arthroplasty or flexor synovectomy, or had undergone surgery on the other hand during the follow-up period: these procedures would also have influenced hand function. In addition, some patients who were initially scheduled only for excision of the distal ulna underwent additional surgery and were, therefore, also subsequently excluded from the study.
The Jebsen hand function test is an objective, standardized test that aims to be representative of the functions of the hand in daily living. It involves a timed series of seven tests: (1) writing a short sentence, (2) turning over cards, (3) picking up small objects and placing them in a container, (4) stacking checkers, (5) simulated eating, (6) moving large empty cans and (7) moving large weighted cans. The tests were carried out separately with both hands and the results recorded as the total time to perform each task in seconds for each hand.
Measures of pain, power and motion
All measurements were performed on the operated hand by the same hand therapist (C.B.) pre-operatively and at 3 and 12 months postoperatively. Early 3-month data were not obtained for three patients.
Pain was measured using a visual analogue scale (1–10) pre-operatively and at 1 year postoperatively. Grip strength was measured using a Jamar dynamometer in 19 patients (20 wrists). Flexion, extension, pronation, supination, and active ulnar and radial deviation were measured using a goniometer.
Statistical analysis
The mean percentage changes in hand function at 12 months, compared to pre-operative values, for both the operated and non-operated hand were compared using the Wilcoxon signed rank test. Range of motion and strength at 3 and 12 months follow-up were compared to preoperative measurements using the paired Student’s t-test, as data on three patients was unavailable at 3 months post-surgery. Pain scores at 12 months post-surgery were compared to preoperative measurements using the paired Student’s t-test.
RESULTS
Jebsen hand function
The results of the Jebsen hand function test in both the operated and non-operated hand are summarized in Table 1. There was an overall greater percentage improvement in writing, simulated feeding, checkers, and moving empty and weighted cans at 12 months in the operated hand compared to the non-operated hand. In particular, tests of simulated feeding showed a threefold greater mean improvement in the operated hand, staking checkers showed an almost 14-fold mean improvement and lifting empty cans showed an almost three-fold mean improvement compared to the non-operated hand at 12 months. Tests of writing improved by a mean of 17% (SD 13) in the operated hand compared to a mean of 12% (SD 14) in the non-operated hand at 12 months and lifting full cans improved by 22% (SD 24) in the operated hand compared to 18% (SD 24) in the non-operated hand at 12 months. Tests of card turning and picking up small objects had improved by almost exactly the same amount in both hands at 1 year.
Range of motion
Range of motion was measured in a group of 19 rheumatoid patients (22 wrists) pre-operatively and at 3 and 12 months postoperatively. Three patients were unavailable for measurement at 3 months following surgery and thus the data are presented for a group of 19 wrists at this time point.
Pronation and supination (Fig 1)
Pre-operatively, the mean forearm pronation was 80° (SD 14°), and this showed a small improvement at 3 months (mean 83°; SD 10°; P = 0.03). The improvement was maintained at 1 year (mean 85°; SD 10°; P = 0.04), although this modest change is unlikely to be of clinical significance. The mean forearm supination prior to surgery was 69° (SD 24°) and this had not improved at 3 months (mean 73°; SD 17°; P = 0.28). At 1 year there was a statistically significant improvement in supination compared to the pre-operative measurements (mean 82°; SD 13°; P = 0.03).
Flexion and extension (Fig 1)
Prior to surgery the mean wrist flexion was 43° (SD 14°). Three months following surgery, there was an almost 50% reduction in wrist flexion (mean 23°; SD 11°; P = 0.0004). Although flexion did improve at 1 year (mean 33°; SD 15°), it was still significantly reduced compared to the pre-operative measurements (P = 0.009). In contrast, wrist extension was increased at 3 months (mean 34°; SD 17°; P = 0.051) compared to pre-operative measurements (mean 27°; SD 27°). This improvement was maintained at 1 year (mean 37°; SD18 °; P = 0.02).
Ulnar and radial deviation (Fig 1)
Mean active ulnar deviation of the wrist prior to surgery was 16° (SD 9°) and this was unchanged at 3 months (mean 15°; SD 7°, P = 0.94). There was a slight increase in active ulnar deviation at 1 year, mean 19° (SD 8°), but this was not significant (P = 0.15). Active radial deviation at 3 months (mean 8°; SD 7°) was significantly less (P = 0.03) than pre-operatively (mean 12 °; SD 7°). However, 1 year following operation the mean radial deviation was 10° (SD 8°), which was not statistically different to preoperative measurements (P = 0.11).
Grip strength and pain
The mean pre-operative grip strength was 15 kg (SD 9) in a group of 19 patients (20 wrists). At 3 months, mean grip strength was reduced to 13 kg (SD 7) (P = 0.75) in the 17 wrists available for measurement. At 1 year grip strength had improved and was significantly greater than pre-operatively (mean 18 kg; SD 9); P = 0.049; n = 20).
The mean (SD) pre-operative pain score, as measured using a visual analogue scale, was 7 (SD 2). This reduced significantly to a mean of 2 (SD 2) at 1 year (P = < 0.0001).
Extensor carpi radialis longus to extensor carpi ulnaris transfer
Six patients also underwent ECRL to ECU transfer. In this subgroup, mean forearm pronation improved from 84° (SD 12°) pre-operatively to 89° (SD 2°) at 1 year. Forearm supination improved from 70° (SD 31°) to 83° (SD 10°), flexion decreased from 52° (SD 11°) to 42° (SD 17°), extension improved from 15° (SD 38°) to 34° (SD 19°). Active ulnar deviation remained the same at 19° (SD 11°) pre-operatively and 19° (SD 9°) at 1 year and active radial deviation decreased slightly from 13° (SD 8°) to 11° (SD 7°) at 1 year. Similarly power improved from 13 kg (SD 10 kg) pre-operatively to 17 kg (SD 9 kg) and visual analogue pain scores reduced from a mean of 9 (SD 2) pre-operatively to 3 (SD 2) at 1 year. Only one patient in this subgroup underwent Jebsen hand function testing and her results were consistent with the remainder of the group.
DISCUSSION
The aim of this study was to assess functional outcome following excision of the distal ulna in rheumatoid patients. Numerous assessments of hand function exist, although there is no clear consensus on which is most representative of hand function (Jarus and Poremba, 1993). The sequential occupational dexterity assessment (Van Lankveld et al., 1996), the Sollerman hand function test (Sollerman and Ejeskar, 1995) and the grip ability test (Dellhag and Bjelle, 1995) were considered unsuitable as all test bimanual dexterity. These tests are thus unable to measure the influence of surgery on each hand separately. We used the Jebsen hand function test as a measure of hand function as it tests unilateral hand function, has been shown to have good test–retest reliability (Jebsen et al., 1969) and is useful in evaluating the rheumatoid hand (Sharma et al., 1994). Hammond and Freeman (1997) studied the relationship between the Jebsen hand function test and the arthritis impact measurement scale 2 (a self-report measure of upper limb daily living function) in a group of 40 rheumatoid patients. They found that the Jebsen hand function test correlated well with all sub-tests of upper limb function, except for writing.
Our results show an improvement in five of the seven components of the Jebsen hand function test at 1 year following surgery. Rheumatoid arthritis is characterized by periods of relapse and remission, and is progressive in nature. However, as the Jebsen hand function test measures hand function in both hands, permitting assessment of changes in function in the operated and the non-operated hand, it enabled us to differentiate between the effects of surgery and changes in hand function caused by variations in disease activity. The times taken to complete all seven tasks by the non-operated hand at 3 months and 12 months were almost identical, but interestingly were less than the baseline preoperative measurements. This is probably because the patients had become familiar with the test (the practice effect) and is another important reason for assessing both hands (Vliet Vlieland et al., 1996).
All seven patients who underwent functional testing had surgery on the wrist which was most painful. Preoperative measurements for the majority of the function tests, except for writing and picking up small objects (for which times were similar), were thus better for the un-operated hand prior to surgery. Six out of the seven patients had surgery on their dominant hand, and this may explain the shorter writing times measured in the affected hand, compared to the non-operated hand, pre-operatively. Our results show that, whilst the majority of tests took longer to complete with the operated hand prior to surgery, most had improved to a greater extent in this hand than in the non-operated hand at 1 year. Tests that showed the greatest improvement in function included simulated feeding, stacking checkers and lifting empty cans. The continued improvement in the operated hands for up to 1 year compared to the non-operated hands suggests that the benefits of surgery are maintained in the long term.
Many patients who were tested for hand function prior to surgery were subsequently excluded from the study, mainly because they had undergone further surgery on either hand in the follow-up period. This demonstrates the difficulty in obtaining long-term functional data on the effects of a particular operation in rheumatoid patients, and explains why we only had complete data on seven patients at 1 year. Although tests for statistical significance are difficult to interpret in this small group, many components of the Jebsen hand function test showed marked improvements in the operated hand.
We also assessed range of motion, grip strength and pain and our data confirm previous reports of decreased pain levels (Fraser et al., 1999; Jackson et al., 1974; Rana et al., 1973; Rasker et al., 1980; Thirupathi et al., 1983), improvements in pronation and supination (Rana et al., 1973; Rasker et al., 1980), decreases in active radial deviation and increases in active ulnar deviation (Fraser et al., 1999), increases in extension and decreases in flexion (Fraser et al., 1999) and improvements in grip strength (Fraser et al., 1999; Rana et al., 1973). Six patients (six wrists) in our study also underwent ECRL to ECU tendon transfer. Results from this subgroup of patients were analysed separately and showed almost exactly the same trends as the group as the whole. While it was not the aim of this study to examine the effectiveness of this tendon transfer, these results suggest that it does not detract from the benefits of extensor synovectomy and excision of the distal ulna.
Footnotes
Acknowledgements
Mr Abhilash Jain was funded by an Arthritis Research Campaign Clinical Research Fellowship (J0535). We would like to thank Kiran Nanchahal, London School of Tropical Hygiene and Medicine, for statistical advice.