- Email: [email protected]
A comparison of different surgical techniques in treating degenerative arthrosis of the carpometacarpal joint of the thumb
A C O M P A R I S O N OF D I F F E R E N T S U R G I C A L T E C H N I Q U E S T R E A T I N G D E G E N E R A T I V E A R T H R O S I S OF T H E C A R P O M E T A C A R P A L J O I N T OF T H E T H U M B
IN
A retrospective study of 98 cases M . LANZETTA and G. FOUCHER
From SOS Main, Strasbourg, Franc e
Between March 1977 and December 1991, 98 surgical procedures on 85 patients were performed at S O S Main Strasbourg for osteoarthrosis of the carpometacarpal joint of the thumb. The mean age was 56 and 90% were female; 13 were operated on bilaterally. 40% had Swanson arthroplasties (group 1), 15% Ashworth-Blatt hemiarthroplasties (group 2), and 45% had soft tissue arthroplasties (group 3). 62 cases were reviewed at an average follow-up of 5 years. Normal thumb range of motion was obtained in all cases, regardless of the technique used. Complete pain relief was achieved in 77% of the cases in group 1, 37.5% in group 2 and 71% in group 3. 15% of group 1 and 50% of group 2 required surgical revision, either for displacement or fracture of the implants. No revision was necessary in group 3. One case of silicone synovitis requiring secondary surgery was noted 6 years after surgery, accounting for 2.9% of the total number of silicone implants reviewed, but radiological signs of silicone synovitis were much more common (56%). Proximal metacarpal migration in group 3 averaged 38% of the initial arthroplasty space, and was not related to the type of ligamentoplasty used, or presence or absence of an "anchovy". The migration increased to 68% of the space if an associated NIP joint arthrodesis was carried out at the same time. Complications included reflex sympathetic dystrophy (4% in group 1 and 14% in group 3). The Ashworth-Blatt hemiarthroplasty failed to gain satisfactory results, but both silicone arthroplasty and soft tissue arthroplasty proved to be useful procedures. However, due to the risk of synovitis the present treatment of choice is soft tissue arthroplasty.
Journal of Hand Surgery (British and European Volume, 1995) 20B: 1:105-110 Methods of surgical treatment for arthrosis of the carpometacarpal ( C M ) j o i n t of the thumb include arthrodesis (Carroll and Hill, 1973), trapeziectomy (Gervis, 1949), soft tissue arthroplasty (Froimson, 1970; Burton and Pellegrini, 1986), silicone arthroplasty (Swanson et al, 1981), partial implant arthroplasty (Ashworth et al, 1977) and total joint replacement (De La Caffini6re, 1991). Few studies have compared the results of different techniques. Amadio et al (1982) compared two series of patients operated on by two different surgeons, one performing soft tissue interposition arthroplasty and the other silicone implant arthroplasty. Burton and Pellegrini (1986) and PeUegrini and Burton (1986) published separate results of soft tissue interposition arthroplasty associated with ligamentoplasty and silicone arthroplasty. In view-of the persistent controversy surrounding the surgical management of this condition, we have reviewed our experience using three different surgical procedures over a 14-year-period.
first consultation was 57 years (range 38-82). There were 65 dominant and 33 non-dominant hands; 13 patients had bilateral procedures (Table 1). The majority of patients complained of pain and weakness interfering with professional activities and daily tasks. Progressive functional impairment and loss of dexterity were also frequently mentioned, and patients were concerned with appearance. Pain was especially related to the use of the hand for pinch and grasp, aggravated by writing and movements like opening jars, but in some cases it was pronounced during very light activities or even at rest. Clinical assessment evaluated range of motion of the thumb, tip pinch and grip strength and functional activities. The Jamar dynamometer was used to quantitate grip and tip pinch strength. Measures of grip and pinch strengths obtained before 1986 with the Martin vigorimeter were converted into Jamar measures by multiplying the original value by a converting factor (37.758 for grip strength and 25.667 for tip pinch strength). The converting factors were obtained by taking bilateral grip and tip pinch strength measurements in 184 adults of both sexes with both devices, and correlating the total mean values. All patients had plain X-rays, including routine views (PA, lateral, and obliques of the thumb), and a PA axial-oblique view so that all the articular surfaces of the trapezium could be visualized (Gruber, 1991). 34 patients had radiological signs of arthritis localized to the CM joint and all the others had more diffuse disease.
MATERIAL A N D M E T H O D S
98 hands of 85 patients with osteoarthrosis of the CM joint of the thumb were treated at SOS Main, Strasbourg between 1977 and 1991. Five of the 98 cases had posttraumatic arthritis, four following a Bennett's fracture and one a fracture of the trapezium. 77 patients were female and eight male. The average age at the time of 105
THE JOURNAL OF HAND SURGERY VOL. 20B No. 1 FEBRUARY 1995
106
Table 1--Total study population--characteristics Diagnosis
Osteoarthritis --primary --secondary
No. of patients'
85 80 5
No. of thumbs
98 93 5
Sex
Bilat.
13 13 --
M
F
9 7 2
89 86 3
Av. age (yrs)
R
L
R
ND
56.6 57.4 42.4
55 51 4
43 42 1
65 61 4
33 32 1
M: male; F: female; R: right; L: left; D: dominant; ND: non-dominant.
Note was made of associated first metacarpal adduction deformity (23% of the cases), and MP joint hyperextension (27.5% of the cases). Associated pathology was diagnosed in 21% of the cases, including carpal tunnel syndrome in 11 (11%), trigger finger or thumb in four (4%), FCR tendonitis in two cases, Dupuytren's disease, Raynaud's syndrome, thrombosis of the ulnar artery and de Quervain's disease in one case each. All patients were treated initially with conservative measures. These included a modification of hand activities, splintage of the basal joints of the thumb and non-steroidal anti-inflammatory medication in some cases. Steroid injections were not used. Failure to achieve relief, or rapid recurrence of symptoms, prompted surgical treatment. Silicone arthroplasty using the Swanson trapezium implant (group-I) was performed in 39 cases from 1977 to 1987; silicone hemiarthroplasty using the AshworthBlatt implant (Ashworth et al, 1977) or an implant carved from the Swanson implant (group 2) was carried out in 15 cases, from 1982 to 1984. From 1977 to 1991, 44 cases were treated by soft tissue arthroplasty (group 3). In each procedure a standard surgical technique was used. For Ashworth implants, the original technique of the authors (Ashworth et al, 1977) was followed. For the two other procedures, through a dorso-radial approach, protecting the superficial branches of the radial nerve and the radial artery, the trapezium was removed piecemeal, avoiding damage to the ligamentous and capsular structures. When performing a Swanson silicone arthroplasty, the trapezoid had to be partially resected to accommodate the implant in 34 cases (87%), and a K-wire was placed through the implant to increase stability in 31 cases (79.5%). The most common sizes of implants were 4 (16 cases) and 3 (12 cases). Some capsular tissue attached to the distal pole of the scaphoid was preserved; during capsular closure, a stitch was placed in this tissue in order to increase the stability of the base of the implant. A capsuloplasty by advancement and radialization of APL tendon was carried out in 19 cases (49%), and a strip of the same tendon served as a source for ligamentoplasty in three cases (8%). Soft tissue arthroplasty was carried out in 44 cases. In this group of patients, a ligamentoplasty using a strip of APL passed through the FCR tendon was performed
(Fig 1) in 34 cases (77%). The remainder of APL was folded and inserted distally and radially. The accessory muscle described by Zancolli (personal communication, 1989) was looked for and resected if present. If there was limited hyperextension of the MP joint, EPB was resected. Tendon interposition was carried out in 35 cases (79.5%), using mostly an "anchovy" of palmaris longus (or in absence of this tendon, the EPB or a strip of APL). The space was left void in nine cases (20.5%). Associated procedures are summarized in Table 2.
Fig 1
Technique of ligamentoplasty used in group 3 passing a strip of A P L through the distal part of the FCR.
Table 2--Association surgical procedures (98 operations, 13 bilateral) At thumb level M P joint K-wire temporary stabilization M P joint arthrodesis Web release Transverse K-wire between first and second metacarpal
8 7 8 3
Miscellaneous Carpal tunnel release De Quervain tendinitis Trigger finger
9 cases 4 cases 2 cases
cases cases cases cases
TECHNIQUESFOR DEGENERATIVE1ST CM ARTHROS1S
107
RESULTS Of the 85 patients (98 cases) treated, 52 patients (62 cases: group 1--26, group 2--eight, group 3--28) were available for review by an independant surgeon, at an average follow-up of 5 years (group 1--7 years, group 2--8.5 years and group 3--2.5 years). Complications are detailed in Table 3. Reflex sympathetic dystrophy (RSD), confirmed by technetium scan, occurred in five of the 62 procedures (8%), but one had long-term consequences related to this complication. Statistical analysis was not possible due to the small size of the group. 31% of the Swanson implants dislocated and 15% fractured, but only 11.5% need to be removed after about 6 years. The score was worse for the Ashworth-Blatt implant (7/8 displaced and the last one fractured); four had to be removed after an average of 3.5 years. Silicone synovitis was more frequent radiologicaly (56%) than clinically (3%); patients complained very rarely of unsatisfactory appearance of their thumb. 15% in group 1 and 50% in group 2 required surgical revision (Table 4). Excellent thumb range of motion (Kapandji grade 9 on a scale of 10) was obtained in all cases, regardless of the technique used. Only extension remained limited in 30% of the cases, especially in patients who had an associated MP joint arthrodesis. Pain relief, especially during da~ly or professional activities, was used as the primary criterion by which results were evaluated.
Complete constant pain relief was achieved in 77% of the cases in group 1, 37.5% of the cases in group 2 and 71% of the cases in group 3 (Table 5). Patients were asked to express the general improvement obtained by surgery and the improvement strictly related to strength on a scale ranging from 0 to 100%. Subjective total improvement compared to the preoperative pattern averaged 91% in group 1, 75% in group 2 and '89.5% in group 3. Strength was judged by the patients to have improved an average of 89.5% in group 1, 82.5% in group 2 and 73% in group 3. Radiographic views showed proximal metacarpal migration after soft tissue arthroplasty by an average of 38% of the initial radiologically measured distance between the distal pole of the scaphoid and the base of the metacarpal. This shifting was not related to the use of a ligamentoplasty. Similarly, no increase in the tendency to migrate proximally was noted when no tendon was interposed as an "anchovy" and the space left void. Proximal migration increased to 68% of the space if an associated MP joint arthrodesis was carried out at the same time (P<0.05). Post-operative grip strength in group 1 was almost equal to the pre-operative value (-1.1% differential), while tip pinch strength improved (+ 3.3% differential), with an overall increase of 1.1% (mean follow-up 7 years). Since in group 2, 50% (4 out of 8) of the reviewed implants were removed, it was not possible to compare
Table 3--Complications
RSD Superficial infection Implant subluxation Implant fracture Radiological silicone synovitis (cystic lesions) Clinical silicone synovitis "Shoulder" cosmetic deformity
Group 1
Group 2
Group 3
Total
1/26 (3.8%) 0
0/8
4/28 (14.2%) 1/28
5/62 (8.1%) 1/62
(3.5%0)
(1.5%)
8/26 (30.7%) 4/26 (15.4%) 17/26 (65.4%°) 1/26 (3.8%) 3/26 (11.5%)
7/8 (87.5%) 1/8 (12.5%) 3/8 (37.5%) 0/8
-
-
3/8 (37.5%)
4/28 (14.2%)
15/34 (44.1%) 5/34 (14.7%) 19/34 (55.9%) 1/34 (2.9%o) 10/62 (16.1%)
0
-
Table 4--Surgical revision. Original procedure
Number of revisions
Cause
Interval
Revision
Group 1
4/26 (15.4%)
Group 2
4/8 (50%)
Displaced 1 Fractured 2 Silicone synovitis 1 Displaced 3 Fractured 1
(7 yrs) (4.5 and 6 yrs) (6 yrs) (1.6 and 6 yrs) (1.5 yrs)
Ablation Ablation Ablation Trapeziectomy Trapeziectomy
Group 3
0/28 (0%)
108
THEJOURNALOF HANDSURGERYVOL.20BNo. 1 FEBRUARY1995 Table 7--Group 3 results. Pre-operative and post-operative grip and tip pinch strength determinations according to the technique used
Table 5--Clinical results: evaluation of pain
~o~
Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Total
1
20 (76.9%) 3 (11.5%) 1 (3.9%) 2 (7.7%) 26
Gro~ 2
~o~ 3
3 (37.5%) 2 (25%) 2 (25%) 1 (12.5%) 8
20 (71.4%) 5 (17.9%) 3 (10.7%) 28
Grip (kg)
Tip pinch (kg)
Average overall differential (%)
16.45 20.10
7.48 9.10
+ 21.91
23.83 24.00
10.50 11.66
+ 5.87
20.40 22.10
9.30 8.90
+ 6.32
Ligamentoplasty and interposition
Pre-op. Post-op. Ligamentoplasty-no interposition
Note: Grade 0=no pain. Grade 1= seasonalpain (i.e. weather). Grade 2=pain duringheavyactivity. Grade 3~pain duringlight activity. Grade4 = constantpain.
Pre-op. Post-op. Interposition-no ligamentoplasty
consistent data regarding pre- and post-operative longterm force measurements (Table 6). In group 3, there was an overall post-operative increase in force of 12.5%, with both grip strength and tip pinch strength exceeding pre-operative levels (+ 13.4% and + 11.6% respectively; mean follow-up 2.5 years). The difference was much more evident in patients who underwent ligamentoplasty combined with a tendon ("anchovy") interposition (+21.91%), compared with the patients who had ligamentoplasty but no tendon interposition (+ 5.87%), or tendon interposition but no ligamentoplasty (+6.32%; Table 7). The only overall (grip+pinch tip strengths) differential decrease, in group 3 (-2.19%), was obtained in the sub-group of patients who had no proximal migration of the first metacarpal, with an unchanged distance between the metacarpal base and the distal surface of the scaphoid at follow-up (P < 0.05). DISCUSSION The treatment of arthrosis of the basal joint of the thumb is still a subject of controversy. A course of conservative treatment should be carried out first' and is frequently effective in avoiding or at least delaying surgical treatment. When surgery is indicated, many techniques have been proposed, and although the majority of them relieve pain and improve function, uncertainty remains concerning which is the best surgical procedure. In the early stages of the disease, reconstruction of
Pre-op. Post-op,
the palmar oblique ligament, as suggested by Eaton et al (1984), may be successful in increasing the stability of the CM joint, which delays the progression of degenerative joint changes. When there is no laxity at an early stage, we favour denervation (Cozzi, 1991), which gives immediate and satisfactory relief from pain, and can be used as palliative treatment in association with conservative measures; we have not included these patients due to short follow-up. Arthrodesis of the trapeziometacarpal joint has become less popular because of its many drawbacks. Following arthrodesis, the mobility remains restricted, with no ability to flatten the hand; any patient with a stiff MP joint, a zig-zag collapse deformity, an adduction deformity of the metacarpal or a pan-arthritis is not a candidate for this operation. Furthermore, arthrodesis usually requires a prolonged period of immobilization, and is technically demanding with a risk of non-union (5 to 50%). Even the benefit in terms of strength has been questioned by Amadio and De Silva (1990) in their study of different procedures in men. Limited resection of the trapezium and arthroplasty of the trapeziometacarpal joint require careful preoperative assessment of all other trapezial articular surfaces, to rule out pan-trapezial arthrosis. Swanson and Ashworth have separately developed silicone implants for limited arthroplasty. When arthritis is more
Table 6--Results. Pre-operative and post-operative grip and tip pinch strength derminations (kg)
Grip
Differential %
Tip pinch
Differential %
Overall differential
22.82 22.57
- 1.1
9.64 9.96
+3.3
+1.1
18.78 21.30
+ 13.4
8.48 9.47
+11.6
+12.5
Group 1
Pre-op. Post-op. Group 3
Pre-op Post-op.
TECHNIQUES FOR DEGENERATIVE 1ST CM ARTHROSIS
extended, bigger implants are inserted after trapeziectomy (Swanson et al, 1981; Poppen and Niebauer, 1978). After satisfactory early experience (Swanson et al, 1981; Ashworth et al, 1977), subluxation and wear have frequently been mentioned. Swanson et al (1981) reported an incidence of radial subluxation of the implants in 14.5%, dislocation in 6% and a "purposeful radial articulation" in 5.5% of the cases, with secondary surgery in 3.3%. According to Pellegrini and Burton (1986) wear and cold flow were more pronounced in more stable implants; this can lead to silicone synovitis. Ashworth et al (1977) reviewing 49 implants (burr hole type), had only two revisions for material failure and no progression of arthritis. Allieu et al (1990), in a long-term follow-up (9.8 years) of 22 Swanson implants found subluxation in 68%, radiological signs of wear in 70%, fractures in 14%, and 31% had radiological signs of synovitis. De La Caffinibre has developed cemented total prostheses providing good, early function, but long-term loosening frequently occurs (five out of 13 after 12 years; De La Caffini~re, 1991). When a total trapeziectomy is performed, the space created can be left void (Gervis, 1973; Burton and Pellegrini, 1986). Total trapeziectomy is simple and improves web contracture, but has been criticized for shortening the thumb, decreasing the strength and providing instability. Filling the space with soft tissue (Froimson, 1970) and adding a ligamentoplasty (Burton and Pellegrini, 1986) have been proposed to avoid these shortcomings. The majority of published papers are devoted to one technique and few compare different techniques. Our retrospective study of three groups of patients treated by Swanson implants (group 1), Ashworth implants (group 2) and trapeziectomy (group 3) has some features of bias. The surgeon was the same but patients were treated at different stages of disease and follow-up is not identical for all groups. Statistical analysis was not always possible due to small samples. Good to excellent pain relief was observed in all groups. Objective strength was superior in group 3 mainly when ligamentoplasty and soft tissue interposition were combined (+ 22%). The same group has no revision operation but an increased rate of RSD. This complication is frequently not even mentioned in publications; others have, found an incidence from 4% (Burton and Pellegrini, 1986; Amadio and De Silva, 1990; Allieu et al, 1990) to 19% (Leviet et al, 1990); this also depends on the criteria used and none of our patients had any long-term problems. In the two other groups the rate of failure of implants was worrying: 31% of Swanson spacers and seven out of eight Ashworth implants were subluxed, 15% of Swanson implants fractured, and half of the Ashworth prostheses were removed. Silicone synovitis was more a radiological problem (56%) than a clinical one (3%). However, all patients must be fully informed of these potential complications, should be followed regularly
109
and advised to seek immediate treatment in case of pain, swelling or stiffness. Among the comparative studies already mentioned, Amadio et al (1982) compared 25 silicone implants with 25 trapeziectomies performed by two surgeons; there were little or no differences in terms of pain relief, strength and motion. Burton and Pellegrini (1986) gave preference to trapeziectomy and ligam e n t o p l a s t y compared to silicone arthroplasty (Pellegrini arid Burton, 1986) with better strength (+ 19% compared to - 4 % ) , better stability, less shortening (11% compared to 25%) and less revision procedure (0 compared to 35%). From our data it appears that our ligamentoplasty and tendon interposition has not prevented proximal metacarpal shifting, which occurs progressively during the first 3 years, but this shortening did not decrease the strength, either grip or tip pinch. On the contrary, the only overall differential decrease in strength was observed in the subgroup of patients who did not show any proximal metacarpal migration. Finally, more recently, Conolly and Lanzetta (1993) compared 16 arthrodeses, 15 trapeziectomies and 53 Swanson arthroplasties, with similar good results (69%, 80%, 73.5%) and rates of complications (31%, 33%, 24.5%). In conclusion, trapeziectomy with ligamentoplasty compares favourably with silicone arthroplasty and is free of such long-term complications as instability, wear, and worrying signs of silicone synovitis. References ALLIEU, Y., LUSSIEZ, B. and MARTIN, B. (t990). Rtsultats ~t long terme de l'implant de Swanson darts le traitement de ta rhizarthrose. Revue de Chirurgie Orthop+dique, 76: 437-441. AMADIO, P. C. and DE SILVA, S. P. (1990). Comparison of the results of trapeziometacarpal arthrodesis and arthroplasty in men with osteoarthritis of the trapeziometacarpal joint. Annals of Hand Surgery, 9: 5: 358-363. AMADIO, P. C , MILLENDER, L. H. and SMITH, R. J. (1982). Silicone spacer or tendon spacer for trapezium resection arthroplasty: Comparison of results. Journal of Hand Surgery, 7: 3: 237-244. ASHWORTH, C. R., BLATT, G., CHUINARD, R. G. and STARK, H. H. (1977). Silicone-rubber interposition arthroplasty of the carpometacarpal joint of the thumb. Journal of Hand Surgery. 2: 5: 345-357. BURTON, R. I. and PELLEGRINI, V. D. (1986). Surgical management of basal joint arthritis of the thumb: Part II: Ligament reconstruction with tendon interposition arthroplasty. Journal of Hand Surgery, l l A : 3: 324 332. CARROLL, R. E. and HILL, N. A. ( t 973). Arthrodesis of the carpo-metaearpat joint of the thumb. Journal of Bone and Joint Surgery, 55B: 2: 292-294. CONOLLY, W. B. and LANZETTA, M. (1993). Surgical management of arthritis of the carpo-metacarpal joint of the thumb. Australian and New Zealand Journal of Surgery, 63: 596-603. COZZI, E. P. D6nervation des Articulations du Poignet et de la Main. Tubiana R. ( Ed ) Trait6 de Chirurgie de la Main. Vol 4. Paris, Masson, 1991: 781-787. DE LA CAFFINII~RE, J. Y. (1991). Rtsultats/t long terme de la proth~se totale traptzo-mttacarpienne dans la rhizarthrose. Revue de Chirurgie Orthoptdique, 77: 312-321. EATON, R. G., LANE, L. B., LITTLER, J. W. and KEYSER, J. J. (1984). Ligament reconstruction for the painful thumb carpometaearpal joint: A long-term assessment. Journal of Hand Surgery, 9A: 5: 692-699. FROIMSON, A. I. (1970) Tendon arthroplasty of the trapeziometacarpal joint. Clinical Orthopaedics and Related Research. 70: 191-199. GERVIS, W. H. (1949). Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. Journal of Bone and Joint Surgery, 31B: 4: 537-539. GERVIS, W. H. (1973). A review of excision of the trapezium for osteoarthritis of the trapezio-metacarpal joint after 25 years. Journal of Bone and Joint Surgery, 55B: 1:56 57.
110 GRUBER, L. (1991). Practical approaches to obtaining hand radiographs and special techniques in hand radiology. Hand Clinics. 7: 1: 1-20. LE VIET, D., KERBOULL, L. and EBELIN, M. (1990). Int6r6t de la trap6zectomie darts le traitement de l'arthrose p6ritrap6zienne. Revue de Chirurgie Orthop~dique. 76: 158-161. PELLEGRINI, V. D. and BURTON, R. 1. (1986). Surgical management of basal joint arthritis of the thumb: Part I: Long-term results of silicone implant arthroplasty. Journal of Hand Surgery, 11A: 3: 309-324. POPPEN, N. K. and NIEBAUER, J. J. (1978). "Tie-in" trapezium prosthesis: long-term results. Journal of Hand Surgery, 3: 5: 445-450.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 1 FEBRUARY 1995 SWANSON, A. B., SWANSON, G., de G. and WATERMEIER, J. J. (1981). Trapezium implant arthroplasty: Long term evaluation of 150 cases. Journal of Hand Surgery, 6: 2: 125-141.
Accepted: 25 May 1994 Guy Foucher, MD, Head, SOS Main Strasbourg, 4 Bd du Pr6sident Edwards, 67000 Strasbourg, France. © 1995 The British Society for Surgery of the Hand
IN
A retrospective study of 98 cases M . LANZETTA and G. FOUCHER
From SOS Main, Strasbourg, Franc e
Between March 1977 and December 1991, 98 surgical procedures on 85 patients were performed at S O S Main Strasbourg for osteoarthrosis of the carpometacarpal joint of the thumb. The mean age was 56 and 90% were female; 13 were operated on bilaterally. 40% had Swanson arthroplasties (group 1), 15% Ashworth-Blatt hemiarthroplasties (group 2), and 45% had soft tissue arthroplasties (group 3). 62 cases were reviewed at an average follow-up of 5 years. Normal thumb range of motion was obtained in all cases, regardless of the technique used. Complete pain relief was achieved in 77% of the cases in group 1, 37.5% in group 2 and 71% in group 3. 15% of group 1 and 50% of group 2 required surgical revision, either for displacement or fracture of the implants. No revision was necessary in group 3. One case of silicone synovitis requiring secondary surgery was noted 6 years after surgery, accounting for 2.9% of the total number of silicone implants reviewed, but radiological signs of silicone synovitis were much more common (56%). Proximal metacarpal migration in group 3 averaged 38% of the initial arthroplasty space, and was not related to the type of ligamentoplasty used, or presence or absence of an "anchovy". The migration increased to 68% of the space if an associated NIP joint arthrodesis was carried out at the same time. Complications included reflex sympathetic dystrophy (4% in group 1 and 14% in group 3). The Ashworth-Blatt hemiarthroplasty failed to gain satisfactory results, but both silicone arthroplasty and soft tissue arthroplasty proved to be useful procedures. However, due to the risk of synovitis the present treatment of choice is soft tissue arthroplasty.
Journal of Hand Surgery (British and European Volume, 1995) 20B: 1:105-110 Methods of surgical treatment for arthrosis of the carpometacarpal ( C M ) j o i n t of the thumb include arthrodesis (Carroll and Hill, 1973), trapeziectomy (Gervis, 1949), soft tissue arthroplasty (Froimson, 1970; Burton and Pellegrini, 1986), silicone arthroplasty (Swanson et al, 1981), partial implant arthroplasty (Ashworth et al, 1977) and total joint replacement (De La Caffini6re, 1991). Few studies have compared the results of different techniques. Amadio et al (1982) compared two series of patients operated on by two different surgeons, one performing soft tissue interposition arthroplasty and the other silicone implant arthroplasty. Burton and Pellegrini (1986) and PeUegrini and Burton (1986) published separate results of soft tissue interposition arthroplasty associated with ligamentoplasty and silicone arthroplasty. In view-of the persistent controversy surrounding the surgical management of this condition, we have reviewed our experience using three different surgical procedures over a 14-year-period.
first consultation was 57 years (range 38-82). There were 65 dominant and 33 non-dominant hands; 13 patients had bilateral procedures (Table 1). The majority of patients complained of pain and weakness interfering with professional activities and daily tasks. Progressive functional impairment and loss of dexterity were also frequently mentioned, and patients were concerned with appearance. Pain was especially related to the use of the hand for pinch and grasp, aggravated by writing and movements like opening jars, but in some cases it was pronounced during very light activities or even at rest. Clinical assessment evaluated range of motion of the thumb, tip pinch and grip strength and functional activities. The Jamar dynamometer was used to quantitate grip and tip pinch strength. Measures of grip and pinch strengths obtained before 1986 with the Martin vigorimeter were converted into Jamar measures by multiplying the original value by a converting factor (37.758 for grip strength and 25.667 for tip pinch strength). The converting factors were obtained by taking bilateral grip and tip pinch strength measurements in 184 adults of both sexes with both devices, and correlating the total mean values. All patients had plain X-rays, including routine views (PA, lateral, and obliques of the thumb), and a PA axial-oblique view so that all the articular surfaces of the trapezium could be visualized (Gruber, 1991). 34 patients had radiological signs of arthritis localized to the CM joint and all the others had more diffuse disease.
MATERIAL A N D M E T H O D S
98 hands of 85 patients with osteoarthrosis of the CM joint of the thumb were treated at SOS Main, Strasbourg between 1977 and 1991. Five of the 98 cases had posttraumatic arthritis, four following a Bennett's fracture and one a fracture of the trapezium. 77 patients were female and eight male. The average age at the time of 105
THE JOURNAL OF HAND SURGERY VOL. 20B No. 1 FEBRUARY 1995
106
Table 1--Total study population--characteristics Diagnosis
Osteoarthritis --primary --secondary
No. of patients'
85 80 5
No. of thumbs
98 93 5
Sex
Bilat.
13 13 --
M
F
9 7 2
89 86 3
Av. age (yrs)
R
L
R
ND
56.6 57.4 42.4
55 51 4
43 42 1
65 61 4
33 32 1
M: male; F: female; R: right; L: left; D: dominant; ND: non-dominant.
Note was made of associated first metacarpal adduction deformity (23% of the cases), and MP joint hyperextension (27.5% of the cases). Associated pathology was diagnosed in 21% of the cases, including carpal tunnel syndrome in 11 (11%), trigger finger or thumb in four (4%), FCR tendonitis in two cases, Dupuytren's disease, Raynaud's syndrome, thrombosis of the ulnar artery and de Quervain's disease in one case each. All patients were treated initially with conservative measures. These included a modification of hand activities, splintage of the basal joints of the thumb and non-steroidal anti-inflammatory medication in some cases. Steroid injections were not used. Failure to achieve relief, or rapid recurrence of symptoms, prompted surgical treatment. Silicone arthroplasty using the Swanson trapezium implant (group-I) was performed in 39 cases from 1977 to 1987; silicone hemiarthroplasty using the AshworthBlatt implant (Ashworth et al, 1977) or an implant carved from the Swanson implant (group 2) was carried out in 15 cases, from 1982 to 1984. From 1977 to 1991, 44 cases were treated by soft tissue arthroplasty (group 3). In each procedure a standard surgical technique was used. For Ashworth implants, the original technique of the authors (Ashworth et al, 1977) was followed. For the two other procedures, through a dorso-radial approach, protecting the superficial branches of the radial nerve and the radial artery, the trapezium was removed piecemeal, avoiding damage to the ligamentous and capsular structures. When performing a Swanson silicone arthroplasty, the trapezoid had to be partially resected to accommodate the implant in 34 cases (87%), and a K-wire was placed through the implant to increase stability in 31 cases (79.5%). The most common sizes of implants were 4 (16 cases) and 3 (12 cases). Some capsular tissue attached to the distal pole of the scaphoid was preserved; during capsular closure, a stitch was placed in this tissue in order to increase the stability of the base of the implant. A capsuloplasty by advancement and radialization of APL tendon was carried out in 19 cases (49%), and a strip of the same tendon served as a source for ligamentoplasty in three cases (8%). Soft tissue arthroplasty was carried out in 44 cases. In this group of patients, a ligamentoplasty using a strip of APL passed through the FCR tendon was performed
(Fig 1) in 34 cases (77%). The remainder of APL was folded and inserted distally and radially. The accessory muscle described by Zancolli (personal communication, 1989) was looked for and resected if present. If there was limited hyperextension of the MP joint, EPB was resected. Tendon interposition was carried out in 35 cases (79.5%), using mostly an "anchovy" of palmaris longus (or in absence of this tendon, the EPB or a strip of APL). The space was left void in nine cases (20.5%). Associated procedures are summarized in Table 2.
Fig 1
Technique of ligamentoplasty used in group 3 passing a strip of A P L through the distal part of the FCR.
Table 2--Association surgical procedures (98 operations, 13 bilateral) At thumb level M P joint K-wire temporary stabilization M P joint arthrodesis Web release Transverse K-wire between first and second metacarpal
8 7 8 3
Miscellaneous Carpal tunnel release De Quervain tendinitis Trigger finger
9 cases 4 cases 2 cases
cases cases cases cases
TECHNIQUESFOR DEGENERATIVE1ST CM ARTHROS1S
107
RESULTS Of the 85 patients (98 cases) treated, 52 patients (62 cases: group 1--26, group 2--eight, group 3--28) were available for review by an independant surgeon, at an average follow-up of 5 years (group 1--7 years, group 2--8.5 years and group 3--2.5 years). Complications are detailed in Table 3. Reflex sympathetic dystrophy (RSD), confirmed by technetium scan, occurred in five of the 62 procedures (8%), but one had long-term consequences related to this complication. Statistical analysis was not possible due to the small size of the group. 31% of the Swanson implants dislocated and 15% fractured, but only 11.5% need to be removed after about 6 years. The score was worse for the Ashworth-Blatt implant (7/8 displaced and the last one fractured); four had to be removed after an average of 3.5 years. Silicone synovitis was more frequent radiologicaly (56%) than clinically (3%); patients complained very rarely of unsatisfactory appearance of their thumb. 15% in group 1 and 50% in group 2 required surgical revision (Table 4). Excellent thumb range of motion (Kapandji grade 9 on a scale of 10) was obtained in all cases, regardless of the technique used. Only extension remained limited in 30% of the cases, especially in patients who had an associated MP joint arthrodesis. Pain relief, especially during da~ly or professional activities, was used as the primary criterion by which results were evaluated.
Complete constant pain relief was achieved in 77% of the cases in group 1, 37.5% of the cases in group 2 and 71% of the cases in group 3 (Table 5). Patients were asked to express the general improvement obtained by surgery and the improvement strictly related to strength on a scale ranging from 0 to 100%. Subjective total improvement compared to the preoperative pattern averaged 91% in group 1, 75% in group 2 and '89.5% in group 3. Strength was judged by the patients to have improved an average of 89.5% in group 1, 82.5% in group 2 and 73% in group 3. Radiographic views showed proximal metacarpal migration after soft tissue arthroplasty by an average of 38% of the initial radiologically measured distance between the distal pole of the scaphoid and the base of the metacarpal. This shifting was not related to the use of a ligamentoplasty. Similarly, no increase in the tendency to migrate proximally was noted when no tendon was interposed as an "anchovy" and the space left void. Proximal migration increased to 68% of the space if an associated MP joint arthrodesis was carried out at the same time (P<0.05). Post-operative grip strength in group 1 was almost equal to the pre-operative value (-1.1% differential), while tip pinch strength improved (+ 3.3% differential), with an overall increase of 1.1% (mean follow-up 7 years). Since in group 2, 50% (4 out of 8) of the reviewed implants were removed, it was not possible to compare
Table 3--Complications
RSD Superficial infection Implant subluxation Implant fracture Radiological silicone synovitis (cystic lesions) Clinical silicone synovitis "Shoulder" cosmetic deformity
Group 1
Group 2
Group 3
Total
1/26 (3.8%) 0
0/8
4/28 (14.2%) 1/28
5/62 (8.1%) 1/62
(3.5%0)
(1.5%)
8/26 (30.7%) 4/26 (15.4%) 17/26 (65.4%°) 1/26 (3.8%) 3/26 (11.5%)
7/8 (87.5%) 1/8 (12.5%) 3/8 (37.5%) 0/8
-
-
3/8 (37.5%)
4/28 (14.2%)
15/34 (44.1%) 5/34 (14.7%) 19/34 (55.9%) 1/34 (2.9%o) 10/62 (16.1%)
0
-
Table 4--Surgical revision. Original procedure
Number of revisions
Cause
Interval
Revision
Group 1
4/26 (15.4%)
Group 2
4/8 (50%)
Displaced 1 Fractured 2 Silicone synovitis 1 Displaced 3 Fractured 1
(7 yrs) (4.5 and 6 yrs) (6 yrs) (1.6 and 6 yrs) (1.5 yrs)
Ablation Ablation Ablation Trapeziectomy Trapeziectomy
Group 3
0/28 (0%)
108
THEJOURNALOF HANDSURGERYVOL.20BNo. 1 FEBRUARY1995 Table 7--Group 3 results. Pre-operative and post-operative grip and tip pinch strength determinations according to the technique used
Table 5--Clinical results: evaluation of pain
~o~
Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Total
1
20 (76.9%) 3 (11.5%) 1 (3.9%) 2 (7.7%) 26
Gro~ 2
~o~ 3
3 (37.5%) 2 (25%) 2 (25%) 1 (12.5%) 8
20 (71.4%) 5 (17.9%) 3 (10.7%) 28
Grip (kg)
Tip pinch (kg)
Average overall differential (%)
16.45 20.10
7.48 9.10
+ 21.91
23.83 24.00
10.50 11.66
+ 5.87
20.40 22.10
9.30 8.90
+ 6.32
Ligamentoplasty and interposition
Pre-op. Post-op. Ligamentoplasty-no interposition
Note: Grade 0=no pain. Grade 1= seasonalpain (i.e. weather). Grade 2=pain duringheavyactivity. Grade 3~pain duringlight activity. Grade4 = constantpain.
Pre-op. Post-op. Interposition-no ligamentoplasty
consistent data regarding pre- and post-operative longterm force measurements (Table 6). In group 3, there was an overall post-operative increase in force of 12.5%, with both grip strength and tip pinch strength exceeding pre-operative levels (+ 13.4% and + 11.6% respectively; mean follow-up 2.5 years). The difference was much more evident in patients who underwent ligamentoplasty combined with a tendon ("anchovy") interposition (+21.91%), compared with the patients who had ligamentoplasty but no tendon interposition (+ 5.87%), or tendon interposition but no ligamentoplasty (+6.32%; Table 7). The only overall (grip+pinch tip strengths) differential decrease, in group 3 (-2.19%), was obtained in the sub-group of patients who had no proximal migration of the first metacarpal, with an unchanged distance between the metacarpal base and the distal surface of the scaphoid at follow-up (P < 0.05). DISCUSSION The treatment of arthrosis of the basal joint of the thumb is still a subject of controversy. A course of conservative treatment should be carried out first' and is frequently effective in avoiding or at least delaying surgical treatment. When surgery is indicated, many techniques have been proposed, and although the majority of them relieve pain and improve function, uncertainty remains concerning which is the best surgical procedure. In the early stages of the disease, reconstruction of
Pre-op. Post-op,
the palmar oblique ligament, as suggested by Eaton et al (1984), may be successful in increasing the stability of the CM joint, which delays the progression of degenerative joint changes. When there is no laxity at an early stage, we favour denervation (Cozzi, 1991), which gives immediate and satisfactory relief from pain, and can be used as palliative treatment in association with conservative measures; we have not included these patients due to short follow-up. Arthrodesis of the trapeziometacarpal joint has become less popular because of its many drawbacks. Following arthrodesis, the mobility remains restricted, with no ability to flatten the hand; any patient with a stiff MP joint, a zig-zag collapse deformity, an adduction deformity of the metacarpal or a pan-arthritis is not a candidate for this operation. Furthermore, arthrodesis usually requires a prolonged period of immobilization, and is technically demanding with a risk of non-union (5 to 50%). Even the benefit in terms of strength has been questioned by Amadio and De Silva (1990) in their study of different procedures in men. Limited resection of the trapezium and arthroplasty of the trapeziometacarpal joint require careful preoperative assessment of all other trapezial articular surfaces, to rule out pan-trapezial arthrosis. Swanson and Ashworth have separately developed silicone implants for limited arthroplasty. When arthritis is more
Table 6--Results. Pre-operative and post-operative grip and tip pinch strength derminations (kg)
Grip
Differential %
Tip pinch
Differential %
Overall differential
22.82 22.57
- 1.1
9.64 9.96
+3.3
+1.1
18.78 21.30
+ 13.4
8.48 9.47
+11.6
+12.5
Group 1
Pre-op. Post-op. Group 3
Pre-op Post-op.
TECHNIQUES FOR DEGENERATIVE 1ST CM ARTHROSIS
extended, bigger implants are inserted after trapeziectomy (Swanson et al, 1981; Poppen and Niebauer, 1978). After satisfactory early experience (Swanson et al, 1981; Ashworth et al, 1977), subluxation and wear have frequently been mentioned. Swanson et al (1981) reported an incidence of radial subluxation of the implants in 14.5%, dislocation in 6% and a "purposeful radial articulation" in 5.5% of the cases, with secondary surgery in 3.3%. According to Pellegrini and Burton (1986) wear and cold flow were more pronounced in more stable implants; this can lead to silicone synovitis. Ashworth et al (1977) reviewing 49 implants (burr hole type), had only two revisions for material failure and no progression of arthritis. Allieu et al (1990), in a long-term follow-up (9.8 years) of 22 Swanson implants found subluxation in 68%, radiological signs of wear in 70%, fractures in 14%, and 31% had radiological signs of synovitis. De La Caffinibre has developed cemented total prostheses providing good, early function, but long-term loosening frequently occurs (five out of 13 after 12 years; De La Caffini~re, 1991). When a total trapeziectomy is performed, the space created can be left void (Gervis, 1973; Burton and Pellegrini, 1986). Total trapeziectomy is simple and improves web contracture, but has been criticized for shortening the thumb, decreasing the strength and providing instability. Filling the space with soft tissue (Froimson, 1970) and adding a ligamentoplasty (Burton and Pellegrini, 1986) have been proposed to avoid these shortcomings. The majority of published papers are devoted to one technique and few compare different techniques. Our retrospective study of three groups of patients treated by Swanson implants (group 1), Ashworth implants (group 2) and trapeziectomy (group 3) has some features of bias. The surgeon was the same but patients were treated at different stages of disease and follow-up is not identical for all groups. Statistical analysis was not always possible due to small samples. Good to excellent pain relief was observed in all groups. Objective strength was superior in group 3 mainly when ligamentoplasty and soft tissue interposition were combined (+ 22%). The same group has no revision operation but an increased rate of RSD. This complication is frequently not even mentioned in publications; others have, found an incidence from 4% (Burton and Pellegrini, 1986; Amadio and De Silva, 1990; Allieu et al, 1990) to 19% (Leviet et al, 1990); this also depends on the criteria used and none of our patients had any long-term problems. In the two other groups the rate of failure of implants was worrying: 31% of Swanson spacers and seven out of eight Ashworth implants were subluxed, 15% of Swanson implants fractured, and half of the Ashworth prostheses were removed. Silicone synovitis was more a radiological problem (56%) than a clinical one (3%). However, all patients must be fully informed of these potential complications, should be followed regularly
109
and advised to seek immediate treatment in case of pain, swelling or stiffness. Among the comparative studies already mentioned, Amadio et al (1982) compared 25 silicone implants with 25 trapeziectomies performed by two surgeons; there were little or no differences in terms of pain relief, strength and motion. Burton and Pellegrini (1986) gave preference to trapeziectomy and ligam e n t o p l a s t y compared to silicone arthroplasty (Pellegrini arid Burton, 1986) with better strength (+ 19% compared to - 4 % ) , better stability, less shortening (11% compared to 25%) and less revision procedure (0 compared to 35%). From our data it appears that our ligamentoplasty and tendon interposition has not prevented proximal metacarpal shifting, which occurs progressively during the first 3 years, but this shortening did not decrease the strength, either grip or tip pinch. On the contrary, the only overall differential decrease in strength was observed in the subgroup of patients who did not show any proximal metacarpal migration. Finally, more recently, Conolly and Lanzetta (1993) compared 16 arthrodeses, 15 trapeziectomies and 53 Swanson arthroplasties, with similar good results (69%, 80%, 73.5%) and rates of complications (31%, 33%, 24.5%). In conclusion, trapeziectomy with ligamentoplasty compares favourably with silicone arthroplasty and is free of such long-term complications as instability, wear, and worrying signs of silicone synovitis. References ALLIEU, Y., LUSSIEZ, B. and MARTIN, B. (t990). Rtsultats ~t long terme de l'implant de Swanson darts le traitement de ta rhizarthrose. Revue de Chirurgie Orthop+dique, 76: 437-441. AMADIO, P. C. and DE SILVA, S. P. (1990). Comparison of the results of trapeziometacarpal arthrodesis and arthroplasty in men with osteoarthritis of the trapeziometacarpal joint. Annals of Hand Surgery, 9: 5: 358-363. AMADIO, P. C , MILLENDER, L. H. and SMITH, R. J. (1982). Silicone spacer or tendon spacer for trapezium resection arthroplasty: Comparison of results. Journal of Hand Surgery, 7: 3: 237-244. ASHWORTH, C. R., BLATT, G., CHUINARD, R. G. and STARK, H. H. (1977). Silicone-rubber interposition arthroplasty of the carpometacarpal joint of the thumb. Journal of Hand Surgery. 2: 5: 345-357. BURTON, R. I. and PELLEGRINI, V. D. (1986). Surgical management of basal joint arthritis of the thumb: Part II: Ligament reconstruction with tendon interposition arthroplasty. Journal of Hand Surgery, l l A : 3: 324 332. CARROLL, R. E. and HILL, N. A. ( t 973). Arthrodesis of the carpo-metaearpat joint of the thumb. Journal of Bone and Joint Surgery, 55B: 2: 292-294. CONOLLY, W. B. and LANZETTA, M. (1993). Surgical management of arthritis of the carpo-metacarpal joint of the thumb. Australian and New Zealand Journal of Surgery, 63: 596-603. COZZI, E. P. D6nervation des Articulations du Poignet et de la Main. Tubiana R. ( Ed ) Trait6 de Chirurgie de la Main. Vol 4. Paris, Masson, 1991: 781-787. DE LA CAFFINII~RE, J. Y. (1991). Rtsultats/t long terme de la proth~se totale traptzo-mttacarpienne dans la rhizarthrose. Revue de Chirurgie Orthoptdique, 77: 312-321. EATON, R. G., LANE, L. B., LITTLER, J. W. and KEYSER, J. J. (1984). Ligament reconstruction for the painful thumb carpometaearpal joint: A long-term assessment. Journal of Hand Surgery, 9A: 5: 692-699. FROIMSON, A. I. (1970) Tendon arthroplasty of the trapeziometacarpal joint. Clinical Orthopaedics and Related Research. 70: 191-199. GERVIS, W. H. (1949). Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. Journal of Bone and Joint Surgery, 31B: 4: 537-539. GERVIS, W. H. (1973). A review of excision of the trapezium for osteoarthritis of the trapezio-metacarpal joint after 25 years. Journal of Bone and Joint Surgery, 55B: 1:56 57.
110 GRUBER, L. (1991). Practical approaches to obtaining hand radiographs and special techniques in hand radiology. Hand Clinics. 7: 1: 1-20. LE VIET, D., KERBOULL, L. and EBELIN, M. (1990). Int6r6t de la trap6zectomie darts le traitement de l'arthrose p6ritrap6zienne. Revue de Chirurgie Orthop~dique. 76: 158-161. PELLEGRINI, V. D. and BURTON, R. 1. (1986). Surgical management of basal joint arthritis of the thumb: Part I: Long-term results of silicone implant arthroplasty. Journal of Hand Surgery, 11A: 3: 309-324. POPPEN, N. K. and NIEBAUER, J. J. (1978). "Tie-in" trapezium prosthesis: long-term results. Journal of Hand Surgery, 3: 5: 445-450.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 1 FEBRUARY 1995 SWANSON, A. B., SWANSON, G., de G. and WATERMEIER, J. J. (1981). Trapezium implant arthroplasty: Long term evaluation of 150 cases. Journal of Hand Surgery, 6: 2: 125-141.
Accepted: 25 May 1994 Guy Foucher, MD, Head, SOS Main Strasbourg, 4 Bd du Pr6sident Edwards, 67000 Strasbourg, France. © 1995 The British Society for Surgery of the Hand