- Email: [email protected]
Save the trapezium: Double interposition arthroplasty for the treatment of stage IV disease of the basal joint
Original Communications Save the Trapezium: Double Interposition Arthroplasty for the Treatment of Stage IV Disease of the Basal Ioint O. Alton Barron, MD, Richard G. Eaton, MD, New York, NY Twenty patients with symptomatic stage IV basal joint (pantrapezial) arthritis were treated with double interposition arthroplasty, a new technique that resurfaces the degenerative trapeziometacarpa[ and scaphotrapezial joints after minimal distal trapezial resection. The results of the 21 procedures were reviewed clinically and radiographically after a mean follow-up period of 34 months (minimum, 1 year). Ninety-one percent of the patients were completely satisfied. Seventy-one percent were entirely free of pain, and an additional 24% noted only occasional, mild, high-stress pain that did not limit their activities. One case was a clinical failure. Objectively, all 21 basal joints were stable, with active range of motion to within 90% of normal. Mean grip strength increased 32%, from a force of 26.1 kg to 33.9 kg (p < .01 ). Key pinch strength increased from a force of 5.3 kg to 6.0 kg (an 11% increase; p = .05). A new method of determining changes in basal joint height indicated a mean decrease in height of 5.3% (range, 0%-12%) at rest and 8.1% (range, 0%-20%) under axial compression. This difference was statistically but not functionally significant. Overall, there were 95% good or excellent results and 1 poor result. We believe this technique is simple and preserves the osseous foundation of the basal joints. These results compare favorably with arthroplasties that include trapezium excision for the treatment of stage IV basal joint arthritis. (J Hand Surg 1998;2:196-204. Copyright 9 1998 by the American Society for Surgery of the Hand.)
Degenerative arthrosis of the thumb trapeziometacarpal (TM) and scaphotrapezial ( S T ) j o i n t s is a c o m m o n problem. 1 While radiographic findings do not always imply clinical symptoms, degenerative
From the Departmentof Orthopedic Surgery, St. Luke's-Roosevelt Hospital Center, New York, NY. Received for publication February6, 1996; acceptedfor publication December4, 1997. No benefits in any form have been received or will be receivedfrom a commercialparty related directly or indirectly to the subject of this article. Reprint requests: O. Alton Barron, MD, CV Starr Hand Surgery Center, RooseveltHospital, 1000 10th Ave, 3rd Floor, New York, NY 10019. Copyright 9 1998by the AmericanSocietyfor Surgeryof the Hand. 0363-5023/98/23A02-000253.00/0 196
The Journal of Hand Surgery
arthritis of 1 or both basal joints is a frequent cause of significant pain and dysfunction. In the 1980s, most surgical options for the treatment of pantrapezial arthritis (i.e., involving multiple trapezial articulations) included complete trapezial excision with or without the interposition of tendon, fascia, or a silicone rubber implant. >6 Reports of limited bone resection with interposition were consistent with the concept that complete trapezial excision might represent overtreatment for degeneration limited to the T M joint. 7-9 Since 1977, one author (R.G.E.) has used tendon interposition arthroplasty after conservative trapezial resection for isolated disease of the T M joint with better than 92% good or excellent results, l~ Recognizing that basal joint arthritis could also
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998
A
B
C
197
D
Figure 1. Radiographic staging system for basal joint disease. (A) Stage I, normal articular contours. Slight joint space widening or laxity is sometimes present. (B) Stage II, slight trapeziometacarpal joint narrowing and joint debris less than 2 mm in diameter. (C) Stage III, marked joint space narrowing or obliteration and joint debris greater than 2 mm in diameter. The scaphotrapezial (ST) joint appears normal. (D) Stage IV, as in stage III, but also with degenerative ST joint changes.
involve the proximal articulation (i.e., the ST joint) of the trapezial load-beating axis led to the development of a radiographic staging system, 1I in which stage III includes disease limited to the TM joint and stage IV includes ST joint degeneration as well (Fig. 1). Occasionally, clinical and radiographic assessment may reveal isolated ST joint arthritis, which represents a focal intercarpal arthritis, does not imply TM joint involvement, and should be treated as an isolated problem, leaving the uninvolved TM joint undisturbed. Initially, stage IV cases (disease of both the TM and ST joints) were effectively treated by complete trapezial excision and a silicone spacer. Recognition of the various problems with silicone implants (e.g., implant subluxation, fracture, cold flow, and silicone synovitis) led to a search for alternative procedures.12 14 Suspensionplasties or ligament reconstruction with tendon interposition arthroplasty after total trapezial excision (the LRTI procedure) are currently the most popular techniques used in the treatment of both stage III and stage IV disease. 15-17 Total trapezium excision, however, is technically challenging, as is the elaborate tendon suspension/
interposition configuration. Furthermore, the potentially untoward effects of any thumb shortening and instability that may follow trapezium excision have been reported. 4'7'17 A decrease in thumb skeletal length will create some degree of relative thenar muscle lengthening, which may diminish thumb pinch strength according to the classic Blix curve of muscle strength. The Blix curve graphically describes a progressive loss of contractile force as optimal muscle fiber tension is altered. The simplicity and excellent long-term results of "single interposition arthroplasty" with minimal trapezial resection 1~have subsequently evolved to a "double interposition arthroplasty." We feel that the most notable advantages of this technique over procedures involving trapezium resection include its simplicity and the less traumatic nature of the technique as well as the preservation of the osseous foundation of the thumb through conservative bone resection. The primary purpose of this study was to review the results of this new procedure performed on the initial series of patients and to compare these results with the current standard for the treatment of stage IV disease of the basal joint. To more precisely
198
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
'\
i~
Figure 2. Pinch lateral radiograph. (Left) Diagram showing the musculotendinous forces serving to axially load the basal joint. (Right) Actual radiograph 2 years after double interposition arthroplasty. Both panels depict how the height ratio (A/B) is measured (see Fig. 4).
quantify the degree of thumb shortening in both resting and dynamic states, a new radiographic measuring technique was developed.
Materials and Methods Twenty patients with stage IV disease, confirmed intraoperatively, underwent 21 double interposition arthroplasties that were reviewed both clinically and radiographically. The minimum and average follow-up periods were 1 and 3 years (range, 1-7 years), respectively. The 8 males and 12 females had a mean age of 67 years (range, 51-83 years). Ninety percent were right-hand dominant, and 55% of the procedures were performed on the dominant side. Clinical assessment included chart review, re-examination, and a personal interview of each patient
by the author (O.A.B.) not involved in the surgery or early postoperative care, thereby providing both objective and subjective clinical data. Objective data included grip and key pinch strength, range of motion, basal joint stability, and tenderness. Subjective data included assessment of pain, satisfaction, and activity limitations.
Radiographic Assessment: Loss of Height To quantify any loss of thumb length resulting from the operative procedure, we devised a radiographic ratio for the comparison of the height of the basal joint space before and after surgery in both the unloaded, resting state and the axially loaded, dynamic state. This method includes the "pinch lateral" x-ray film depicted in Fig. 2. This x-ray film was taken with the hand
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998 199
/
Figure 3. Determination of the height ratio (A/B). (Left) Diagram showing measurements from the level of the subchondral bone at appropriate locations. (Right) Height ratio of a basal joint after surgery.
in the same position as for a standard lateral film of the thumb, but with the patient engaged in a thumb tip pinch under maximal effort. If the patient's degree of effort is questionable, a pinch dynamometer can be used while the x-ray is being taken to document the magnitude of the effort. The change in height from before surgery to the time of the latest follow-up visit is calculated from a "height ratio." As shown in Fig. 3, this is the ratio of proximal phalangeal length to the distance between the distal pole of the scaphoid and the distalmost point of the metacarpal head. By including the proximal phalangeal length as the constant over the
length of the follow-up period, errors due to magnification are eliminated. Furthermore, the denominator of the ratio is measured in a manner that bypasses the inevitably irregular contours of the diseased or surgically altered metacarpal base and trapezium. To provide baseline data for comparing height ratio data obtained before and after surgery, normal ratios were calculated from the x-ray fihns of 40 individuals without evidence of disease of the basal joint.
Operative Technique Double interposition arthroplasty (Fig. 4) adheres to the basic technique reported by Eaton et al.~l for
200
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
1
\
2
4
3
(i /
B
Figure 4. Double interposition arthroplasty. (A) 1, Wagner incision; 2, scaphotrapezial (ST) joint arthrotomy and inspection (no bone resection); 3, trapeziometacarpal (TM) joint arthrotomy with limited bone resection (horns only); 4, gouge track made dorsal to volar and perpendicular to the long axis of the metacarpal. (B) 1 and 2, 60% to 70% of flexor carpi radialis (FCR) harvested and distally based free end passed volar to dorsal through the gouge track; 3, interposition of the FCR remnant into the TM joint followed by tension placing of the ligament reconstruction; 4, interposition of the tendon into the ST joint.
the treatment of stage III disease. Through the Wagner (volar) approach (a radially based curvilinear incision, as shown in Fig. 4A), the thenar musculature is elevated from the volar-radial aspect of the carpus and is the first metacarpal to gain access to the ST and TM joints, the base of the first metacarpal, and the flexor carpi radialis tendon (FCR). The ST joint is inspected through a volar-radial arthrotomy. With longitudinal traction across the intact, as yet undisturbed, carpometacarpal joint, an assessment of its articular cartilage can be made. Palpation of this interspace with a dental probe is useful when visualization is limited. A decision is made regarding the necessity of resurfacing based on
the condition of the articular cartilage. Significant fibrillation, erosion, or eburnation are all indications for resurfacing this second joint later in the procedure. The previously described TM joint interposition arthroplasty is performed after minimal bony resection (limited to the volar and dorsal horns of the trapezial saddle). A maximum length strip of FCR tendon is obtained. The distal one third is left anchored at its insertion at the index metacarpal, while the free proximal two thirds are detached and set aside for the interposition implants. The distal FCR is passed through the channel at the base of the metacarpal to create a new volar ligament. ~ It has
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998 201
sufficient length to pass dorsally through the metacarpal base and eventually be sutured to the abductor pollicis longus insertion on the metacarpal. The carpometacarpal tendon implant is prepared to create a double layer of FCR tendon to match the dorsal-tovolar width of the partially resected distal trapezium. This tendon implant is inserted into the TM joint and temporarily stabilized with an absorbable suture passed dorsally through the capsule and tied over the skin. The ST joint is then resurfaced with a rectangular layered graft of remaining FCR tendon using a similar pull-through suture technique. If there is insufficient remaining FCR to create a layered implant of the appropriate dimensions, additional autogenous tendon must be obtained, usually from the flexor carpi ulnaris. In the present series, no bone was resected at the ST joint before resurfacing (see Discussion). Temporary longitudinal Kirschner wire (Kwire) fixation and short arm thumb spica casting provide immobilization for 4 weeks after surgery. Any hyperextension or "collapse" deformity at the metacarpophalangeal (MCP) joint greater than 20 ~ should be corrected with a capsulodesis, which recesses the volar plate and tightens the radial collateral ligament. 18 By preventing MCP hyperextension, the effective transmission of axial force along the thumb ray is enhanced. Volar MCP capsulodesis was performed on 5 thumbs of 4 patients in the present series. Postoperative rehabilitation includes emphasis on extension-abduction for the first week of mobilization and progressive flexion-adducti0n over the following 4 to 6 weeks. Emphasis also must be placed on maintaining MCP joint flexion and avoiding hyperextension during early active range of motion exercises. M i n i m i z i n g Bias and Error
During clinical evaluation, no patient interviews or re-examinations were performed by the operating surgeon or in the surgeon's presence, thereby encouraging patient candor. Use of the same grip and pinch dynamometer eliminated differences between readings, which we have encountered using different devices from the same manufacturer. A questionnaire ensured that data were recorded in a consistent manner. A standardized radiographic technique was taught to specific radiology technicians and proved quite reproducible. The method of determining the height ratio corrects for magnification and eliminates the
effect of slight differences in rotation of the TM joint. Statistical analysis was performed using the Student's t-test for comparison of objective data obtained before and after surgery.
Results The results were classified as excellent, good, fair, or failure as in previous reports of single interposition arthroplasty.11 A rating of excellent implied no pain, pinch strength of at least 90% of that of the contralateral thumb, and no instability. A result was considered good when pain was only occasional, occurring after prolonged activity, and pinch strength of at least 70% of that of the contralateral limb. Patients considered to have a fair result had frequent pain with routine use and less than 70% comparative pinch strength. These patients, however, felt they had improved from their status before surgery. A failure exhibited no improvement following surgery. Ninety-one percent of the patients expressed complete satisfaction with the procedure. Seventy-one percent were completely free of pain. Although still satisfied, an additional 24% noted occasional, mild pain, usually after above-average stress, that did not limit their activities. One patient complained of frequent, moderate pain that was centered not over her basal joint but over her MCP joint, which has been undergoing gradually progressive hyperextension collapse. She is currently considering an MCP joint fusion. This patient had less than 20 ~ of MCP hyperextension before surgery and was not considered a candidate for volar MCP stabilization. Objectively, all 21 basal joints were stable with range of motion within 90% of normal. Grip strength increased significantly (p < .001), an average of 30% (from a force of 26.1-33.9 kg). Pinch strength increased less significantly (p = .05), an average of 12% (from a force of 5.3-6.0 kg). Height ratios for the normal hands ranged from 0.45 to 0.56 (mean, 0.50 _+ 0.02 SD). In our stage IV study grou p , the ratio was 0.515 before surgery, implying a 4% average loss .of height due to this two-joint arthritic process alone. Loss of height resulting from the surgical procedure was determined by comparing stressed and unstressed heightratios obtained before and after surgery. As noted above, the average ratio before surgery was 0.515; this increased to 0.544 in the thumb at rest after surgery. With power pinch, the ratio further increased to 0.56. An increasing height
202
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
ratio implies a decreasing denominator (i.e., basal joint height). These ratios translate to a mean loss of basal joint height of 5.3% (range, 0% to 12%) at rest and 8.1% (range, 0% to 20%) under axial load. While this "loaded" versus unloaded difference is statistically significant (p < .001) and supports the validity of this measurement technique, such small decreases did not appear to be functionally significant. Complications were few and did not affect outcomes. In 2 patients, the temporary K-wire migrated in the early postoperative period, necessitating premature removal. Two other patients noted mild thenat paresthesias distal to the incision that had completely resolved at 2 and 10 months after surgery. Based on these results, 15 of 21 (71%) thumbs were rated as excellent and 5 of 21 (24%) were rated as good. The 1 case of MCP joint collapse was considered a failure. Disease of the basal joint predominately affects females, possibly due to greater baseline ligamentous laxity. The higher percentage of males in the present study (40%) allowed the data to be analyzed according to gender. The only significant difference between the 2 groups was in the loss of basal joint height. Compared with the thumbs of the female patients, the unstressed and stressed thumbs from the male patients had on average a 50% and 48% greater loss of height, respectively, which possibly reflects the generally greater strength of males. Again, the clinical relevance of this notable mathematical difference remains questionable. Discussion Pain at the base of the thumb can derive from I (TM joint, stage III) or both (TM and ST joints, stage IV) of the trapezial articulations along the loadbearing axis of the thumb. The radiographic and anatomic study of North and Eaton~ of arthritic basal joints suggests that only half of the patients with TM joint involvement have concomitant disease of the ST joint and that the appearance of the ST joint on x-ray fihn correlated poorly with the anatomic findings. Because it was a cadaver study, North and Eaton did not know the clinical status of these patients at the time of death. Because of this poor correlation, it is our policy to inspect the ST joint in patients who are to undergo arthroplasty and who have evidence of stage [II disease on x-ray film. The presence of cartilage degeneration would cause such a patient to be classified as stage IV, and double
interposition arthroplasty would therefore become the procedure of choice. Interposition arthroplasty is performed to effectively debride or replace an irregular or eburnated articular surface; this concept applies to both the TM and ST joint articulations. The clinically tender ST joint that is found to be irregular on inspection at the time of surgery contributes to generalized basal joint pain. Even if not tender before surgery, however, an ST joint that is degenerative to some degree probably has the potential to become painful in a variable, sometimes brief, period of time. Our results indicate that the addition of the rather straightforward ST joint interposition based on intraoperative findings ensures that all potential problems have been addressed. The techniques currently popular for the treatment of basal joint arthritis do not differentiate between stage IIt and stage IV disease and include resection of the entire trapezium for isolated carpometacarpal disease (stage III). In our opinion, these procedures are technically more complex and have more potential complications. A less-complicated procedure that largely preserves the osseous anatomy is more appealing. Furthermore, in stage IV (disease of the TM and ST joints), resurfacing the involved articulations while retaining the trapezium appears to be the simplest approach. This approach would eliminate the potential for shortening and subluxation that accompanies any form of trapeziectomy. As with our single interposition arthroplasty, which includes volar ligament reconstruction, any lateral shift of the metacarpal base would be detected with the basal joint stress view. There was no lateral shift detected with this approach. Consideration for the reconstruction of the second joint (the ST joint) must be further clarified. In our series, no bone was resected at the ST joint since after arthrotomy and joint distraction, adequate space existed to insert the double tendon implant. There was, however, a frequent sense of uncertainty because the flat implant tended to slide off the dome of the distal scaphoid, even though a longitudinal K-wire seemed to impale it in place. Despite this uncertainty, no patients had pain or focal tenderness at this level. Even though x-ray films taken after surgery often showed some narrowing (but without obliteration) of this interface, there was no pain attributable to this. Nevertheless, our current technique includes a conservative leveling of the dome of the scaphoid to produce a more planar and potentially more adherent surface to which the interposed ten-
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998
don can be applied. The 1- to 2-mm loss of distal scaphoid does not appear to compromise our effort to maintain maximum thumb length. The arthritic ST joint may respond well to such limited interposition because, unlike the highly mobile TM joint, its stresses are largely axial, and such a pseudoarthrosis solves its problems. Glickel et al. ~') reported that even if a mildly to moderately worn ST joint was not resurfaced as part of the interposition arthroplasty, a good or excellent result was still achieved at a minimum of 5 years. This study helped to formulate our current approach, in which the ST joint that exhibits more significant cartilage erosion or eburnation is resurfaced. The extrusion of either of the interposition autografts dorsally is a theoretical possibility and one that we have considered and attempted to identify. Several observations make this possibility unlikely. First, the dorsal capsules of both the ST and TM joints are not disturbed, their integrity remaining intact. The dorsal capsule provides a natural barrier to extrusion as the graft is pulled dorsally with the absorbable sutures to abut the capsule. Second, both the TM and ST joints are most directly palpable from the dorsal direction, making any graft extrusion most easily palpable there. The fact that we have never palpated a mass consistent with a dorsally extruded graft, even with close observation in the early postoperative period when extrusion is most likely to occur, coupled with the presence of a wider TM joint space on the intraoperative x-ray film also speaks against graft extrusion. Because we have not needed to revise any double interposition arthroplasties thus far, and because a relatively thin graft is used to resurface the ST joint, we cannot be certain of the ultimate fate of the graft. Late postoperative examination of the ST joint, however, routinely showed no evidence of pain. We believe that an accurate, reproducible measurement of metacarpal migration is the ultimate proof of the stability of a basal joint arthroplasty construct. The radiographic technique presented here appears to provide a precise means by which to quantify proximal metacarpal migration by comparing preoperative height ratios with those obtained at the latest postoperative follow-up visit. Dorsal migration of the metacarpal base is minimal due to the anatomic restoration of volar stabilization using FCR-volar ligament reconstruction, m.~ The only current data that allow any comparison of the relative merits of trapezium retention versus excision are those of the recently published, long-term
203
follow-up study of the LRTI procedure by Tomaino et al. '6 By direct x-ray measurement, these investig a t o r s / b u n d a mean loss of height of 13% between early postoperative and late postoperative values in thumbs not under an axial load. How much additional loss of height would occur with dynamic loading remains unknown. In that series, a loss of height greater than 20% was noted in 5 of 24 (21%) patients. To be acceptable as an operative alternative, however, any new procedure must exhibit a clinical efficacy that at least approximates that of the current standard of treatment. Tomaino et al. ~6 reported 95% " e x c e l l e n t " pain relief and significant increases in grip and tip pinch strength. The results of double interposition arthroplasty at a mean follow-up period of 3 years (91% complete patient satisfaction and 95% total or near-total pain relief) compare favorably with the above results. Grip and pinch strengths after surgery were also comparable. Our height ratio measurements compared preoperative with the latest postoperative x-ray films, as described above. In the unloaded state, our maximum loss of height was 12% (average, 5.3%). Even under axial load, only 2 patients demonstrated the maximum 20% loss of height seen in this series. It is unlikely, however, that further shortening will occur with the trapezium preserved.
References 1. North ER, Eaton RG. Degenerative joint disease of the trapezium: a comparative radiographic and anatomic study. J Hand Surg 1983;8:160-167. 2. Eaton RG. Replacement of the trapezium for arthritis of the basal articulations. A new technique with stabilization by tenodesis. J Bone Joint Surg 1979;61A:76-82. 3. Froimson AL. Tendon arthroplasty of the trapeziolnetacarpal joint. Clin Orthop 1970~70:19l-199. 4. Gervis WH. Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. J Bone Joint Surg 1949;42B: 537-553. 5. Murley AHG. Excision of the trapezium in osteoarthritis of the first carpometacarpal. J Bone Joint Surg 1960;42B: 502-507. 6. Swanson AB. Disabling arthritis of the base of the thumb. Treatment by resection of the trapezium and flexible (silicone) implant arthroplasty. J Bone Joint Surg 1972;54A: 456-471. 7. Ashworth CR, Blatt G, Chuinard RG, Stark HH. Siliconerubber interposition arthroplasty of the carpometacarpal joint of the thumb. J Hand Surg 1977;2:345-357. 8. Kessler I, Axer A. Arthroplasty of the first carpometacarpal joint with a silicone implant. Plast Reconstr Surg 1971;47: 252-256.
204
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
9. Millender LH, Nalebuff EA, Amadio PC, Philips C. Interpositional arthroplasty for rheumatoid carpometacarpal joint disease. J Hand Surg 1978;3:533-541. 10. Eaton RG, Glickel SZ, Littler JW. Tendon interposition arthroplasty for degenerative arthritis of the trapeziometacarpal joint of the thumb. J Hand Surg 1985;10A: 645-658. 11. Eaton RG, Lane LB, Littler JW, Keyser JJ. Ligament reconstruction for the painful thumb carpometacarpal joint: a long-term assessment. J Hand Surg 1984;9A: 692 699. 12. Carter PR, Benton LJ, Dysert PA. Silicone rubber carpal implants: study of the incidence of late osseous complications. J Hand Surg 1986;11A:639-644. 13. Peimer CA, Medige J, Eckert BS, Wright JR, Howard CS. Reactive synovitis after silicone arthroplasty. J Hand Surg 1986; 11A:624- 638. 14. Pelligrini VD Jr, Burton RI. Surgical management of basal joint arthritis of the thumb. Part I. Long-term results of
15.
16.
17. 18.
19.
silicone implant arthroplasty. J Hand Surg 1986;11A:309324. Burton RI, Pelligrini VD Jr. Surgical management of basal joint arthritis of the thumb. Part II. Ligament reconstruction with tendon interposition arthroplasty. J Hand Surg 1986;11A:324-332. Tomaino MM, Pelligrini VD Jr, Burton RI. Arthroplasty of the basal joint of the thumb. J Bone Joint Surg 1995;77A: 346 -355. Thompson JS. Suspensionplasty. J Orthop Surg Techniques 1989;4:1-13. Eaton RG, Floyd WE III. Thumb metacarpophalangeal capsulodesis: an adjunct procedure to basal joint arthroplasty for collapse deformity of the first ray. J Hand Surg 1988;13A:449-453. Glickel SZ, Kornstein AN, Eaton RG. Long-term follow-up of trapeziometacarpal arthroplasty with coexisting scaphotrapezial disease. J Hand Surg 1992;17A: 612-620.
Degenerative arthrosis of the thumb trapeziometacarpal (TM) and scaphotrapezial ( S T ) j o i n t s is a c o m m o n problem. 1 While radiographic findings do not always imply clinical symptoms, degenerative
From the Departmentof Orthopedic Surgery, St. Luke's-Roosevelt Hospital Center, New York, NY. Received for publication February6, 1996; acceptedfor publication December4, 1997. No benefits in any form have been received or will be receivedfrom a commercialparty related directly or indirectly to the subject of this article. Reprint requests: O. Alton Barron, MD, CV Starr Hand Surgery Center, RooseveltHospital, 1000 10th Ave, 3rd Floor, New York, NY 10019. Copyright 9 1998by the AmericanSocietyfor Surgeryof the Hand. 0363-5023/98/23A02-000253.00/0 196
The Journal of Hand Surgery
arthritis of 1 or both basal joints is a frequent cause of significant pain and dysfunction. In the 1980s, most surgical options for the treatment of pantrapezial arthritis (i.e., involving multiple trapezial articulations) included complete trapezial excision with or without the interposition of tendon, fascia, or a silicone rubber implant. >6 Reports of limited bone resection with interposition were consistent with the concept that complete trapezial excision might represent overtreatment for degeneration limited to the T M joint. 7-9 Since 1977, one author (R.G.E.) has used tendon interposition arthroplasty after conservative trapezial resection for isolated disease of the T M joint with better than 92% good or excellent results, l~ Recognizing that basal joint arthritis could also
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998
A
B
C
197
D
Figure 1. Radiographic staging system for basal joint disease. (A) Stage I, normal articular contours. Slight joint space widening or laxity is sometimes present. (B) Stage II, slight trapeziometacarpal joint narrowing and joint debris less than 2 mm in diameter. (C) Stage III, marked joint space narrowing or obliteration and joint debris greater than 2 mm in diameter. The scaphotrapezial (ST) joint appears normal. (D) Stage IV, as in stage III, but also with degenerative ST joint changes.
involve the proximal articulation (i.e., the ST joint) of the trapezial load-beating axis led to the development of a radiographic staging system, 1I in which stage III includes disease limited to the TM joint and stage IV includes ST joint degeneration as well (Fig. 1). Occasionally, clinical and radiographic assessment may reveal isolated ST joint arthritis, which represents a focal intercarpal arthritis, does not imply TM joint involvement, and should be treated as an isolated problem, leaving the uninvolved TM joint undisturbed. Initially, stage IV cases (disease of both the TM and ST joints) were effectively treated by complete trapezial excision and a silicone spacer. Recognition of the various problems with silicone implants (e.g., implant subluxation, fracture, cold flow, and silicone synovitis) led to a search for alternative procedures.12 14 Suspensionplasties or ligament reconstruction with tendon interposition arthroplasty after total trapezial excision (the LRTI procedure) are currently the most popular techniques used in the treatment of both stage III and stage IV disease. 15-17 Total trapezium excision, however, is technically challenging, as is the elaborate tendon suspension/
interposition configuration. Furthermore, the potentially untoward effects of any thumb shortening and instability that may follow trapezium excision have been reported. 4'7'17 A decrease in thumb skeletal length will create some degree of relative thenar muscle lengthening, which may diminish thumb pinch strength according to the classic Blix curve of muscle strength. The Blix curve graphically describes a progressive loss of contractile force as optimal muscle fiber tension is altered. The simplicity and excellent long-term results of "single interposition arthroplasty" with minimal trapezial resection 1~have subsequently evolved to a "double interposition arthroplasty." We feel that the most notable advantages of this technique over procedures involving trapezium resection include its simplicity and the less traumatic nature of the technique as well as the preservation of the osseous foundation of the thumb through conservative bone resection. The primary purpose of this study was to review the results of this new procedure performed on the initial series of patients and to compare these results with the current standard for the treatment of stage IV disease of the basal joint. To more precisely
198
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
'\
i~
Figure 2. Pinch lateral radiograph. (Left) Diagram showing the musculotendinous forces serving to axially load the basal joint. (Right) Actual radiograph 2 years after double interposition arthroplasty. Both panels depict how the height ratio (A/B) is measured (see Fig. 4).
quantify the degree of thumb shortening in both resting and dynamic states, a new radiographic measuring technique was developed.
Materials and Methods Twenty patients with stage IV disease, confirmed intraoperatively, underwent 21 double interposition arthroplasties that were reviewed both clinically and radiographically. The minimum and average follow-up periods were 1 and 3 years (range, 1-7 years), respectively. The 8 males and 12 females had a mean age of 67 years (range, 51-83 years). Ninety percent were right-hand dominant, and 55% of the procedures were performed on the dominant side. Clinical assessment included chart review, re-examination, and a personal interview of each patient
by the author (O.A.B.) not involved in the surgery or early postoperative care, thereby providing both objective and subjective clinical data. Objective data included grip and key pinch strength, range of motion, basal joint stability, and tenderness. Subjective data included assessment of pain, satisfaction, and activity limitations.
Radiographic Assessment: Loss of Height To quantify any loss of thumb length resulting from the operative procedure, we devised a radiographic ratio for the comparison of the height of the basal joint space before and after surgery in both the unloaded, resting state and the axially loaded, dynamic state. This method includes the "pinch lateral" x-ray film depicted in Fig. 2. This x-ray film was taken with the hand
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998 199
/
Figure 3. Determination of the height ratio (A/B). (Left) Diagram showing measurements from the level of the subchondral bone at appropriate locations. (Right) Height ratio of a basal joint after surgery.
in the same position as for a standard lateral film of the thumb, but with the patient engaged in a thumb tip pinch under maximal effort. If the patient's degree of effort is questionable, a pinch dynamometer can be used while the x-ray is being taken to document the magnitude of the effort. The change in height from before surgery to the time of the latest follow-up visit is calculated from a "height ratio." As shown in Fig. 3, this is the ratio of proximal phalangeal length to the distance between the distal pole of the scaphoid and the distalmost point of the metacarpal head. By including the proximal phalangeal length as the constant over the
length of the follow-up period, errors due to magnification are eliminated. Furthermore, the denominator of the ratio is measured in a manner that bypasses the inevitably irregular contours of the diseased or surgically altered metacarpal base and trapezium. To provide baseline data for comparing height ratio data obtained before and after surgery, normal ratios were calculated from the x-ray fihns of 40 individuals without evidence of disease of the basal joint.
Operative Technique Double interposition arthroplasty (Fig. 4) adheres to the basic technique reported by Eaton et al.~l for
200
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
1
\
2
4
3
(i /
B
Figure 4. Double interposition arthroplasty. (A) 1, Wagner incision; 2, scaphotrapezial (ST) joint arthrotomy and inspection (no bone resection); 3, trapeziometacarpal (TM) joint arthrotomy with limited bone resection (horns only); 4, gouge track made dorsal to volar and perpendicular to the long axis of the metacarpal. (B) 1 and 2, 60% to 70% of flexor carpi radialis (FCR) harvested and distally based free end passed volar to dorsal through the gouge track; 3, interposition of the FCR remnant into the TM joint followed by tension placing of the ligament reconstruction; 4, interposition of the tendon into the ST joint.
the treatment of stage III disease. Through the Wagner (volar) approach (a radially based curvilinear incision, as shown in Fig. 4A), the thenar musculature is elevated from the volar-radial aspect of the carpus and is the first metacarpal to gain access to the ST and TM joints, the base of the first metacarpal, and the flexor carpi radialis tendon (FCR). The ST joint is inspected through a volar-radial arthrotomy. With longitudinal traction across the intact, as yet undisturbed, carpometacarpal joint, an assessment of its articular cartilage can be made. Palpation of this interspace with a dental probe is useful when visualization is limited. A decision is made regarding the necessity of resurfacing based on
the condition of the articular cartilage. Significant fibrillation, erosion, or eburnation are all indications for resurfacing this second joint later in the procedure. The previously described TM joint interposition arthroplasty is performed after minimal bony resection (limited to the volar and dorsal horns of the trapezial saddle). A maximum length strip of FCR tendon is obtained. The distal one third is left anchored at its insertion at the index metacarpal, while the free proximal two thirds are detached and set aside for the interposition implants. The distal FCR is passed through the channel at the base of the metacarpal to create a new volar ligament. ~ It has
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998 201
sufficient length to pass dorsally through the metacarpal base and eventually be sutured to the abductor pollicis longus insertion on the metacarpal. The carpometacarpal tendon implant is prepared to create a double layer of FCR tendon to match the dorsal-tovolar width of the partially resected distal trapezium. This tendon implant is inserted into the TM joint and temporarily stabilized with an absorbable suture passed dorsally through the capsule and tied over the skin. The ST joint is then resurfaced with a rectangular layered graft of remaining FCR tendon using a similar pull-through suture technique. If there is insufficient remaining FCR to create a layered implant of the appropriate dimensions, additional autogenous tendon must be obtained, usually from the flexor carpi ulnaris. In the present series, no bone was resected at the ST joint before resurfacing (see Discussion). Temporary longitudinal Kirschner wire (Kwire) fixation and short arm thumb spica casting provide immobilization for 4 weeks after surgery. Any hyperextension or "collapse" deformity at the metacarpophalangeal (MCP) joint greater than 20 ~ should be corrected with a capsulodesis, which recesses the volar plate and tightens the radial collateral ligament. 18 By preventing MCP hyperextension, the effective transmission of axial force along the thumb ray is enhanced. Volar MCP capsulodesis was performed on 5 thumbs of 4 patients in the present series. Postoperative rehabilitation includes emphasis on extension-abduction for the first week of mobilization and progressive flexion-adducti0n over the following 4 to 6 weeks. Emphasis also must be placed on maintaining MCP joint flexion and avoiding hyperextension during early active range of motion exercises. M i n i m i z i n g Bias and Error
During clinical evaluation, no patient interviews or re-examinations were performed by the operating surgeon or in the surgeon's presence, thereby encouraging patient candor. Use of the same grip and pinch dynamometer eliminated differences between readings, which we have encountered using different devices from the same manufacturer. A questionnaire ensured that data were recorded in a consistent manner. A standardized radiographic technique was taught to specific radiology technicians and proved quite reproducible. The method of determining the height ratio corrects for magnification and eliminates the
effect of slight differences in rotation of the TM joint. Statistical analysis was performed using the Student's t-test for comparison of objective data obtained before and after surgery.
Results The results were classified as excellent, good, fair, or failure as in previous reports of single interposition arthroplasty.11 A rating of excellent implied no pain, pinch strength of at least 90% of that of the contralateral thumb, and no instability. A result was considered good when pain was only occasional, occurring after prolonged activity, and pinch strength of at least 70% of that of the contralateral limb. Patients considered to have a fair result had frequent pain with routine use and less than 70% comparative pinch strength. These patients, however, felt they had improved from their status before surgery. A failure exhibited no improvement following surgery. Ninety-one percent of the patients expressed complete satisfaction with the procedure. Seventy-one percent were completely free of pain. Although still satisfied, an additional 24% noted occasional, mild pain, usually after above-average stress, that did not limit their activities. One patient complained of frequent, moderate pain that was centered not over her basal joint but over her MCP joint, which has been undergoing gradually progressive hyperextension collapse. She is currently considering an MCP joint fusion. This patient had less than 20 ~ of MCP hyperextension before surgery and was not considered a candidate for volar MCP stabilization. Objectively, all 21 basal joints were stable with range of motion within 90% of normal. Grip strength increased significantly (p < .001), an average of 30% (from a force of 26.1-33.9 kg). Pinch strength increased less significantly (p = .05), an average of 12% (from a force of 5.3-6.0 kg). Height ratios for the normal hands ranged from 0.45 to 0.56 (mean, 0.50 _+ 0.02 SD). In our stage IV study grou p , the ratio was 0.515 before surgery, implying a 4% average loss .of height due to this two-joint arthritic process alone. Loss of height resulting from the surgical procedure was determined by comparing stressed and unstressed heightratios obtained before and after surgery. As noted above, the average ratio before surgery was 0.515; this increased to 0.544 in the thumb at rest after surgery. With power pinch, the ratio further increased to 0.56. An increasing height
202
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
ratio implies a decreasing denominator (i.e., basal joint height). These ratios translate to a mean loss of basal joint height of 5.3% (range, 0% to 12%) at rest and 8.1% (range, 0% to 20%) under axial load. While this "loaded" versus unloaded difference is statistically significant (p < .001) and supports the validity of this measurement technique, such small decreases did not appear to be functionally significant. Complications were few and did not affect outcomes. In 2 patients, the temporary K-wire migrated in the early postoperative period, necessitating premature removal. Two other patients noted mild thenat paresthesias distal to the incision that had completely resolved at 2 and 10 months after surgery. Based on these results, 15 of 21 (71%) thumbs were rated as excellent and 5 of 21 (24%) were rated as good. The 1 case of MCP joint collapse was considered a failure. Disease of the basal joint predominately affects females, possibly due to greater baseline ligamentous laxity. The higher percentage of males in the present study (40%) allowed the data to be analyzed according to gender. The only significant difference between the 2 groups was in the loss of basal joint height. Compared with the thumbs of the female patients, the unstressed and stressed thumbs from the male patients had on average a 50% and 48% greater loss of height, respectively, which possibly reflects the generally greater strength of males. Again, the clinical relevance of this notable mathematical difference remains questionable. Discussion Pain at the base of the thumb can derive from I (TM joint, stage III) or both (TM and ST joints, stage IV) of the trapezial articulations along the loadbearing axis of the thumb. The radiographic and anatomic study of North and Eaton~ of arthritic basal joints suggests that only half of the patients with TM joint involvement have concomitant disease of the ST joint and that the appearance of the ST joint on x-ray fihn correlated poorly with the anatomic findings. Because it was a cadaver study, North and Eaton did not know the clinical status of these patients at the time of death. Because of this poor correlation, it is our policy to inspect the ST joint in patients who are to undergo arthroplasty and who have evidence of stage [II disease on x-ray film. The presence of cartilage degeneration would cause such a patient to be classified as stage IV, and double
interposition arthroplasty would therefore become the procedure of choice. Interposition arthroplasty is performed to effectively debride or replace an irregular or eburnated articular surface; this concept applies to both the TM and ST joint articulations. The clinically tender ST joint that is found to be irregular on inspection at the time of surgery contributes to generalized basal joint pain. Even if not tender before surgery, however, an ST joint that is degenerative to some degree probably has the potential to become painful in a variable, sometimes brief, period of time. Our results indicate that the addition of the rather straightforward ST joint interposition based on intraoperative findings ensures that all potential problems have been addressed. The techniques currently popular for the treatment of basal joint arthritis do not differentiate between stage IIt and stage IV disease and include resection of the entire trapezium for isolated carpometacarpal disease (stage III). In our opinion, these procedures are technically more complex and have more potential complications. A less-complicated procedure that largely preserves the osseous anatomy is more appealing. Furthermore, in stage IV (disease of the TM and ST joints), resurfacing the involved articulations while retaining the trapezium appears to be the simplest approach. This approach would eliminate the potential for shortening and subluxation that accompanies any form of trapeziectomy. As with our single interposition arthroplasty, which includes volar ligament reconstruction, any lateral shift of the metacarpal base would be detected with the basal joint stress view. There was no lateral shift detected with this approach. Consideration for the reconstruction of the second joint (the ST joint) must be further clarified. In our series, no bone was resected at the ST joint since after arthrotomy and joint distraction, adequate space existed to insert the double tendon implant. There was, however, a frequent sense of uncertainty because the flat implant tended to slide off the dome of the distal scaphoid, even though a longitudinal K-wire seemed to impale it in place. Despite this uncertainty, no patients had pain or focal tenderness at this level. Even though x-ray films taken after surgery often showed some narrowing (but without obliteration) of this interface, there was no pain attributable to this. Nevertheless, our current technique includes a conservative leveling of the dome of the scaphoid to produce a more planar and potentially more adherent surface to which the interposed ten-
The Journal of Hand Surgery / Vol. 23A No. 2 March 1998
don can be applied. The 1- to 2-mm loss of distal scaphoid does not appear to compromise our effort to maintain maximum thumb length. The arthritic ST joint may respond well to such limited interposition because, unlike the highly mobile TM joint, its stresses are largely axial, and such a pseudoarthrosis solves its problems. Glickel et al. ~') reported that even if a mildly to moderately worn ST joint was not resurfaced as part of the interposition arthroplasty, a good or excellent result was still achieved at a minimum of 5 years. This study helped to formulate our current approach, in which the ST joint that exhibits more significant cartilage erosion or eburnation is resurfaced. The extrusion of either of the interposition autografts dorsally is a theoretical possibility and one that we have considered and attempted to identify. Several observations make this possibility unlikely. First, the dorsal capsules of both the ST and TM joints are not disturbed, their integrity remaining intact. The dorsal capsule provides a natural barrier to extrusion as the graft is pulled dorsally with the absorbable sutures to abut the capsule. Second, both the TM and ST joints are most directly palpable from the dorsal direction, making any graft extrusion most easily palpable there. The fact that we have never palpated a mass consistent with a dorsally extruded graft, even with close observation in the early postoperative period when extrusion is most likely to occur, coupled with the presence of a wider TM joint space on the intraoperative x-ray film also speaks against graft extrusion. Because we have not needed to revise any double interposition arthroplasties thus far, and because a relatively thin graft is used to resurface the ST joint, we cannot be certain of the ultimate fate of the graft. Late postoperative examination of the ST joint, however, routinely showed no evidence of pain. We believe that an accurate, reproducible measurement of metacarpal migration is the ultimate proof of the stability of a basal joint arthroplasty construct. The radiographic technique presented here appears to provide a precise means by which to quantify proximal metacarpal migration by comparing preoperative height ratios with those obtained at the latest postoperative follow-up visit. Dorsal migration of the metacarpal base is minimal due to the anatomic restoration of volar stabilization using FCR-volar ligament reconstruction, m.~ The only current data that allow any comparison of the relative merits of trapezium retention versus excision are those of the recently published, long-term
203
follow-up study of the LRTI procedure by Tomaino et al. '6 By direct x-ray measurement, these investig a t o r s / b u n d a mean loss of height of 13% between early postoperative and late postoperative values in thumbs not under an axial load. How much additional loss of height would occur with dynamic loading remains unknown. In that series, a loss of height greater than 20% was noted in 5 of 24 (21%) patients. To be acceptable as an operative alternative, however, any new procedure must exhibit a clinical efficacy that at least approximates that of the current standard of treatment. Tomaino et al. ~6 reported 95% " e x c e l l e n t " pain relief and significant increases in grip and tip pinch strength. The results of double interposition arthroplasty at a mean follow-up period of 3 years (91% complete patient satisfaction and 95% total or near-total pain relief) compare favorably with the above results. Grip and pinch strengths after surgery were also comparable. Our height ratio measurements compared preoperative with the latest postoperative x-ray films, as described above. In the unloaded state, our maximum loss of height was 12% (average, 5.3%). Even under axial load, only 2 patients demonstrated the maximum 20% loss of height seen in this series. It is unlikely, however, that further shortening will occur with the trapezium preserved.
References 1. North ER, Eaton RG. Degenerative joint disease of the trapezium: a comparative radiographic and anatomic study. J Hand Surg 1983;8:160-167. 2. Eaton RG. Replacement of the trapezium for arthritis of the basal articulations. A new technique with stabilization by tenodesis. J Bone Joint Surg 1979;61A:76-82. 3. Froimson AL. Tendon arthroplasty of the trapeziolnetacarpal joint. Clin Orthop 1970~70:19l-199. 4. Gervis WH. Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. J Bone Joint Surg 1949;42B: 537-553. 5. Murley AHG. Excision of the trapezium in osteoarthritis of the first carpometacarpal. J Bone Joint Surg 1960;42B: 502-507. 6. Swanson AB. Disabling arthritis of the base of the thumb. Treatment by resection of the trapezium and flexible (silicone) implant arthroplasty. J Bone Joint Surg 1972;54A: 456-471. 7. Ashworth CR, Blatt G, Chuinard RG, Stark HH. Siliconerubber interposition arthroplasty of the carpometacarpal joint of the thumb. J Hand Surg 1977;2:345-357. 8. Kessler I, Axer A. Arthroplasty of the first carpometacarpal joint with a silicone implant. Plast Reconstr Surg 1971;47: 252-256.
204
Barron and Eaton / Double Arthroplasty for Basal Joint Arthritis
9. Millender LH, Nalebuff EA, Amadio PC, Philips C. Interpositional arthroplasty for rheumatoid carpometacarpal joint disease. J Hand Surg 1978;3:533-541. 10. Eaton RG, Glickel SZ, Littler JW. Tendon interposition arthroplasty for degenerative arthritis of the trapeziometacarpal joint of the thumb. J Hand Surg 1985;10A: 645-658. 11. Eaton RG, Lane LB, Littler JW, Keyser JJ. Ligament reconstruction for the painful thumb carpometacarpal joint: a long-term assessment. J Hand Surg 1984;9A: 692 699. 12. Carter PR, Benton LJ, Dysert PA. Silicone rubber carpal implants: study of the incidence of late osseous complications. J Hand Surg 1986;11A:639-644. 13. Peimer CA, Medige J, Eckert BS, Wright JR, Howard CS. Reactive synovitis after silicone arthroplasty. J Hand Surg 1986; 11A:624- 638. 14. Pelligrini VD Jr, Burton RI. Surgical management of basal joint arthritis of the thumb. Part I. Long-term results of
15.
16.
17. 18.
19.
silicone implant arthroplasty. J Hand Surg 1986;11A:309324. Burton RI, Pelligrini VD Jr. Surgical management of basal joint arthritis of the thumb. Part II. Ligament reconstruction with tendon interposition arthroplasty. J Hand Surg 1986;11A:324-332. Tomaino MM, Pelligrini VD Jr, Burton RI. Arthroplasty of the basal joint of the thumb. J Bone Joint Surg 1995;77A: 346 -355. Thompson JS. Suspensionplasty. J Orthop Surg Techniques 1989;4:1-13. Eaton RG, Floyd WE III. Thumb metacarpophalangeal capsulodesis: an adjunct procedure to basal joint arthroplasty for collapse deformity of the first ray. J Hand Surg 1988;13A:449-453. Glickel SZ, Kornstein AN, Eaton RG. Long-term follow-up of trapeziometacarpal arthroplasty with coexisting scaphotrapezial disease. J Hand Surg 1992;17A: 612-620.