Trapeziectomy and Suture-Button Suspensionplasty for Basilar Thumb Arthritis: Is It Enough to Prevent First Ray Subsidence?

Abstract

Introduction  Trapeziectomy and suture-button suspensionplasty (SBS) are a novel option to treat end-stage trapeziometacarpal (TMC) osteoarthritis. Our purpose is to evaluate our outcomes with this technique and in this setting, with a minimum of 18 months of follow-up.

Materials and Methods  Twenty-eight patients were included, operated between 2016 and 2018. We recorded demographic data, preoperative Eaton stage, follow-up and operative times. The patients completed the Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) questionnaire and tip pinch, key pinch, and grip strength were measured. First metacarpal subsidence was calculated, and postoperative complications were documented.

Results  The average follow-up was 34 months. The mean QuickDASH was 32 at the final follow-up. The average strength results were 20 kg for grip, 3.6 kg for tip pinch, and 4.2 kg for key pinch. The rate of first ray subsidence was 10.7%. We encountered three complications: a hardware intolerance, a second metacarpal fracture, and a suture rupture. There was one reoperation to remove an implant.

Conclusion  Trapeziectomy and SBS functional results are similar to other techniques, with less subsidence of the first ray and allowing for early mobilization and fast recovery. This procedure is a safe and promising option in the treatment of TMC osteoarthritis, with good medium-term outcomes.

Introduction

Trapeziometacarpal (TMC) joint is the second most common site of hand arthritis and its prevalence starts to increase at the age of 40 years, being present in ~25% of the postmenopausal women. 1 2 Despite this, treatment is only needed when there is pain and decreased motion, interfering with daily life activities. 3

There is a multitude of surgical procedures described for the treatment of this condition, including trapeziectomy alone, hematoma distraction arthroplasty, arthrodesis, trapeziectomy and ligament reconstruction with or without tendon interposition, prosthetic joint replacement, or suspension ligamentoplasty, with no consensus in which one is the gold standard. 3 4 5 6

Trapeziectomy, the root treatment in current thumb basilar arthritis surgery, creates adequate pain relief and has few complications, becoming an option for the older patients. 2 7

In a younger patient, however, a good range of motion associated with strength and precision is required to fine and gross motor function, which is not currently addressed with the former technique. 7

Subsidence of the first metacarpal, as seen in trapeziectomy alone, decreases the lever arm of the thumb and may limit pinch strength and result in impingement between the scaphoid and the first metacarpal. 2 Meta-analyses have supported simple trapeziectomy with stabilization of the thumb carpometacarpal (CMC) space for 4 weeks with a Kirschner (K)-wire to allow a hematoma to form and prevent first ray subsidence; however, pain, infection, and delay in rehabilitation are disadvantages of this technique. 7

In 1986, ligament reconstruction and tendon interposition (LRTI) arthroplasty was described by Burton and Pellegrini, using flexor carpi radialis (FCR) tendon. 8 Soon, LRTI with FCR and abductor pollicis longus (APL) tendons became the most widely performed techniques for thumb basilar arthritis. 1

Recently described procedures mix trapeziectomy with suspensionplasty between the first and second metacarpals to prevent index subsidence into the newly formed TMC space, either using a suture-button 7 9 or just a high-strength orthopaedic suture. 10 In our approach, we perform a trapeziectomy followed by suspension of the first and second metacarpals using a single strand of a nonabsorbable suture with Mini TightRope suture button. This technique offers similarly good results when compared with previously described techniques when it comes to pain reduction, restoration of mobility, and strength. The suture-button suspensionplasty (SBS) provides the added benefit of eliminating the need for K-wire fixation, for casting and for tissue reconstructions to heal, allowing for early mobilization at 7 to 10 days. 2 7 9 11 12 This feature can lead to increased patient satisfaction, faster recovery, and return to full activity. 7 Furthermore, suspensionplasty with a button obviates the need to sacrifice a tendon, thus causing less morbidity. 1

All new techniques have some risks. Potential complications of this approach have been reported: index metacarpal fractures, tissue reaction leading to osteolysis, extensor pollicis longus tendon rupture, suspensionplasty collapse with subluxation of the thumb metacarpal and pain, overtightening of the suture button with resulting symptomatic impingement of the first and second metacarpal bases, and discomfort from superficially located hardware. 5 13 14 15 16

Our purpose is to evaluate our results with trapeziectomy and suspensionplasty with a suture button in the treatment of thumb basilar arthritis, with a minimum of 18 months of follow-up. Our primary outcomes were first ray subsidence into the trapezial space and rate of complications.

Materials and Methods

After we obtained the Institutional Review Board approval, we performed a retrospective review of 28 patients (28 thumb–s) with at least 1 year of follow-up after trapeziectomy and SBS, using Mini TightRope CMC system (Arthrex, Naples, Florida, United States), for end-stage refractory TMC arthritis, between January 2016 and December 2018.

All surgeries were performed by the Hand Team, which includes at least one fellowship-trained hand surgeon. Patients' medical charts and radiographs were reviewed. Details including patient demographics, dominance, complications, and postoperative parameters were recorded. Postoperative grip, key pinch, and tip pinch strength were obtained using a dynamometer (Jamar, Cambridge). For this evaluation, we compare our results to the strength norm tables based on age, gender, and dominance, obtaining a differential between these norms and the values for our patients. Patients were asked to complete the Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) questionnaire at the latest follow-up visit.

The metacarpal subsidence was obtained indirectly by calculating pre- and postoperative trapezial space heights (TSHs), with the postsurgical value being expressed as a percentage of the preoperative, as described by Yao and Cheah. 12 This means that 100% of postsurgical TSH indicates 0% of subsidence. As in other studies, the trapezial height was measured in a lateral radiograph of the thumb, drawing three lines: one connecting the corners of the first metacarpal base, one joining the two corners of the scaphoid, and a third line along the metacarpal central axis, bisecting the former two. The TSH was measured as the length along the third line that lies between the lines at the metacarpal base and distal scaphoid. We used the Picture Archiving and Communication System software for the measurement (SECTRA, Linköping, Sweden).

Of the 28 patients, 4 did not respond to our last contact for testing hand strength and 3 failed to complete the QuickDASH questionnaire. For these last three individuals, we have just evaluated metacarpal subsidence with at least 1 year of follow-up and collected data about demographics and complications until the last visit.

Surgical Technique

For the trapeziectomy, we use a dorsoradial skin incision over the TMC joint, with 3 to 4 cm of length. Capsulotomy is done just volar to APL insertion on the first metacarpal, with care to avoid damaging the dorsal radial sensory nerve. Next, trapeziectomy is performed with the aid of an osteotome and a rongeur forceps ( Fig. 1 ).

Fig. 1.

Dorsoradial approach and trapeziectomy.

A second (2 cm) incision is made dorsally between the second and third metacarpal bases to prepare the exit point of the guidewire in the index metacarpal. To do so, we elevate the dorsal interosseous muscle subperiosteally from the ulnar aspect of the second metacarpal to view its base. With the hand in a neutral position, a 1.1-mm tapered suture passing wire is drilled from the base of the first metacarpal, just beneath a portion of the radial aspect of the APL tendon, to the second metacarpal metadiaphyseal junction, in a 40-degree angle trajectory ( Fig. 2 ). Then, the suture of the Mini TightRope is passed through the loop in the guidewire from radial to ulnar, leaving the button in the first metacarpal base ( Fig. 3 ). A second button is passed down to the index metacarpal cortex. The thumb is then reduced to its anatomical position with axial traction, extension and palmar abduction, and the suture is tied in the second webspace. To have the correct tensioning, we interpose a clamp, with 1 to 2 mm of thickness, between the button and the base of the first metacarpal before tightening the knot ( Fig. 4 ).

Fig. 2.

Guidewire trajectory.

Fig. 3.

Insertion of the suture button.

Fig. 4.

Tightening of the knot with a clamp interposition.

The wounds are closed with an absorbable suture and a Robert Jones bandage is kept for 14 days. At the first follow-up visit, at 2 weeks postoperatively, the immobilization is removed, and the patients are referred to physical therapy. We restrict strength activities to 50% of grip power until week 6.

Results

Twenty-eight patients were included in our study, 24 females (85.7%) and 4 males (14.3%), with a mean age of 63 years (range, 51–78 years). Radiographic evaluation of the disease showed Eaton stage III in 19 thumbs (68%) and stage IV in 9 (32%). The average tourniquet time was 60 minutes (ranging from 45 to 75 minutes). The mean follow-up time was 34 months (range, 18–42 months). In 20 cases (71.4%), the operated hand was the dominant one (the right in all).

The mean postoperative QuickDASH was 32 (fluctuating from 4.5 to 90.9) and was available for 25 patients (3 failed to answer the questionnaire). The patient with the worst QuickDASH had a complication of intolerance to the hardware and two comorbidities: a psychiatric disease and a cervicobrachial syndrome. We also analyzed the comorbidities that could have impaired these functional scores ( Tables 1 and 2 ). Forty-four per cent of the cases had QuickDASH scores below 30 and 26% had scores above 40.

Table 1. Patients demographics and outcomes.

Patient	Age (y)	Gender	Operated side	Dominant side	Eaton stage	Follow-up (mo)	Trapezial height (%)	Grip strength (kg)	Differential grip (kg)	Tip pinch (kg)	Differential tip (kg)	Key pinch (kg)	Differential key (kg)
1	59	F	Right	Right	III	42	56.2	22	6	3.5	2.3	3.5	3.5
2	70	F	Right	Right	IV	41	100	10	10	1	4.4	2	4.2
3	70	F	Right	Right	III	41	98.2	22	2	5	0.4	5.5	0.7
4	56	F	Right	Right	III	41	105	16	12	2	3.8	2.5	4.5
5	71	F	Left	Right	IV	41	97.2
6	53	F	Right	Right	III	40	84.3	12	16	1	4.9	1	6.3
7	63	F	Right	Right	III	40	100
8	54	F	Right	Right	IV	40	89.4	24	4	4	1.9	5	2.3
9	65	F	Right	Right	III	40	91.3	14	10	2	3.5	2.5	3.7
10	58	F	Left	Right	III	39	87.1	18	11	4.5	1.2	5	1.9
11	59	F	Left	Right	III	38	93.2	10	16	1	4.7	3	3.9
12	74	F	Right	Right	IV	38	69	22	0	5.5	0	5.5	0.7
13	72	F	Left	Right	IV	37	53	11	8	3.5	1.7	3.5	2.3
14	63	F	Right	Right	III	36	72.4	18	6	2.5	3	3	3.6
15	59	F	Right	Right	III	36	94.9	24	4	4	1.8	4	3
16	65	F	Right	Right	IV	34	85.7	22	2	3	2.5	3.5	2.9
17	52	F	Right	Right	III	34	84.8	18	10	4.5	1.3	4.5	1.5
18	73	F	Right	Right	III	33	93.9
19	51	F	Left	Right	III	33	100	26	0	3.5	2.1	3.5	3.6
20	67	M	Right	Right	III	31	86.8	38	2	5.5	2.8	7	2.9
21	62	M	Left	Right	III	30	89.2	28	10	3	5.3	4.5	5.9
22	67	F	Left	Right	IV	29	86.9	19	4	6	0	7	0
23	78	M	Left	Right	IV	25	93.8	22	10	4	2.8	6	2.5
24	57	F	Right	Right	III	24	100
25	53	F	Left	Right	III	23	98.1	18	8	3.5	2.1	5	2.1
26	59	F	Left	Right	III	21	92.3	10	16	1	4.7	2.5	4.4
27	59	F	Left	Right	IV	18	95	16	10	2.4	3.3	3	3.9
28	66	M	Right	Right	III	18	103.8	40	0	9.5	0	9	0.9
Mean	63					34	89.3	20	7.4	3.6	2.5	4.2	3

Table 2. Postprocedure QuickDASH and complications.

Patient	Age (y)	Gender	Follow-up (mo)	Postoperative QuickDASH	Complications	Comorbidities
Abbreviation: QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand.
1	59	F	42	90.9	Implant intolerance	Psychiatric disease, cervicobrachialgia
2	70	F	41	65.9
3	70	F	41	20
4	56	F	41	17.5
5	71	F	41
6	53	F	40	72.5		Rotator cuff tendinopathy
7	63	F	40	5
8	54	F	40	9.1		Rotator cuff tendinopathy
9	65	F	40	31.8
10	58	F	39	22.7
11	59	F	38	18.2
12	74	F	38	9.1
13	72	F	37	18.2	Suture rupture
14	63	F	36	38.6	Second metacarpal fracture
15	59	F	36	31.8
16	65	F	34	34.1
17	52	F	34	43.2
18	73	F	33
19	51	F	33	31.8
20	67	M	31	31.8
21	62	M	30	20.4
22	67	F	29	31.8
23	78	M	25	43.2		Rotator cuff tendinopathy, De Quervain tenosynovitis + basilar thumb arthritis on contralateral hand
24	57	F	24
25	53	F	23	18.2
26	59	F	21	50		Rheumatic disease with multiple osteoarthritis
27	59	F	18	29.5
Mean	63		34	32

Average strength results for the 24 patients who completed the evaluation were: 20 kg for grip, 3.6 kg for tip pinch, and 4.2 kg for key pinch. Comparing with standard tables for individuals of the same age, gender, and dominance, we calculated the differential for these values: grip strength differential was 7.4 kg, tip pinch was 2.5 kg, and key pinch was 3.0 kg ( Table 1 ).

Mean TSH was 89.3% (range, 53–105%), indicating that the metacarpal subsidence rate was 10.7%.

We had to remove the implant from one patient due to hardware intolerance. One patient had a second metacarpal fracture at 2 months of follow-up ( Fig. 5 ), treated with a cast without further complications. In one case, we had a failed suspensionplasty because of a suture rupture ( Fig. 6 ).

Fig. 5.

Radiograph showing second metacarpal fracture after suture-button suspensionplasty.

Fig. 6.

Radiograph showing suture rupture of the button suspensionplasty.

Discussion

Undoubtedly, the most commonly performed surgery worldwide for refractory thumb basilar arthritis is trapeziectomy and LRTI. 6 Nevertheless, none of the several techniques available showed superiority over the others in terms of pain reduction, function, complications, or first metacarpal subsidence. 6 17

Trapeziectomy and suspensionplasty of the first and index metacarpals with a suture button as we perform showed good functional and radiographic outcomes in the treatment of basilar thumb arthritis Eaton stages III and IV. None of our patients had Z deformities of the thumb, obviating the need for simultaneous metacarpophalangeal joint procedures.

Regarding complications, our rate was 10.7% even though the revision rate was only 3.6%.

At the present, no surgical procedure is proven to be superior but complications do differ between techniques. Trapeziectomy with LRTI seems associated with a slightly higher complication rate (between 10 and 19%) than trapeziectomy alone. 6 10 18 Regarding CMC arthrodesis, the nonunion rate may be as high as 21% and reoperations are very frequent. 18 In hematoma and distraction arthroplasty techniques, with the use of K-wires complication rates up to 14% may be expected. 12 19 Avant et al found similar complication rates between SBS and APL suspensionplasties. 3 When comparing suspensionplasty and pyrolytic carbon hemiarthroplasty, Vitale et al found that the last group had an increased risk of adverse events (38 vs. 12.5%). 20 DeGeorge et al did not find significant differences in the incidence of postoperative complications between trapeziectomy with FCR–APL tendon transfer with and without SBS. 1 One study compared trapeziectomy alone and with SBS using Mini TightRope CMC and found fewer complications in the trapeziectomy group. 21 Roman et al had a mean complications rate of 10.5% with trapeziectomy and SBS, 22 which is very close to ours. Also, revision rates of 5.3 to 6.6% are described for arthoscopic 23 and open trapeziectomy and SBS. 12 21 22 23 To summarize, we believe that trapeziectomy alone has the lowest complication rate, with the disadvantage of loss of strength making the trapeziectomy with SBS a better option.

The index metacarpal fractures described in this procedure may be a possible result of an increased angle of fixation or overtensioning of the wire, which may restrict the natural motion of the thumb metacarpal, placing increased forces on the second metacarpal. 13 Osteolysis around the suture is another explanation. 13 Additionally, we do not use the C-Ring aiming guide because it is not available in the standard set and it needs to be ordered separately, adding more cost to the surgery. This may have led to incorrect positioning of the passing wire in our patient and/or to overdrilling to achieve a centered position. All described cases of fracture presented at least 6 weeks after the procedure 13 15 24 which points to a stress fracture.

Like in other series, 12 21 we had a case of implant intolerance, due to bad positioning of the button. It was our first case; we now know that placing the suture button beneath a portion of the radial aspect of the abductor pollicis brevis belly decreases the chance of postoperative hardware prominence. 14 One patient showed a failure of the suture suspensionplasty, a complication already described in a case report. In this article, the authors hypothesized that this may have resulted from continuous abrasion of the suture knot in the tunnel, drilled in a partially reduced position. 16 In our case, the break did not affect the outcome, since the patient was asymptomatic and refused removal of the button. We had no cases of infection, complex regional pain syndrome nor dorsal radial sensory nerve injuries, as described in other series.

Concerning suture-button orientation, most studies show that obliquely guiding the Mini TightRope provides a vector of tension that is better able to prevent subsidence. 9 Hozack et al also found that a low angle trajectory of the suture, aiming the meta-diaphyseal junction of the second metacarpal, provided a more stable construct regarding subsidence and angular motion. 25 In our experience and regarding our low first metacarpal subsidence rate, the best position of the suture is in a 40-degree angle, from the thumb metacarpal base.

It is also important to provide enough tension to prevent subsidence of the first ray while avoiding overtensioning, which may cause impingement between the first and second metacarpals and decrease the thumb range of motion. We use the interposition of a small clamp (2–3 mm) to avoid overtightening of the suture.

About the first ray subsidence, we had a mean of 10.7% of subsidence. This is a lower result than that described in the previous series in which TSH decrease ranged from 25 to 33%. 5 9 12 Trapezial space preservation is thought to be crucial for thumb strength, preventing thumb instability, collapse, and consequently, pain in a newly formed metacarpal–scaphoid joint. 2 25 26

Some studies show that comparing 6 weeks, 12 weeks, and 1 year postoperatively radiographs, there is no change in TSH with suture-button systems. 22 Since these results are sustained over time, an increase in strength is expected with a longer follow-up. 12 This phenomenon is not true for trapeziectomy alone or even LRTI with FCR tendon, showing the superiority of the SBS technique. 22 26 Furthermore, there is some evidence that a good outcome is reached, in terms of strength and range of motion, as long as the metacarpal subsidence is not complete and there is not a Z-deformity of the thumb. 12

Regarding function, we had a mean QuickDASH of 32 points, which is a functional score meaning that most of our patients were able to do their activities of daily living without impairment. These outcomes are similar to those found in the Cochrane review for trapeziectomy alone and with LRTI, in which the mean DASH score was 31 for both procedures. 6 Also, in a retrospective study, with a mean 9.5 months of follow-up, with a sample size similar to ours, Avant et al had an average of 37 points in postsurgical DASH score. 3 Yao and Cheah, with 5 years of follow up had a mean postoperative QuickDASH score of 9.2, 12 so we could expect improvements with a longer follow-up.

We compared our postprocedure strength values to standardized tables, having good outcomes in this field. The mean differential strength for grip was 7.4 kg, for tip pinch was 2.5 kg, and key pinch was 3 kg. Our results are in line with other case series published using the same surgical procedure. Yao and Cheah had a mean postsurgical grip strength of 19 kg and a mean tip pinch of 6 kg, with 64 months of follow up. 12 Roman et al obtained 24, 4.8, and 3.8 kg for grip, key, and tip pinch strengths, respectively, at 20 months of follow-up. 22 Avant et al, in a comparative study with a mean follow-up of 18 months, presented identical data: 17 kg for grip strength, 4 kg for tip pinch, and 5 kg for key pinch, with no differences between suture-button and APL suspensionplasties. 3 Parry and Kakar, using a double system of Mini TightRope had overlapping results with 24 kg for grip strength and 6 kg for tip pinch, with 17 months of follow-up. 5

Recent studies show the advantage of dual suture-button fixation, compared with a single-one or ligament reconstruction and tendon interposition procedures, in terms of pain relief, increased strength, preserved range of motion, and maintenance of TSH. 2 5

According to our experience, SBS has several advantages. First, in our department, the surgical time was shortened from near 90 minutes in LRTI to 60 minutes with SBS. Also, the risk of complications and morbidity is less when a tendon graft is used. Another benefit is that there is no need for immobilization with a cast, like almost all other surgical options. This leads to a faster recovery because our patients can start physical therapy as soon as 15 days after surgery, decreasing the risk of stiffness and chronic pain.

We also believe that a cost analysis comparing the price of the implant with the expenses of an extended rehabilitation program after a long period of immobilization would be in favor of the use of trapeziectomy and SBS. Having in mind that implant price oscillates between 250 27 and 495 9 21 dollars and that physical/occupational therapy after thumb CMC procedures cost rounds 85 to 95 dollars per patient and session, 28 we think that the early rehabilitation afforded by the SBS device represents an accelerated recovery, compared with the techniques that imply immobilization, requiring fewer sessions until full recovery and so being cost-effective.

This study has some limitations. First, it is a retrospective study without a comparative group. Second, we do not have preoperative QuickDASH scores to compare with postprocedure results. We also were unable to obtain strength evaluation for four patients and postoperative QuickDASH in three cases. Also, some patients had upper arm comorbidities that affected the evaluation of the hand itself, modifying the QuickDASH results. Another potential bias was the trapezial height measurement which is dependent on the radiograph views.

Conclusion

Trapeziectomy and SBS functional results are similar to previously described techniques, including LRTI, with less subsidence of the first ray. This feature combined with the possibility of early mobilization can lead to excellent long-term outcomes in terms of function and strength. Furthermore, the complication rate is identical or even lower than other surgical approaches. So, we conclude that this procedure is a safe and promising option in the treatment of end-stage basilar thumb arthritis.