Modified Graner's Technique with\without Vascularized Capitate Lengthening for Kienböck Disease Stage IIIb—A Pilot Study

Abstract

Background  Intercarpal fusions are used to treat stage IIIb Kienböck disease. They increase force transfer across the radioscaphoid articulation with predisposition to arthritis.

Description of Technique  This technique is excision of lunate followed by proximal transfer of capitate, with scaphocapitate and triquetrocapitate fusion to increase area of load transfer mimicking wrist hemiarthroplasty. Our purpose is to evaluate mid-term results of this technique.

Patients and Methods  A prospective case series study was conducted on 11 patients with stage IIIb and IIIc. In seven cases, transfer of the capitate was performed by osteotomizing the capitate just distal to its waist, proximal migration to replace the excised lunate then bone grafting. In four cases, proximal transfer of vascularized pedicled capitate was done. Clinical outcome measures included pain (visual analog scale), grip strength, range of motion, and functional evaluation by modified Mayo wrist score and scoring system of Evans. Radiological outcome measures included healing of fusion mass, progression of the disease, and occurrence of avascular necrosis to the capitate.

Results  Follow-up period averaged 54 months. Scaphocapitate fusion healing averaged 11 weeks. Union of the lengthened capitate occurred in 10 patients only. There was postoperative improvement in pain scores, grip, Evans, and modified Mayo wrist score. There was postoperative decrease in wrist flexion and extension. One patient showed resorption of the capitate head with progressive radioscaphoid arthritis-necessitated wrist fusion.

Conclusion  The mid-term results of this technique may be satisfactory due to low incidence of degenerative arthritis in the radioscaphoid joint. However, longer follow-up with recruiting larger number of patients is needed.

Lunate avascular necrosis was reported by Robert Kienböck in 1910 ¹ and termed as Kienböck disease. Since this report, the etiology and treatment have been controversial. A variety of procedures have been provided, which include lunate revascularization, lunate excision and replacement, limited carpal fusions, joint leveling procedures, proximal carpectomy, and wrist fusion. ²

Limited wrist fusion is the most common procedure used to treat Kienböck stage IIIb. Although early reports were largely favorable, more recent studies suggest an increasing awareness of the limitations and potential complications of limited wrist fusion. ^{3 4}

Recently, several authors ^{5 6 7} have investigated the biomechanical effect of scaphotrapeziotrapezoid (STT) fusion on the wrist. They found that STT fusion reduced load in the lunate fossa of the radius and increased load in the scaphoid fossa of the radius. The concern brought out by these results is that the triangular fibrocartilage complex fossa and radiolunate fossa will be unloaded, but the increased radio-scaphoid forces may lead to premature wear of this fossa. This change in the mechanics of the wrist with the resultant nonphysiological loading is a possible explanation.

Graner's procedure is logical as it aims at creating a new radiocarpal articulation, either by the fusion of the lunate with all carpus except trapezium (Graner I) or by replacing the lunate with the head of the capitate with pancarpal fusion except trapezium (Graner II) or replacing the lunate with the head of the capitate with just capitohamate fusion (Graner III). ⁸ We suppose that the creation of a new radiocapitate articulation as in Graner II and III procedure may decrease overloading of the radioscaphoid joint following STT or scaphocapitate (SC) arthrodesis in Kienböck disease.

In comparison to Takase and Imakiire ⁹ who did modify Graner's procedure with nonvascularized capitate osteotomy and pancarpal arthrodesis, we modified Graner's procedure by performing excision of the lunate followed by proximal transfer of the capitate with the fusion of the elongated capitate to both the scaphoid and the triquetrum. This newly formed fusion mass has the shape of an inverted T mimicking an autogenous wrist hemiarthroplasty. In the later cases, we conducted a vascularized proximal transfer of the capitate based on the dorsal branch of anterior interosseous artery as devised by Lu et al. ¹⁰ They excised the lunate and proximally shifted the vascularized capitate for advanced Kienböck disease.

The presumed advantages of the index technique are that it may decrease the nonphysiological loading of radioscaphoid column in limited wrist arthrodesis by creating a weight sharing radiocapitate articulation. In addition, excision of the lunate will remove a potential source of pain. Third, transfer of the capitate in a vascularized fashion will confer a lower risk for avascular necrosis and higher integration rate for the formed fusion mass.

The purpose of this study was to assess the mid-term functional and radiological results of autogenous wrist hemiarthroplasty (modified Graner's technique) for Kienböck disease stage (IIIb and IIIc).

Patients and Methods

Prospective case series study comprised of 11 patients with stage IIIb and IIIc Kienböck disease to whom lunate excision and SC fusion with capitate lengthening (four cases of them with vascularized capitate pedicle) were conducted in our institute between October 2014 and September 2016. The exclusion criteria were radiocarpal arthritis or open physis.

The average age of the patients was 33 years (range: 18–53 years). Five patients were females and six were males. Six patients were in stage IIIB and five were stage IIIC. The dominant hand was affected in five patients and the nondominant hand was affected in six patients. Six patients were manual workers, four house makers, and one student. All patients had given informed consent, and the study protocol was approved by the Institutional Review Board in our university.

The clinical outcome measurements included analysis of pain using visual analog scale (VAS), ¹¹ assessment of wrist motion using a goniometer, and assessment of grip strength using a Jamar dynamometer. Range of motion (ROM) and grip strength were expressed as a percentage of the uninvolved side. The ability to perform different daily activities including work performance was analyzed using modified Mayo wrist score (MMWS) ¹² and scoring system of The Evans ¹³ ( Table 1 ).

Table 1. The scoring system of Evans.

Criterion	Score(points)
Little or no discomfort in the wrist	1
Grip > 70% healthy side	1
Flexion/extension arc > 60 degrees	1
Return to previous occupation	1

Evans et al ¹³ considered good (4 points), fair (3 points), poor (2 points), or very poor (1 or 0 points).

Radiologically, the patients were evaluated by plain X-rays (posteroanterior and lateral views) of the wrist preoperatively for the assessment of the stage of the disease according to Lichtman classification, ¹⁴ radioscaphoid angle, and the presence of radiocarpal arthritis. All patients were scanned preoperatively by magnetic resonance imaging to confirm the diagnosis.

Postoperatively, X-rays were evaluated to assess healing of the transferred capitate and SC fusion sites, progression of arthritis, or development of AVN of the capitate. CT was obtained in most of the patients pre- and postoperatively to confirm findings of the X-rays.

Surgical Technique

Seven cases were treated with osteotomy of the waist of the capitate, proximal transfer of the head capitate, and bone grafting. In the last four cases, we transferred the capitate based on the dorsal branch of the anterior interosseous artery as described by Lu et al . ¹⁰

The procedure was performed under tourniquet control without exsanguination; a 4-cm longitudinal dorsal incision was made overlying the capitate. The extensor tendons of the fourth compartment were retracted after opening of extensor retinaculum and the capsule of the midcarpal joint was exposed. In the vascularized cases, two dorsal longitudinal incisions of the capsule of the midcarpal joint 1.5 cm apart were made (to include the dorsal branch of anterior interosseous artery that can be identified at the floor of fourth compartment) and one transverse limb 2 mm proximal to capitate base (carpo metacarpal joint easily identified by needle) with fluoroscopic confirmation. The capitate transverse osteotomy was done 2 mm proximal to its base to ensure inclusion of the vascular pedicle to the capitate in the proximal segment. Release of the volar ligaments was followed while keeping the dorsal capsular pedicled flap attached to proximal capitate. The proximal part of the capitate was elevated with the dorsal capsule, then the capsular incision was extended and the distal joint surface of the lunate was exposed. The tourniquet was released to check capitate vascularity.

The necrotic lunate was excised. Complete denuding of articular surface between scaphoid and capitate, hamate facing the capitate, and triquetrum facing the lunate were performed. The mobile segment of the capitate could be easily transferred more proximally to fit into the lunate fossa of the radius. The triquetrum was transfixed to the capitate with Kirschner wires (K-wires). Corticocancellous BG from the distal radius was inserted to fill the capitate void and the denuded cartilage. Multiple K-wires were used to stabilize the SC construct. In some cases, SC fusion was fixed with a Herbert screw. Intraoperative fluoroscopy was used to check the alignment and fixation.

The capsule with dorsal intercarpal ligament and extensor retinaculum was closed with absorbable suture. The tourniquet was released, and closure of the skin was conducted ( Fig. 1A–G ).

Fig. 1.

Operative techniques ( A ) Dorsal incision, capsular flap, excision of necrotic lunate. ( B ) Capitate osteotomy and lengthening. ( C ) Bone grafting of the capitates void. ( D ) Fixation of capitate osteotomy and carpal fusions with Kirschner wires. ( E ) Closure of the capsular flap. ( F ) Deflation of tourniquet to check vascularity of the pedicled capitates (one of the vascularized groups).

Rehabilitation

Postoperatively, a short-arm splint was applied and light activity with the fingers was encouraged. Stitches removed after 2 weeks, and the splints were removed 6 to 8 weeks after surgery. The patients were seen every 2 weeks till healing of the osteotomy and fusion sites, and then they were seen monthly to evaluate clinical improvement and signs of capitate devascularization or collapse. Physical rehabilitation started following union and pin removal with removable short arm splint applied. Rehabilitation comprised gradual ROM exercises, grip strength, and pain control modalities. Strenuous exercises were allowed after 3 months.

Postoperative changes in VAS, ROM, grip strength, MMWS, and Evans score in addition to radiocarpal joint changes were compared with preoperative values using the paired-samples Student's t -test. The level of significance was set at p ≤ 0.05. All data analysis was calculated with statistical software SPSS version 15 for Windows (SPSS Inc, Chicago, IL).

Results

Follow-up period averaged 4.5 years (range: 49–60 months). Union of the SC fusion was obtained in all 11 cases in an average of 11 weeks (range: 8–12 weeks). Healing of the capitate lengthening osteotomy occurred in 10 cases and resorption occurred in 1 case. Integration of the capitate with the triquetrum occurred in all cases except three cases (including the case with resorbed proximal capitate). Table 2 summarizes the functional results of the patients.

Table 2. The functional results of the patients.

Clinically	Preoperatively	1-year FU	Final FU	p -value (between preop and final FU)
Pain (VAS)	4.6 (3–7)	2.3 (1–6)	1.2 (0–6)	< 0.001
ROM (% of contralateral side)	Flex 59 (35–80%) Ext 75.4 (60–90%)	Flex 42.3 (30–60%) Ext 61.8 (40–80%)	Flex 44 (30–60%) Ext 63.1 (40–80%)	< 0.001 < 0.05
Grip strength (%of contralateral side)	60.5% (50–80%)	70.9% (50–90%)	74.5% (60–90%)	< 0.001
Modified Mayo wrist score	50.5 (25–60) poor	58.6 (30–70%)	68.6 (50–80) fair	< 0.05
Scoring System of Evans	1.8 (0–3) Fair 1 Poor 7 Very poor 3	2.7 (1–4) Good 1 Fair 6 Poor 2 Very poor 1	3 (1–4) Good 4 Fair 5 Poor 1 Very poor1	< 0.001

Abbreviations: Ext, extension; Flex, flexion; FU, follow-up; ROM, range of motion; VAS, visual analog scale.

The VAS improved significantly from 4.6 preoperatively to 2.3 at 1 year follow-up and 1.2 at the final follow-up ( p  < 0.001).

In comparison to contralateral side, flexion decreased from 59% (35–80) preoperatively to 42.3% (30–60) postoperatively and to 44% (35–60) at final follow-up ( p  < 0.001). Extension decreased from 75.4% (60–90) preoperatively to 61.8% (40–80) postoperatively and to 63.1% at final follow-up ( p  < 0.05)

Grip strength improved significantly from 60.5% (50–80) of the contralateral side preoperatively to 70.9% (50–90) postoperatively and to 74.5% (60–90) at the final follow-up ( p  < 0.001)

Functional outcome as reflected by MMWS improved from preoperative value of 50.5 (25–65) to 58.6 (30–70) postoperatively and to 62.7(45–75) at final follow-up with ( p  < 0.05). Also, Evans score improved from 1.8 (average, 1–3) preoperatively to 2.7 (average, 1–4) and to 2.9 (average, 1–4) at final follow-up with ( p  < 0.001).

We found that there were no significant changes from 1 year postoperative to the final follow-up results except in Evans scoring that improved significantly.

Radiocarpal joint arthritis was not detected radiographically through the whole follow-up period except in one case which developed advanced osteoarthritis following capitate resorption and collapse.

Complications

One patient developed superficial wound infection that healed on empirical antibiotics for 1 week. One patient developed resorption of the capitate with progressive radiocarpal arthritis. Two patients had incomplete integration of the capitate with the fusion mass (nonunion between the lengthened capitate and the triquetrum).

Discussion

Graner's procedure was used previously for Kienböck disease. The pendulum shifted to STT or SC fusion mostly due to simpler surgical technique with similar (if not better) functional outcome. The described technique can be considered a midway between both techniques.

Interestingly, the worst-case scenario in this technique will be resorption or AVN of the head of the capitate. Both scenarios are tolerable from the medical point of view. The former situation can mimic SC fusion with lunate excision. In the later, the patient will have SC fusion with necrotic head of capitate instead of necrotic lunate.

This study has some limitations. First, this is not a randomized controlled study. Second, measurement bias possibility as the observer who collected and analyzed data was not masked. Third, this is a medium-term study (may be too early to determine if the procedure decreases the risk of arthritis at the radioscaphoid articulation). Finally, the number of patients with vascularized capitate lengthening was small to delineate definitely the benefit of adding vascularized capitate transfer.

In our study, mean VAS was 1.2 at final follow-up. In the study of Facca et al, ⁸ Graner II procedure was conducted on four patients with late stage Kienböck disease. The patients were followed up for 25 years. Their mean postoperative VAS was 3.25. The series of Takase and Imakiire ⁹ (modified Graner's procedure) included 15 patients with a mean follow-up of 5 years. They concluded a good result in 11 cases, moderate result in 2 cases, and bad result in 2 cases. All the X-rays showed arthritis with little or no clinical expression. Hierner and Wilhelm ¹⁵ also used transverse capitate osteotomy with progressive lengthening by callotasis. Their results were favorable; decreased pain, conservation of wrist motion, and increased wrist strength.

In the study of Lu et al, ¹⁰ 23 cases were followed up for 1 year with no pain except residual pain existed in labor work.

The technique of Lu et al ¹⁰ is innovative and simpler than our technique as they did not combine SC fusion with the vascularized capitate transposition. We believe that adding the SC fusion represents a (back up) midcarpal fusion if the capitate transposition failed. Moreover, we assume it will provide more even distribution of load on both the radiocapitate and radioscaphoid articulations. We are skeptical that the technique of Lu et al ¹⁰ may overload the radiocapitate articulation in the long term (their follow-up is 1 year).

In the study of Minami et al, ⁴ 23 STT fusions were done for Kienböck disease with excision of the lunate. A follow-up period averaged 84 months with 5 patients who had postoperative pain; 4 patients were dissatisfied with the results.

Tambe et al ¹⁶ compared total versus limited wrist fusion in advanced Kienböck disease with average follow-up of 66.8 months. In the limited carpal fusion group (radiolunate fusions, STT fusions, SC, and capitohamate), VAS improved only from 8.1 to 7.2 postoperatively.

Grip strength improved significantly to reach 75% of the strength of the unaffected side. In the study of Facca et al, ⁸ the wrist strength was 80.9% of the opposite side. In the study of Takase and Imakiire, ⁹ the grip power of the affected hand reached on average 80% compared with the contralateral side after 12 months from surgery. In study of Hierner and Wilhelm, ¹⁵ the mean power grip and pinch grip strength improved from 15.4 and 7.2 kg to 21.8 and 7.2 kg, respectively, at the 5-year follow-up. In the study of Lu et al, ¹⁰ the grip power of the affected hand reached on average 70% compared with the contralateral side. In Watson et al ¹⁷ study of long-term follow-up after STT fusion, grip strength was 68% of the nonoperated side. Arthritis developed in 1.8% of their patients and the overall complication rate was 13.4%. It is speculated that these differences may be due to differences in hand-dominance distribution among different series, and/or due to differences in follow-up periods between these studies. According to Tang and Imbriglia, ¹⁸ it takes more than 2 years before grip strength to plateau. However, what is consistent from the results of all studies that the postoperative grip strength is expected to be less than the contralateral side at the intermediate-term follow-up.

Regarding ROM, flexion decreased to 44% and extension to 63% of contralateral side. In the study of Facca et al, ⁸ the range of movement was half of the opposite side. In the study of Takase and Imakiire, ⁹ there was moderate limitation of wrist ROM that did not affect the usual level of activity. In study of Hierner and Wilhelm, ¹⁵ the mean ROM at the 2-year interval was 29 degrees for extension and 27 for flexion. In the study of Lu et al, ¹⁰ the arc of motion ranged on average from 35 degrees of flexion to 45 degrees of extension. In Watson et al ¹⁷ study, the ROM was 69% of flexion and 66% of extension of the nonoperated side.

This salvage type of surgery leads to expected decrease in wrist motion. However, the absolute ROM in our patients is comparable to that reported in previous studies.

After resorption of the transferred capitate in one case, we opted to perform a vascularized transfer of the capitate according to the technique of Lu et al. ¹⁰ We believe this may decrease the possibility for AVN of the capitate and may increase the integration of the capitate in the fusion mass. No AVN or resorption of the head of the capitate occurred in the 4 cases of the vascularized group in comparison to one case of capitate resorption out of 7 in the nonvascularized group. However, the limited number in both groups does not allow us to draw solid conclusion.

There was no evidence of nonunion in this series compared with a high rate of nonunion in STT procedure that was 24% in McAuliffe et al ¹⁹ study. We attribute this to the wider surface area available for fusion in this technique. Also, no progressive radiocarpal arthritis developed in our study except in one case, while in study of Kleinman and Carroll, ²⁰ 9/24 developed progressive arthrosis following STT fusion.

Cases are presented in Figs. 2 to 4 .

Fig. 2.

Case 1—A 24-year-old manual worker presented with left-hand wrist pain. ( A, B ) Preoperative X-ray posteroanterior view. magnetic resonance imaging, ( C, D ) 2 months follow-up X-ray, ( E, F ) 8 months follow-up X-ray, ( G, H ) clinical range of motion.

Fig. 4.

Case 3—A 42-year-old male manual worker , heavy smoker, right hand dominant hand stage III. ( A , B ) Postoperative X-ray, ( C, D ) 10 weeks follow-up, ( E, F ) 11-month follow-up.

Fig. 3.

( A, B ) Case 2—A 40-year-old manual worker presented with left-hand wrist pain. Five years follow-up.