Arthroscopic versus open cancellous bone grafting for scaphoid delayed/nonunion in adults (SCOPE-OUT): study protocol for a randomized clinical trial

A scaphoid fracture is the most common injury to the carpal bones. The incidence is 107–151/100.000 per year and fractures are predominantly sustained by males in their twenties [1]. Scaphoid non-union is defined as a lack of healing 6 months after injury and develops in 5–25% of cases after non-operative treatment [2, 3]. Delayed union is defined as incomplete healing 2–6 months after injury. However, some potential for a union probably exists, especially in nondisplaced fractures, otherwise this condition is associated with a transition into a persistent non-union [4]. The risk of non-union increases with delayed diagnosis and treatment, displaced fractures, proximal pole fractures, smoking, poor vascularity and advancing age [5‐8].

The scaphoid is primarily covered with cartilage and has a retrograde blood supply. The dorsal branch of the radial artery accounts for 80% and a separate volar branch for 20% of the extramedullary blood supply. The proximal pole is only supplied by the intramedullary flow. Compromised blood supply can explain the potential of non-union and avascular necrosis of the proximal pole. The healing process can be complicated by volar angulation of the fracture leading to humpback deformity. This will disrupt carpal kinematics and result in lunate instability and dorsal intercalated segment instability (DISI). Untreated scaphoid non-union can lead to degenerative changes, called scaphoid non-union advanced collapse (SNAC), and irreversible impairment such as pain and altered hand function.

X-ray is commonly applied to evaluate the scaphoid, although CT scans are reportedly superior in terms of displacement, angulation and union [9]. Different measurements to describe the angulation and deformity of the scaphoid are suggested. The height length ratio (HLR) and dorsal cortical angle (DCA) (Fig. 1) are found to be the most reliable measurements [10, 11].

Surgical treatment of scaphoid delayed/non-union is technically demanding and often results in a long period with a supportive bandage until union is established. Current treatment strategies for delayed union and non-union include vascularized or non-vascularized bone graft with internal fixation. Kirchner wires or screws have been the gold standard for fixation. A meta-analysis found an average of 94% union with screw fixation and 77% with K-wire fixation in unstable non-union [12]. Open CC graft reconstruction is often preferred [13, 14]. However, there is an increased donor site morbidity from the iliac crest, and the outcome is not superior to distal radius C graft [15]. Arthroscopic reconstruction with C chips and internal fixation is predominantly applied in delayed union and stable non-union [16‐18]. The advantages of arthroscopy include thorough wrist assessment, evaluation of concomitant ligamentous injury and minimal trauma to the ligament structures, joint capsule and the tenuous blood supply [19]. Recent studies have reported an 86–100% union rate, also in patients with cystic changes and/or translation > 2 mm, and/or humpback deformity [20‐24]. A retrospective study, with 62 cases, found union in 96% arthroscopic group vs. 97% in open surgery with no difference in the functional results, however, time to union was not recorded [16]. No RCT comparing arthroscopic or conventional open C graft technique have been performed.

The objective of this study protocol is to describe the methodology for a randomized controlled trial comparing time to union and functional outcome scores of arthroscopic assisted C chips reconstruction or open C graft reconstruction for scaphoid delayed/non-union. The standard protocol items recommendations for interventional trials (SPIRIT) statement 2013 have been followed for the completion of this protocol (Additional file 1) and the items from the WHO trial registry are found in within the protocol.

A total of 88 patients with scaphoid delayed/non-union are randomized to either:

All Danish citizens aged 18–68 years with scaphoid delayed/non-union referred to the hand surgy unit, Herlev and Gentofte University Hospital, will be offered participation in the study. Other hand unit departments in the Capital Region of Denmark will forward their referrals to our department for inclusion. The physician will review the medical records and assess whether the patients fulfil the inclusion/exclusion criteria. If the patient is found eligible, they will receive oral and written information of the study (Fig. 2) and the possibility to bring a bystander for the preparation consultation. The patients will be given a minimum of 24 h deliberation period. At the preparation consultation with the primary investigator, the patient will once again receive oral and written information, and trial risks of the study, and subsequently the consent will be obtained. The patients will simultaneously be allocated to one of the treatment groups. Patients can withdraw their consent at any time. The Helsinki Declaration will be followed [25]. The informed consent will give the primary investigator access to information about sex, age, comorbidity, trauma, time from fracture to treatment, history of smoking, occupation, hand dominance, pain, ongoing workers’ compensation and insurance claim. Baseline range of motion, grip strength, Key-pinch strength, pain in rest and activity (VAS) and Q-DASH will be recorded (Fig. 2).

Patients will undergo a clinical examination and CT scan of the wrist to describe angulation (HLR and DCA), displacement, localization of non-union, presence of cysts and degenerative changes. All patients with involvement of the proximal pole will undergo gadolinium-enhanced MRI to clarify vascularity, but the final assessment of avascular necrosis of the proximal pole will be performed perioperatively. If punctate bleeding after the tourniquet being off for at least 5 min of expectation cannot be visible, the patient will then be excluded from the study and will be simultaneously operated on with a vascular bone graft or a salvage procedure.

The operations will be performed at the Orthopedic Department at Herlev and Gentofte University-Hospital, Denmark, which is the largest referral hand unit in Denmark. Annually 1–25 patients are treated for scaphoid delayed/non-union. With forwarded referrals included, we estimate a 3-year inclusion period from 01–01-2023 to 01–01-2026.

Based on the sample size calculation, a total number of 88 patients will be allocated into two groups of equal size.

The randomization is done in the outpatient clinic and the patients will be informed about the operative treatment. The randomization is done using an irreversible application in Research Electronic Data Capture (REDCap). Patients will be allocated in a 1:1 ratio by block randomization, stratified for proximal pole fracture (yes /no), dislocation (> / < 2 mm) and smoking (yes/ no). An independent statistician generated the randomization sequence for REDCap, and thus the allocation is concealed to the research staff and physicians.

In this observer-blinded RCT, union is assessed by a blinded musculoskeletal radiologist. QDASH is a patient-reported survey, without the involvement of surgeons or research staff. Other secondary outcomes will be measured by an independent observer. The study will not be blinded to the operating theatre staff, surgeons, physiotherapists or patients.

All patients will be operated on by 4 highly experienced hand surgeons of the same experience level. Two surgeons will perform the open technique and 2 surgeons the arthroscopic technique. The surgeons are using the technique which they are familiar with to avoid a learning curve affecting the outcomes.

Patients are awake in a supine position with the arm anaesthetized as standard. The arm is attached to a wrist traction tower with vertical traction of 5–8 kg force through plastic finger trap devices. An arm tourniquet is applied. Through 3/4 and 4/5 portals using dry-arthroscopy, the radio-carpal joint is inspected for cartilage damage, synovitis and ligament injuries. Mid-carpal radial (MCR) and mid-carpal ulnar (MCU) portals are applied to examine the non-union site. The non-union site is debrided with a shaver until healthy-looking bone with punctate bleeding. The C chips graft is harvested from the ipsilateral distal radius through a 2 cm incision with a biopsy needle, using multiple biopsy probes. After arthroscopy, the wrist is taken out of traction and the non-union is reduced under the C-arm image intensifier and a 1.2-mm K-wire is inserted percutaneously, either in a retrograde or anterograde manner. The scaphoid is prepared with a drill, the length is measured and the graft is inserted with a trochar through the MCR portal. An Acutrak mini compression screw (Acumed^tm, Hillsboro, USA) is inserted. The final position is confirmed under fluoroscopy. The non-union site is inspected, and spilled bone graft material is impacted in the gap. Finally, the skin is sutured, and a bandage is applied.

Patients are under general anaesthesia with the arm anaesthetized with a regional block anaesthesia. A longitudinal volar incision, lateral to the flexor carpi radialis (FCR) curved distal over the scaphoid tuberosity is made. The FCR is retracted to the ulnar side and the volar capsule and ligaments are incised longitudinally to expose the scaphoid. The non-union is debrided to a healthy-appearing bone. A C graft is harvested from the ipsilateral iliac crest through a 4-cm incision. The cancellous graft is prepared and placed in the cavity. Under fluoroscopy, a 1.2-mm K-wire is inserted in a retrograde or anterograde manner. After drilling, the length is measured and an Acutrak mini compression screw (Acumed^tm, Hillsboro, USA) is inserted. The final position is confirmed under fluoroscopy. The capsule and ligaments are repaired, the skin is sutured and the immobilizing bandage is applied.

Rehabilitation will be identical between the groups. Stitches are removed 2 weeks postoperatively. Patients will be provided a thumb/wrist splint for 2 weeks, followed by the application of removable orthosis for another 4 weeks allowing the beginning of light non-weight-bearing exercises. The wrist will be allowed mobilization without restrictions if union is established on a postoperative CT scan and with the absence of scaphoid tenderness on the clinical examination. Return to work will be accepted when the union is established.

We will record all complications related to the operative treatment (tendon, ligament, nerve or arterial injury, infection, complex regional pain syndrome, haematoma or hardware failure). Reoperations, defined as revision surgery and secondary surgery due to no union will be noted.

All patients will be followed for 10 years. Union will be measured in 2 weeks intervals 6–16 weeks after operative treatment. Clinical outcomes and patient-reported outcomes will be measured after 1.5, 3, 6, 12 and 24 months. Online questionnaires will be sent after 5 and 10 years (Fig. 2).

The patients evaluated for inclusion will be reported in a consort diagram (Fig. 3). The statistical analysis will be conducted on an intention-to-treat basis. Outcome scores will be analysed according to the initial study group assignment regardless of possible cross-over. If crossover does occur, a secondary per-protocol analysis will be conducted.

Patients drop-out/loss to follow-up will be recorded and the reason will be noted. The patient is included in the analysis with their latest follow-up if at least a 6-week evaluation is available. Patients who undergo revision/secondary surgery will remain in the study and will be followed according to the index procedure. Their results before and after the revision procedure are included in the analysis. If challenges in the trial conduct occur, they will be presented by the primary investigator every second week to the research group. Any protocol modifications will be communicated to all investigators, observers and trial registries immediately. No regular audit is planned, apart from possible independent inspection visits by the Data Protection Agency and the National Committee on Health Research Ethics.

The operative techniques are used routinely in our department and are considered safe, therefore interim analysis will not be conducted. At follow-up, patients will be asked about complications, and journal notes about hospital contacts related to the surgical procedure are recorded.

Today, operative treatment for scaphoid delayed union or non-union is performed with the conventional open technique or the arthroscopic technique based on the preference of the surgeon. Based on non-randomized studies both techniques appear safe and with no major differences in functional outcome or adverse events [16, 35]. There is, however, disagreement regarding the treatment of patients with a large deformity. Hegazy et al. [36] and Sayegh and Strauch [13] found superior correction and superior ROM, Q-DASH and Mayo wrist scores in patients treated with a structural graft compared with patients treated with a non-structural graft. Other studies found no difference in deformity correction [15, 37, 38].

The arthroscopic technique is potentially less invasive with minimal donor site morbidity and potentially faster time to union [15] because of minimal trauma to the ligament structures, joint capsule and the tenuous blood supply [19]. It may also have advantageous osteogenic properties compared to a structural graft [39]. Differences in time to union between the open technique and the arthroscopic technique have, to our knowledge, never been investigated.

X-ray is commonly applied to monitor union after operative treatment for scaphoid delayed/non-union. Consolidation on at least three out of four views is a sign of union [40]. X-ray is not reliable to determine fracture displacement and deformity [41], and the wrong position of the hand and “overlining” in the fracture line can complicate the evaluation of union. CT scans are reportedly more accurate and reliable in the assessment of union [26, 42‐44]. We aim to provide a more accurate measurement of time to union with repeated CT scans.

The study differentiates from standard treatment with up to 7 scheduled CT scans in patients with prolonged union time, compared to 1–3 CT scans and 1–3 X-rays as a standard follow-up. The background radiation in Denmark is 3 mSv each year. Each CT scan is performed with the involved wrist above the head with a mean radiation dose of 0.03 mSv. The wrist contains minimal radiation-sensitive red bone marrow and the cumulative radiation dose is maximal 0.21 mSv, (category IIA, international commission on radiation Protection), corresponding to 21 days of background radiation [45]. The radiation dose of each scaphoid X-ray is 0.0002 mSv.

We aim to decrease observer and selection bias through blinding and randomization, stratified for tobacco-smoking habits, fracture displacement (> / < 2 mm) and proximal pole fractures. By adding 20% to the sample size, we compensate for dropouts and attempt to avoid Type-II errors. Potential dropouts will be minimized with permission to contact patients by mail or phone if they do not show up for follow up-visit or if data is missing.

The patients will not be blinded to their treatment allocation. Sham surgery can be used to blind patients regarding skin incisions and subsequently the surgical technique. However, all patients treated with open bone grafting using a cancellous graft taken from the iliac crest will be operated on in general anaesthesia, while arthroscopically treated patients undergoing the treatment in regional block anaesthesia. Furthermore, the use of graft from the iliac crest in one group and not the other makes it difficult to blind patients as donor-side pain is expected. Finally, we do not expect patients to have an expectation of open- or arthroscopic-assisted graft reconstruction to be superior to the other.

Observer bias is minimized with a blinded musculoskeletal radiologist for the primary outcome, to prevent observer bias from favouring one of the interventions.

The secondary outcome (Q-DASH) is reported by the patient independently, thus with no implication to the assessor. Other secondary outcomes are evaluated by an independent, non-blinded observer.

The current standard treatment at our department includes follow-up until the union is established. In this study, we will have additional follow-up examinations at 6 months and at 1 and 2 years. It has the risk of keeping the patients in the role of being ill. However, the patients can have a feeling of extra good care with the possibility to address uncertainty or problems more easily.

Patients are at risk of being treated inferior with an intervention which is deemed inferior with the study analysis, but nothing a priori suggests which intervention is the better one.

The trial will provide high-quality evidence regarding time to union and short- and long-term functional outcomes of open and arthroscopic assisted graft reconstruction for scaphoid delayed/non-union.

The results from the study can contribute to establish a treatment algorithm for scaphoid delayed/non-union together with results from other studies.