The primary aim of this study was to identify the rate of osteoarthritis in scaphoid fracture non-union. We also aimed to investigate whether the incidence of osteoarthritis correlates with the duration of non-union(interval), and to identify the variables that influence the outcome. We retrospectively reviewed 273 scaphoid fracture non-union presented between 2007 and 2016. Data included patient demographics, interval, fracture morphology, grade of osteoarthritis (Kellgren-Lawrence) and scaphoid non-union advanced collapse (SNAC), and overall health-related quality of life. Patients were divided into two groups (SNAC and Non-SNAC). Group differences were analysed using Mann-Whitney U test and association with Pearson's correlations. A two-sided p-value of <0.05 was considered significant. The scaphoid fracture non-union were confirmed on CT scans (n=243) and plain radiographs (n=35). The subjects were 32 females and 260 males with the mean age of 33.8 years (SD, 13.2). The average interval was 3.1 years (range, 0–45 years). Osteoarthritis occurred in 58% (n=161) of non-unions, and 42% (n=117) had no osteoarthritis. In overall, 38.5% (n=107) had SNAC-1, 9% (n=25) with SNAC-2, and 10.4% (n=29) presented with SNAC-3. The mean interval in the non-SNAC group was 1.2 years, and in SNAC 1,2, and 3 were 2.6, 6.8, and 11.1 years, respectively. The average summary index in SNAC and non- SNAC groups was 0.803 and 0.819, respectively. Our results also showed a significant correlation between advanced osteoarthritis and proximal fracture non-unions(P<0.05). We concluded that there is no clear correlation between the interval and the progression of osteoarthritis. SNAC was more likely to occur in fractures aged 2 years or older

Longstanding un-united scaphoid fractures or scapholunate insufficiency can progress to degenerative wrist osteoarthritis (termed scaphoid non-union advanced collapse (SNAC) or scapho-lunate advanced collapse (SLAC) respectively). Scaphoid excision and partial wrist fusion is a well-established procedure for the surgical treatment of this condition. In this study we present a novel technique and mid-term results, where fusion is reserved for the luno-capitate and triquetro-hamate joints, commonly referred to as bicolumnar fusion. The purpose of this study was to report functional and radiological outcomes in a series of patients who underwent this surgical technique. This was a prospective study of 23 consecutive patients (25 wrists) who underwent a bicolumnar carpal fusion from January 2014 to January 2017 due to a stage 2 or 3 SNAC/SLAC wrist, with a minimum follow-up of one year. In all cases two retrograde cannulated headless compression screws were used for inter-carpal fixation. The clinical assessment consisted of range of motion, grip and pinch strength that were compared with the unaffected contralateral side where possible. Patient-reported outcome measures, including the DASH and PRWE scores were analysed. The radiographic assessment parameters consisted of fusion state and the appearance of the radio-lunate joint space. We also examined the relationship between the capito-lunate fusion angle and wrist range of motion, comparing wrists fused with a capito-lunate angle greater than 20° of extension with wrists fused in a neutral position. The average follow-up was 2.9 years. The mean wrist extension was 41°, flexion 36° and radial-ulnar deviation arc was 43° (70%, 52% and 63% of contralateral side respectively). Grip strength was 40 kg and pinch strength was 8.9 kg, both 93% of contralateral side. Residual pain for activities of daily living was 1.4 (VAS). The mean DASH and PRWE scores were 19±16 and 29±18 respectively. There were three cases of non-union (fusion rate of 88%). Two wrists were converted to total wrist arthroplasty and one partial fusion was revised and healed successfully. Patients with an extended capito-lunate fusion angle trended toward more wrist extension but this did not reach statistical significance (P= 0.07). Wrist flexion did not differ between groups. Radio-lunate joint space narrowing progressed in 2 patients but did not affect their functional outcome. After bicolumnar carpal fusion using retrograde headless screws, patients in this series maintained a functional flexion-extension arc of motion, with grip-pinch strength that was close to normal. These functional outcomes and fusion rates were comparable with standard 4-corner fusion technique. A capito-lunate fusion angle greater than 20° may provide more wrist extension but further investigation is required to establish this effect. This technique has the advantage that compression screws are placed in a retrograde fashion, which does not violate the proximal articular surface of the lunate, preserving the residual load-bearing articulation. Moreover, the hardware is completely contained, with no revision surgery for hardware removal required in this series