[Clinical application of three-dimensional O-arm navigation system in treating patients with dystrophic scoliosis secondary to neurofibromatosis type Ⅰ].

Objective: To investigate the clinical outcomes and the accuracy of O-arm-navigation system assisted pedicle screw insertion in dystrophic scoliosis secondary to neurofibromatosis type Ⅰ(NF-1). Methods: A retrospective study was conducted in 41 patients with dystrophic NF-1-associated thoracic scoliosis who were surgically treated at Department of Orthopaedics, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School between June 2012 and October 2014 with more than 18 months follow-up. The patients were then divided into two groups: 18 patients were under the assistance of O-arm-navigation-based pedicle screw insertion (O-arm group) and the remaining 23 patients' pedicle screws insertion were conducted by free-hand (free-hand group). The X-ray and CT were analyzed to investigate the correction rate and safety of pedicle insertion. t-test was used to analyze measurement data and χ(2) test was used to analyze accuracy of screw insertion between the two groups. Results: The mean coronal Cobb angle was 63.2°±8.7° in the O-arm group and 66.9°±7.4° in the free-hand group (P>0.05), which was then corrected into 23.1°±6.8° and 30.2°±7.6°(t=2.231, P=0.031) after surgery respectively.Operation time was (265.0±70.3)minutes and estimated blood loss was (1 024±465)ml in the O-arm group. Operation time and estimated blood loss was (243.0±49.6)minutes and (1 228±521)ml respectively in the free-hand group, which had no significant difference between the two groups. However, the implant density was higher in the O-arm group than that in the free-hand group ((64.1±10.8)% vs.(44.3±15.3)%)(t=4.652, P=0.000). The O-arm group comprised 122 screws, of which 72.9% were excellent, 22.1% were good and 4.9% were bad. The free-hand group comprised 136 screws and 48.5% of them were excellent, 33.8% were good and 17.6% were bad.Accuracy of pedicle screw insertion was higher in the O-arm group than that in the free-hand group(χ(2)=10.140, P<0.05). By June 2016, the average follow-up period was (20.9±3.4)months(ranging from18 to 26 months), including (20.3±3.1)months in the O-arm group and (21.4±5.5)months in the free-hand group. At last follow-up point, coronal correction loss was significantly higher in the free-hand group than that in the O-arm group (6.3°±2.6° vs. 4.4°±1.6°)(t=2.719, P=0.009). Conclusions: Compared with free-hand technique, O-arm-navigation technique could enhance accuracy of pedicle insertion and implant density of dystrophic region in dystrophic NF-1-associated scoliosis patients, which result in a better correction rate and less correction loss. Besides, the advantage of O-arm-navigation do not increase operative time and estimate blood loss.