Khanduja Vikas, Lawrence John E, Audenaert Emmanuel
Department of Trauma & Orthopaedics, Addenbrooke's Hospital, Cambridge, United Kingdom.
Department of Trauma & Orthopaedics, Addenbrooke's Hospital, Cambridge, United Kingdom.
Arthroscopy. 2017 Mar;33(3):566-571. doi: 10.1016/j.arthro.2016.09.028. Epub 2016 Dec 16.
To test the construct validity of the hip diagnostics module of a virtual reality hip arthroscopy simulator.
Nineteen orthopaedic surgeons performed a simulated arthroscopic examination of a healthy hip joint using a 70° arthroscope in the supine position. Surgeons were categorized as either expert (those who had performed 250 hip arthroscopies or more) or novice (those who had performed fewer than this). Twenty-one specific targets were visualized within the central and peripheral compartments; 9 via the anterior portal, 9 via the anterolateral portal, and 3 via the posterolateral portal. This was immediately followed by a task testing basic probe examination of the joint in which a series of 8 targets were probed via the anterolateral portal. During the tasks, the surgeon's performance was evaluated by the simulator using a set of predefined metrics including task duration, number of soft tissue and bone collisions, and distance travelled by instruments. No repeat attempts at the tasks were permitted. Construct validity was then evaluated by comparing novice and expert group performance metrics over the 2 tasks using the Mann-Whitney test, with a P value of less than .05 considered significant.
On the visualization task, the expert group outperformed the novice group on time taken (P = .0003), number of collisions with soft tissue (P = .001), number of collisions with bone (P = .002), and distance travelled by the arthroscope (P = .02). On the probe examination, the 2 groups differed only in the time taken to complete the task (P = .025) with no significant difference in other metrics.
Increased experience in hip arthroscopy was reflected by significantly better performance on the virtual reality simulator across 2 tasks, supporting its construct validity.
This study validates a virtual reality hip arthroscopy simulator and supports its potential for developing basic arthroscopic skills.
Level III.
测试虚拟现实髋关节镜模拟器髋关节诊断模块的结构效度。
19名骨科医生在仰卧位使用70°关节镜对健康髋关节进行模拟关节镜检查。医生被分为专家（进行过250例或更多髋关节镜手术的医生）或新手（进行过少于此数量髋关节镜手术的医生）。在中央和周边间隙内可视化21个特定目标；9个通过前侧入路，9个通过前外侧入路，3个通过后外侧入路。紧接着进行一项测试关节基本探针检查的任务，其中通过前外侧入路探测一系列8个目标。在任务过程中，模拟器使用一组预定义指标评估医生的表现，包括任务持续时间、软组织和骨骼碰撞次数以及器械移动距离。不允许对任务进行重复尝试。然后使用Mann-Whitney检验比较新手组和专家组在这两项任务中的表现指标来评估结构效度，P值小于0.05被认为具有显著性。
在可视化任务中，专家组在完成时间（P = 0.0003）、与软组织碰撞次数（P = 0.001）、与骨骼碰撞次数（P = 0.002）以及关节镜移动距离（P = 0.02）方面优于新手组。在探针检查中，两组仅在完成任务的时间上存在差异（P = 0.025），其他指标无显著差异。
在虚拟现实模拟器上两项任务的表现明显更好，这反映出髋关节镜经验增加，支持其结构效度。
本研究验证了虚拟现实髋关节镜模拟器，并支持其在培养基本关节镜技能方面的潜力。
III级。
