Prospective, randomized assessment of the acquisition, maintenance, and loss of laparoscopic skills.

Background: Laparoscopic skills are difficult to learn. We, therefore, assessed the factors involved in skill acquisition, maintenance, and loss in 2 prospective, randomized studies.

Methods: In study 1, 24 laparoscopic novices were randomly assigned to a control condition who performed the laparoscopic assessment task; Massed condition who trained on virtual reality (VR) simulation during 1 day or Interval condition who had the same amount of VR training distributed over 3 consecutive days. All groups also completed a novel laparoscopic box-trainer task on 5 consecutive days. In study 2, 16 laparoscopic novices were randomly assigned to a Practice or a No-practice condition. All subjects were required to train on a VR simulation curriculum for the same duration and skill attainment level. The week after completion of training, subjects in the Practice condition were allowed 1 complete practice trial on the simulator. Both groups completed the same tasks 2 weeks after completion of the training.

Results: In study 1, the Interval trained group showed the fastest rate of learning and on completion of training significantly outperformed both the Massed and Control groups (P < 0.0001). In study 2, both groups showed significant skills improvement from training trial T1 to T3 (P < 0.0001). The subjects in the Practice group maintained or improved their skills at 1 week but those in the No practice group showed significant decline of skills at 2 weeks after training completion (P < 0.0001).

Conclusions: Laparoscopic skills are optimally acquired on an Interval training schedule. They significantly decline with 2 weeks of nonuse.