- Teodor P Grantcharov (), research fellowaa,
- Linda Bardram, consultant surgeonb,
- Funch-Jensen Peter, head of departmenta,
- Jacob Rosenberg, consultant surgeonb
- a Department of Surgical Gastroenterology L, Aarhus University, Kommunehospitalet, DK-8000 Aarhus, Denmark
- b Department of Surgical Gastroenterology 435, University of Copenhagen, Hvidovre Hospital, DK-2650 Hvidovre, Denmark
- Correspondence to: T P Grantcharov
Participants, methods, and results
The study was carried out in a gastroenterological surgical unit at a teaching hospital. A night shift started at 3 30 pm and finished at 9 am the following day. A total sleep time of less than three hours was necessary for inclusion in the study.
All 14 surgeons in training at our department— 11 men and three women—participated in the study. The median age was 34 (range 24-43) and the median time since graduation was six years (1-11 years). All trainees had similar, limited experience in laparoscopic surgery; the median number of cholecystectomies they had performed was 0 (0–5). All participants received identical pretraining on the minimally invasive surgical trainer-virtual reality (MIST-VR, Mentice Medical Simulation, Gothenburg, Sweden) by performing nine repetitions of six tasks. 1 2 The laparoscopic surgical skills of the 14 trainees were assessed on the 10th repetition of the task, which was performed during normal daytime working hours and again at 9 30 am after a night on call with impaired sleep. The period between the first and 10th repetition on the MIST was predetermined to be no longer than one month.
We analysed the data using non-parametric analysis (Wilcoxon test). We examined the difference between scores for error of motion, time of motion, and economy of motion measured during the 10th repetition of the task in the daytime and after a night on call.
The median total sleep time during the night on call was 1.5 hours (0-3 hours). After a night on call the time taken to complete the virtual laparoscopic tasks (P≤0.006) increased significantly for tasks 1, 3, 4, 5, and 6 (5.4 v 7.6 seconds, 5.6 v 7.8 seconds, 6.7 v 8.1 seconds, 15.0 v 18.1 seconds, and 18.2 v 23.8 seconds, respectively), and after a night shift surgeons performed significantly more errors in tasks 1 and 6 (0.6 v 1.0, P=0.01; and 1.4 v 3.5, P=0.005, respectively). The number of unnecessary movements for tasks 5 and 6 increased significantly after a night on call (7.8 v 9.4, P=0.008 and 6.1 v 8.2, P=0.004, respectively). (Data for all six tasks and a description of each task can be found on the BMJ's website.) The figure shows data from task 6. This task includes elements from most of the other tasks, is the most complex, and requires the highest levels of concentration and coordination. Previous studies have found that this task correlated best with surgical performance in vivo.3
Surgeons show impaired speed and accuracy in simulated laparoscopic performance after a night on call in a surgical department. Our results are consistent with the findings of Taffinder et al.4
Previous studies have shown that effects of sleep deprivation on cognitive performance do not become consistently apparent until after 36–40 hours of total lack of sleep.5 Our results show that significant deficits in psychomotor performance occur after 17 hours on call with disturbed night sleep. Factors connected with surgical work, such as emergency workload, stress, and emotional demands, may potentiate the effects of sleep deprivation alone.
Further studies should determine how long it takes for surgeons' laparoscopic performance to recover after an extended period on duty and should be aimed at developing and evaluating countermeasures that can maximise alertness and reduce fatigue.
Contributors: TPG and JR had the idea for the study and the design. TPG performed data collection, data entry, and statistical analysis, and compiled the first draft of the paper. PFJ and LB were involved in the design of the study, and revised critically and approved the final manuscript. JR is guarantor of the study.
Funding Sygekassernes Helsefond, Copenhagen, Denmark.
Competing interests None declared.
Data for all six tasks can be found on the BMJ's website