109 A 1986 The Carry-over Effects of Triazolam Compa- red with Nitrazepam and Placebo in Acute Emergency Driving Situations and in Mono- tonous Simulated Driving Hans Laurell and Jan Törnroos Reprint from Acta Pharmacologica et toxicologica 1986 v Väg06/7 Efi/( Statens väg- och trafikinstitut (VTI) * 581 01 Linköping [St]tlltet Swedish Road and Traffic Research Institute * S-581 01 Linköping Sweden Acta pharmacol. et toxicol. 1986, 58, 182186. From the National Swedish Road and Traffic Research Institute, (V.T.I.), S-58 101 Linköping, Sweden The Carry-over Effects of Triazolam Compared with Nitrazepam and Placebo in Acute Emergency Driving Situations and in Monotonous Simulated Driving Hans Laurell and Jan Törnros (Received October 9, 1985; Accepted January 9, 1986) Abstract: Eighteen healthy volunteers of both sexes, aged 2034, were tested in the morning while undertaking real car driving avoidance manoeuvres and during monotonous simulated driving after 1 and 3 nights of medication with triazolam 0.25 mg, nitrazepam 5 mg or placebo. The study was a double-blind, randomized, cross-over study, where a minimum of 7 days wash-out separated the 3 treatment periods. Nitrazepam was found to impair performance in the simulated task after 1 but not after 3 nights of medication. Performance in the triazolam condition was not signicantly different from the other conditions on this task on either day. However, after one night of medication triazolam tended to score worse than placebo but better than nitrazepam. In real car driving a tendency was noted for nitrazepam to score worst, whereas the difference between placebo and triazolam was hardly noticeable. The same tendency appeared on both days. Keywords: Driver performance drug driving simulator. In recent years several new benzodiazepine deriva- When comparing the possible residual effects tives have been developed in search of hypnotics of various hypnotics on car driving performance, with effective hypnotic activity but which do not two aspects of driving stand out as particularly produce unwanted residual effects. This has great interesting, namely monotonous driving and social implications especially in societies where emergency avoidance situations. By employing transportation mainly relies upon private cars. A such test situations, the study would be able to state of continual alertness is of critical impor- yield information that concerns performance, re- tance when driving a car. The the wide use how- lated to driving in heavy city traffic as well as on ever, of sleeping pills may give rise to hang-over empty highways. effects that can impair driving performance. Triazolam is a triazolobenzodiazepine differing from most others in that it is active in very low doses and has a short mean elimination half-life Materials and Methods of 2.3 hrs (Eberts et al. 1981). Thus, it would Eighteen healthy volunteers of both sexes, aged 2034, reduce the risk of residual effects. who possessed a current Swedish Car Driving License, In this study, a comparison is made between were enrolled in the study. The trial employed a double-blind, randomized, cross triazolam and nitrazepam (with a mean elimin- over design. The study medications, triazolam 0.25 mg, ation half-life of 29 hrs) (Kangas et al. 1979) and nitrazepam 5 mg and placebo were each administered placebo. for 3 consecutive nights with a minimum of 7 days BENZODIAZEPINE EFFECTS IN SIMULATED DRIVING 183 between each treatment period. Assessments were made immediately. The time from stimulus appearance to ap- in the morning after 1 and 3 nights of medication. plication of the brake was the BRAKE REACTION TIME. Simulated driving task. In order to detect any changes 2. Auditory signals. An auditory tone, called for the same in susceptibility to driver fatigue, a monotonous 2.5 hrs reaction as the visual stimuli already described. driving task was created in a driving simulator. The Both the visual and auditory stimuli were presented driving compartment of the simulator consisted of a real, randomly at intervals of between 10120 sec. truncated car body, complete with all controls, which were parts of a feed-back system of which a digital- Real car driving. In the real car driving situation, in analogue computer was the brain. contrast to the simulated monotonous task an emer- A continually changing road pattern and landscape gency, evasive manoeuvre, was used (see fig. 2). was randomly generated by a computer. A large tele- The driving task was largely the same as the one used vision unit interpreted the signals and presented them in earlier studies on the effects of alcohol on driving Visually on a wide screen in front of the driver. The performance (Laurell 1977; Laurell & Törnros 1983). programme allowed for an everchanging pattern of road On observing an avoidance signal from an apparatus and landscape which included not only bends and mounted on the front of the vehicle, the driver had to straights but also up and down-hill sections. Technologi- carry out an avoidance manoeuvre and, in doing this, cal adaptions further mimicked real driving conditions try to avoid knocking over pylon cones which were by simulating Vibrations on the road surface which, placed along the avoidance path. The tolerance on either should the driver accidentally stray off the road onto side of the car was approximately 15 cm. The number the verge, increased in intensity. of cones knocked over was employed as a measure of Activation of the controls by the driver caused the driver performance. The signal could be presented at appropriate change in the visual scenario to occur i.e. either one of four positions in the cone setting. If present- turning, braking, etc. The driver was instructed to stay ed above the left headlight position, the signal appeared on the right side of the road and to maintain a steady as an obstacle in front of and to the left of the car, thus 90 km/hr throughout the test. requiring an avoidance manoeuvre to the right, and vice The computer Continually monitored the drivers per- versa. formance and functioning and at the termination of a The order of presentation of positions was random- test run presented a detailed assessment of the par- ized for each subject and each treatment condition. In ameters under study, brake reaction time, time off road each session, the course was negotiated 10 times plus 2 or across white line etc. blank runs by each subject. Two warm-up trials identical Whilst driving the simulator the subject was exposed to a test trial preceded each session. The total of 14 trials to a number of stimuli which, once perceived, called for lasted 30 min. a specic action to be taken. These events were pre- The road surface conditions are liable to considerable programmed and there was no way that the driver could variations so the test area was sprinkled with water at predict their presentation in advance. regular intervals to maintain constant driving con- ditions. Vehicle speed was controlled by an automatic speed control system. 1. Visual signals. Lights were placed both in the central In order to keep motivation at a high level throughout visual elds and to the periphery (see g. 1). When the investigation, the amount of payment was made they were illuminated the driver had to apply the brake dependent upon performance. For each trial the driver \ w/ \ / Stimulus positions Fig. 1. Stimulus presentation in the driving simulator. Drivers View. 184 HANS LAURELL AND JAN TÖRNROS _ san" /% . _: 0 . o o o 0 o o o o o o o o o o o o o o o o o o .4. reflector position . -o o . '|' .l. t+|-c _..I.|. O | 2 3 m | Fig. 2. Specications of pylon arrangement. had at his disposal a sum of 25 SEK (approximately 3.3 cedure was repeated to examine the effects of possible US$). The sum of money was reduced for each cone drug accumulation. knocked over. In the simulated monotonous driving task long reaction times (> 1.5 sec.) caused minor nancial penalties. Results Procedure. All subjects practiced the simulator task for Real car driving. one hour and the real car driving task for at least 2.5 hrs or until a minimum level of performance had been Although the differences did not achieve statistical reached. significance, (drug effect: F(2,34)=2.05; n.s. The capsules were taken at ll p.m. after which the Drug >< day interaction:F(2,34)< 1; n.s.) the subjects went to bed. After a standardized breakfast in worst performance on both assessment days l and the morning, they were brought to the simulator for 3 was found with nitrazepam, whilst triazolam testing at 8 a.m. Immediately after having completed the simulator test they took on the actual car driving task and placebo appeared to have about the same which lasted for 30 min. Two days later the whole pro- effect (fig. 3). BENZODIAZEPINE EFFECTS IN SIMULATED DRIVING 185 [_____I NITRAZEPAM 5 MG % TRIAZOLAM 0,25 MG reaction time performance in the simulated task. Very few signals were missed, a total of 29 DAY 1 1 DAY 3 8 8 (out of 8100). The nitrazepam condition, however, \ \\ s 91 & 93 scored worst on both days (10 and 6 missed sig nals, respectively). The triazolam condition scored best on both days (2 on both days) with the pla- PERFORMANCE PERFORMANCE cebo condition scoring in between (5 and 4 missed signals respectively). Very few subjects caused PLACEBO PLACEBO these misses and it was not considered meaningful THE THE to test these differences with respect to statistical OF OF signicance. °/o % . 3. Actual car driving (total means, 18 subjects). 2"! oo Discussion With the small doses of the test drugs administer- Monotonous simulated driving task.