Abstract
The first International Symposium on the Pre-Analytical Phase was held in Tampere, Finland in July 1995. Following this symposium, a second symposium was scheduled for 1996. This symposium was held in Keble College, Oxford, UK in July of 1996 and was attended by 58 delegates from 14 countries.
Automation Themes
Total laboratory automation, including robotics systems, is being introduced in selected laboratories with a high workload (more than 1000 specimens per day). Speakers gave illustrations of pre-analytical implications such as the design, labeling and handling of blood tubes; methods for biohazard containment within automated sample handling; and the cost-benefit of the exercise.
Tube design for automated handling requires consideration of the optimal tube size; the method of labeling each tube including location of bar codes; and the design of the tube cap to allow robotic piercing without the risk of aerosol generation when decapping. When there is sample aliquoting from primary to secondary tubes in the laboratory, there should be automatic transfer of the patient identification from the primary tube. To reduce the need for aliquoting, however, some automated laboratory systems require the phlebotomist to draw multiple primary tubes to serve different workstations, but this was felt to run counter to the need to reduce the volume of blood taken from patients. Also, there are cost benefits when multiple analytes can be measured in one primary tube.
Total laboratory automation requires rapid transport of blood tubes and it was recommended that rapid transit systems should be considered as a continuum that starts with transport from clinical areas to the laboratory and ends with automated specimen preparation and transport to the analyzer. While overhead pneumatic or rail-based systems are becoming popular for transport to the laboratory, robotic systems at bench height were more likely to remain the preferred option within the laboratory. The importance of continuous tracking and traceability of individual tubes was emphasized throughout this continuum from blood collection to analysis. This is important not only for ensuring sample identification but also for monitoring quality parameters for each tube such as centrifugation conditions (e.g. centrifuge temperature, speed, time, video recording of tube breakage) and test sample characteristics (e.g. haemolysis, microclots, icterus, lipaemiae, sample volume, and the effectiveness of refrigeration to below 15 C to reduce evaporative loss with the analytical system).
Two concepts of pre-analytical automation were presented. The sequential concept is based on a linear conveyor track with multiple instruments clustered around the track line. The selective concept, in contrast, relies on modular robotic systems that are used to automate discrete islands of the test process. While the Japanese systems of total laboratory automation with sequential linear testing may only become cost effective when the workload exceeds 1000–2000 specimens per day, the modular system was thought to have wider application by allowing selective automation in smaller laboratories. Indeed, selective automation of part of the pre-analytical phase (specimen preparation or front-end automation) might be the first choice for many smaller laboratories.
The discussion in this symposium session concluded that design requirements for the pre-analytical phase were as important as the design of the automated laboratory itself. Otherwise the full cost-benefit of automation would not be realized and the bottleneck would merely be transferred to the pre-analytical stage. While the clinical and cost-benefits arising from laboratory automation are difficult to quantify at this early stage of their development, workflow simulation can be used to estimate the effect of varying workload, workflow, transport, manpower and instrument configurations. Processing bottlenecks at various combinations of workload, test mix and manpower can then be assessed. Such modeling can also be used to optimize the physical layout of automated laboratories.