European Multicentre Evaluation of the ESAT 6660.

The analytical performance of the glucose analyser ESAT 6660 from Eppendorf was studied according to the ECCLS guidelines and partly according the SFBC protocol in a multicentre evaluation involving laboratories from three European countries. The glucose determination in serum and in haemolysate was studied. The following results were obtained. 1. The precision was as good as or better than the precision of the comparison instruments. The coefficients of variation were between 1.1 and 3.4% for the between-days imprecision and between 0.35 and 1.45% for the within-run imprecision experiment. 2. The recovery of control sera values compared with the hexokinase method was between 94.3 and 102.6%. 3. With patient specimens as good agreement was found between the results obtained with the ESAT 6660 and the different comparison instruments (ASTRA, Hitachi 737 and ACP 5040). 4. A drift effect of 1.1-2.3% occurred in 5 of 21 experiments, depending on the individual enzyme membrane. 5. Sample carry-over was not observed. 6. A linearity between 0.5 and 50 mmol/l was found, exceeding the manufacturer's claims. 7. Several different endogenous and exogenous interferences were investigated. No interfering effect was detected for endogenous substances. A positive interference was observed by ascorbic acid at a concentration above 350 mg/l. 8. The practicability of the instrument was judged as very good. It was considered as a disadvantage that the instrument is not capable of piercing sample lids. Also the numeration of samples is not very convenient.