Medical analysis has considerably improved the diagnostic possibilities by increasing the number of individual analyses of a patient's serum. In order to limit the total amount of a patient's blood to be collected for this purpose, it is becoming increasingly important to analyse liquid volumes of only a few microlitres. Such small liquid volumes are dispensed with manually controlled piston-operated pipettes and microlitre syringes and with similar automated systems. In accordance with the relevant standards, all these devices are calibrated using water as the dispensing liquid. There are doubts, however, about the validity of these calibrations as far as the dispensing of serum is concerned. The dispensing behaviour of water and serum was therefore investigated by PTB.
The PTB study is the first to confirm and quantify differences in the dispensing behaviour. Positive displacement pipettes do not react very strongly to a change of the dispensing liquid. The differences found in the dispensed volume at nominal volumes greater than 10 µl were smaller than 1% and sometimes even smaller than 0,1%. In the case of air displacement pipettes, the differences in the dispensed volume were, however, considerably larger, reaching values of about 5%. Therefore, calibration with the serum to be dispensed is urgently required at least for air displacement pipettes.
The gravimetric calibration of dispensing devices with serum requires that the density of the serum be exactly known in each individual case. What has been known to date is that this density varies around a statistical mean value of about 1,024 g/cm3. The PTB study now presents a very simple way to determine this density from the content of protein and salt, the main components of serum, with a relative uncertainty of only 3 · 10-4 . As these quantities are routinely measured in almost every serum analysis, no additional sophisticated density measurement is required.
