Temperature control and volume measurement in clinical analysers

Temperature control and volume measurement are two instrumental factors that, at different stages of the analytical cycle (both in manual and automatic work), have a significant impact on the quality (accuracy and precision) of the analytical result. This paper reports work carried out by two committees which prepared guidelines on the definition and control of these two instrumental factors of analytical reliability.

The AHCTC's starting point was an attempt to define the parameters which describe temperature control and to establish the requirements for such parameters in the clinical laboratory.
The time-course of the temperature in a liquid mass (of temperature To), which is transferred in an environment thermostated [1] at a set-point temperature (Ts), is shown schematically in figure 1. The AHCTC expressed the opinion [2] that the periodical temperature variation which occurs after the set-point has been approached is better defined in terms of'permissible deviation [from the set-point]' than in terms of accuracy and precision. The 'equilibration time' and the 'permissible deviation' (figure 1) are the parameters which, together with a third, 'temperature uniformity', describe temperature control in clinical analysis. The latter refers not only to the temperature uniformity among different tubes and/or cuvettes in a rack, in a block or in a rotor, but also to temperature uniformity in a single cuvette (or tube) [3]. The next step was to define the requirements for temperature control, and the permissible deviation was regarded as the most critical parameter for this. Considering that multi-purpose analysers are used to perform different tests, and assuming that the narrower tempera-  Tm.CVp ATm Kt which relates the permissible deviation (ATm) to the assay temperature (Tm), to the partial error due to temperature variation (CVp) and to a factor (Kt), which is a function of the energy of activation of the enzyme.
Assuming, for the most commonly measured enzyme activities, that activation energy values are in the range 35-60 kj/mol [4], the values for permissible deviation listed in table 2 were calculated.
Following this approach, temperature requirements are not given in fixed terms as previously [5], but in relation I71 instrumental factors of the analytical variability (like pH control, volume measurement, absorbance measurement, and wavelength calibration) are not strictly enough controlled to allow for lower than, say, 4% overall error, it may result in a useless effort to achieve better than +0.2 K (or +0.2 C) temperature control. A temperature control within +0.05 K should be aimed at, which will give an overall error in kinetic measurements of 1%.
The verification of such a narrow permissible temperature deviation requires a high-quality thermometer, such as a standard platinum resistance thermometer [6]. The high price of these sophisticated systems leads to the question of cost/benefit; one which is difficult to address.
The verification tools (thermometers) may be calibrated with reference to the fixed points of the International Practical Temperature Scale-1968 (IPTS-68) or to some suggested fixed points for the life sciences [6], as listed in table 4. Among these, the rubidium triple point standard [8] may be especially useful for checking the accuracy of thermometers intended for measuring temperatures near to 37 C. In addition to those listed in table 4, the second or third order fixed points (6,9) listed in table 5 may be of help in the range 22.4-42.7C. In practical work, although the recommended procedure for calibrating thermistor thermometers includes comparison with a recently calibrated platinum resistance thermometer [10], comparison at three temperature values with a certified mercury-in-glass thermometer, in a well equilibrated water-bath [11], may be an acceptable procedure. allow the temperature of the incubation mixture (Ts) at the start of the kinetic measurement (ts) to be: Ts Tm + A Tm. Further work of the AHCTC will involve to the preparation ofguidelines to verifying how the expected in-the-cell temperature control is being achieved. Basically, for this operation, appropriate tools and feasible methods for checking calibration are needed. The tools are listed in table 3." the Pt-100 resistance thermometer may represent a 'reference method' [6], but well calibrated thermistors, featuring very low thermal mass and thin enough probes to be introduced in small-volume cuvettes, may be adequate in terms of accuracy and sensitivity. The use of thermochromic solutions [7] may allow temperature verification in fixed or rotating spectrophotometric cuvettes which are unaccessible to the probe, but the inherent accuracy and the sensitivity of this measuring system have to be proved. Other methods, listed in parentheses in table 3, have no practical application at the moment. Table 3. Tools for monitoring the temperature in reaction mixtures.
(Radiation thermometry) (Liquid crystal thermography) 26"87 C Na2SO4.10H20/Na2SO4 (trsp): 32"37 C n-icosane (tp): 34  Italy). Because of the changed structure of the EPI, the WPSDD ended before any guideline was produced; nevertheless, collaborative efforts produced some results and these are discussed here.   [12], and are also being considered in draft documents by ISO [13].   3[a]). A diluting operation is obtained; if the specimen volume is set to 0, dispensing is achieved. In a second mode ( figure 3[b]) the system is filled with a washing or inert fluid and both the reagent and the specimen are aspirated and distributed through the tip. Here again diluting and dispensing operations can be performed. These two operating modes include the presence of a valve. POVAs can also be operated in a without-valve mode, as shown in figure 3(c). This last operating mode also allows sampling operations to be easily performed.
REAGENT WASHING OR Table 7. Devices for sampling, diluting and dispensing which are commonly used in mechanized analysis.
(2) Peristaltic pumps. In automated work, liquid-displacement systems are most commonly used; they can be operated in three main modes (figure 3). The system may be filled with the reagent (or diluent): the desired amounts of reagent and of specimen are aspirated, respectively, from a reservoir Since the reagent-to-sample volume ratio is frequently high, the systems operate with two pistons (syringes) of different diameter, but single-piston instruments may also be operated in three modes cited above.
POVAs' pistons are driven in different ways, see table 9. Significant improvements have recently been achieved in this by the introduction of, microprocessor-controlled stepper motor driven pistons; the additional monitoring, via external sensors, ofthe fluid's level may lead to further reliability of volume measurements in the zl range. The performance characteristics ofvolumetric apparatus may be defined by precision and accuracy of volume measurements, and by carry-over and dead volume. These characteristics may be verified by means of 'primary' and 'secondary' methods [14], as shown in table 10. Since the secondary methods are based upon comparison with gravimetrically verified volumetric equipment, the assessment ultimately relies upon gravimetry. Specifications concerning balance performance, and calibration by means of NBS standards, procedures and statistical calculations have been published 14]. In the secondary methods, the use of potassium dichromate [14] and ofEvans blue [15] solutions has been suggested. For the secondary-method verification of samplers intended for serum measurement, the use of iso-viscosity dye solution, such as Evans-blue-dyed-serum, has been suggested 16]: this may be particularly important for the verification of air-displacement POVAs.