[ Pobierz całość w formacie PDF ]
INTERNATIONAL UNION OF PURE AND APPLIED CHEMISTRY
ANALYTICAL, APPLIED, CLINICAL, INORGANIC AND
PHYSICAL CHEMISTRY DIVISIONS
INTERDIVISIONAL WORKING PARTY FOR HARMONIZATION OF
QUALITY ASSURANCE SCHEMES FOR ANALYTICAL LABORATORIES*
IUPAC/ISO/AOAC INTERNATIONAL/EURACHEM
HARMONISED GUIDELINES FOR THE USE OF RECOVERY INFORMATION IN
ANALYTICAL MEASUREMENT
(Technical Report)
Resulting from the Symposium on Harmonisation of Quality Assurance
Systems for Analytical Laboratories, Orlando, USA, 4-5 September 1996
held under the sponsorship of IUPAC, ISO and AOAC INTERNATIONAL
Prepared for publication by
MICHAEL THOMPSON
1
, STEVEN L R ELLISON
2
, ALE
Õ
FAJGELJ
3
, PAUL WILLETTS
4
AND ROGER WOOD
4
1
Department of Chemistry, Birkbeck College (University of London), London WC1H 0PP, UK
2
Laboratory of the Government Chemist, Queens Road, Teddington, Middlesex TW11 0LY, UK
3
International Atomic Energy Agency, Agency’s Laboratories Seibersdorf, A-2444 Seibersdorf,
Austria
4
MAFF Food Science Laboratory, Norwich Research Park, Colney, Norwich NR4 7UQ, UK
*Membership of the Working Party during 1995-99 was as follows:
Chairman
:
M. Parkany,1995-97, (Switzerland); A. Fajgelj, 1997- , (IAEA, Austria);
Members
: T.B. Anglov (Denmark); K. Bergknut (Norway); K.G. Boroviczeny (Grmany);
Carmen Camara (Spain); K. Camman (Germany); Jyette Molin Christensen (Denmark); S. Coates
(AOAC Int., USA); W.P. Cofino (The Netherlands); P. De Bievre (Belgium); T.D. Geary (Australia);
T. Gills (USA); A.J. Head (UK); J. Hlavay (Hungary); D.G. Holcombe (UK); P.T. Holland (New
Zealand); W. Horwitz (USA); A. Kallner (Sweden); H. Klich (Germany); J. Kristiansen (Denmark);
Helen Liddy (Australia); E.A. Maier (Belgium); H. Muntau (Italy); C. Nieto De Castro (Portugal);
E. Olsen (Denmark); Nancy Palmer (USA); S.D. Rasberry (USA); M. Thompson (UK); M.J. Vernengo
(Argentina); R. Wood (UK).
Republication of this report is permitted without the need for formal IUPAC permission on the
condition that an acknowledgement, with full reference together with IUPAC copyright symbol (ª
1995 IUPAC) is printed. Publication of a translation into another language is subject to the
additional condition of prior approval from the relevant IUPAC National Adhering Organization.
Harmonised guidelines for the use of recovery information in analytical measurement
(Technical Report)
Synopsis.
ISO, IUPAC and AOAC INTERNATIONAL have co-operated to produce agreed
protocols or guidelines on the “Design, Conduct and Interpretation of Method Performance
Studies” [1] on the “Proficiency Testing of (Chemical) Analytical Laboratories”
[2] and on “Internal
Quality Control in Analytical Chemistry Laboratories” [3]. The Working Group that produced
these protocols/guidelines was asked to prepare guidelines on the use of recovery information in
analytical measurement. Such guidelines would have to outline minimum recommendations to
laboratories producing analytical data on the internal quality control procedures to be employed.
A draft of the guidelines was discussed at the Seventh International Symposium on the
Harmonisation of Quality Assurance Systems in Chemical Laboratory, sponsored by
IUPAC/ISO/AOAC INTERNATIONAL, held in Orlando, USA, 4-5 September 1996 .
Proceedings from that Symposium are available [4].
The purpose of these guidelines is to outline the conceptual framework needed for considering those
types of analysis where loss of analyte during the analytical procedure is inevitable. Certain
questions cannot be satisfactorily addressed, and hence remain irreducibly complex, unless such a
conceptual framework is established. The questions at issue involve (a) the validity of methods for
estimating the recovery of the analyte from the matrix of the test material, and (b) whether the
recovery estimate should be used to correct the raw data to produce the test result. The types of
chemical analysis most affected by these considerations are those where an organic analyte is present
at very low concentrations in a complex matrix.
2
CONTENTS
1.
Introduction
2.
Definitions and Terminology used in the Guidelines
3.
Procedures for Assessing Recoveries
3.1
Recoveries from Reference Materials
3.2
Recovery information from Surrogates
3.2.1
Isotope Dilution
3.2.2
Spiking
3.2.3
Internal Standards
3.3
Matrix Mismatch
3.4
Concentration of Analyte
4.
Should Recovery information be Used to Correct Measurements?
4.1
Arguments for Correction
4.2
Arguments against Correction
4.3
Rational and Empirical Methods
5.
Estimation of Recovery
5.1
Representative Recovery Studies
5.2
Internal Quality Control
6.
Uncertainty in Reporting Recovery
6.1
Estimating Uncertainty in a Recovery
7.
Conclusions
8.
Recommendations
9.
References
Appendix: Uncertainty in Reporting Recovery
3
FOREWORD
It is recognised that the use of recovery information to correct/adjust analytical results is a
contentious one for analytical chemists. Different sectors of analytical chemistry have different
practices. Formal legislative requirements with regard to the use of recovery factors also vary
sector-to-sector. It is the aim of IUPAC, however, to prepare general Guidelines which may be
seen to aid the preparation of the “best estimate of the true result” and to contribute to the
comparability of the analytical results reported.
This document attempts to give Guidelines that are intended to be general in their scope and give
recommendations that reflect common practice best able to achieve the above. However, specific
sectors of analytical chemistry will need to develop these Guidelines for their own requirements and
the recommendations are not, therefore, to be seen as binding for all areas of analytical chemistry.
1.
INTRODUCTION
The estimation and use of recovery is an area where practice differs among analytical chemists. The
variations in practice are most obvious in the determination of analytes such as veterinary drug
residues and pesticide residues in complex matrices, such as foodstuffs and in environmental
analysis. Typically, such methods of analysis rely on transferring the analyte from the complex
matrix into a much simpler solution that is used to present the analyte for instrumental
determination. However, the transfer procedure results in loss of analyte. Quite commonly in such
procedures a substantial proportion of the analyte remains in the matrix after extraction, so that the
transfer is incomplete, and the subsequent measurement gives a value lower than the true
concentration in the original test material. If no compensation for these losses is made, significantly
discrepant results may be obtained by different laboratories. Even greater discrepancies arise if
some laboratories compensate for losses and others do not.
Recovery studies are clearly an essential component of the validation and use of all analytical
methods. It is important that all concerned with the production and interpretation of analytical
results are aware of the problems and the basis on which the result is being reported. At present,
however, there is no single well-defined approach to estimating, expressing and applying recovery
information. The most important inconsistency in analytical practice concerns the correction of a
raw measurement, which can (in principle) eliminate the low bias due to loss of analyte. The
difficulties involved in reliably estimating the correction factor deter practitioners in some sectors of
analysis from applying such corrections.
In the absence of consistent strategies for the estimation and use of recovery information, it is
difficult to make valid comparisons between results produced in different laboratories or to verify
the suitability of those data for the intended purpose. This lack of transparency can have important
consequences in the interpretation of data. For example in the context of enforcement analysis, the
difference between applying or not applying a correction factor to analytical data can mean
respectively that a legislative limit is exceeded or that a result is in compliance with the limit. Thus,
where an estimate of the
true concentration
is required, there is a compelling case for compensation
for losses in the calculation of reported analytical result.
These Guidelines provide a conceptual framework for consistent decisions on the estimation and use
of recovery information in the various sectors of analytical science.
4
2.
DEFINITIONS AND TERMINOLOGY USED IN THE GUIDELINES
General analytical terminology is assumed to be accepted when these Guidelines are read, but
specific definitions of the terms most pertinent to the Guidelines are given below:
Recovery:
Proportion of the amount of analyte, present in or added to the analytical portion of the
test material, which is extracted and presented for measurement.
Surrogate:
Pure compound or element added to the test material, the chemical and physical
behaviour of which is taken to be representative of the native analyte.
Surrogate Recovery
:
Recovery of a pure compound or element specifically added to the test
portion or test material as a spike. (Sometimes called "marginal recovery".)
Native Analyte:
Analyte incorporated into the test material by natural processes and manufacturing
procedures (sometimes called “incurred analyte”). Native analyte includes “incurred analyte” and
“incurred residue” as recognised in some sectors of the Analytical Community. It is so defined to
distinguish it from analyte added during the analytical procedure.
Empirical Method of Analysis
:
A method that determines a value which can be arrived at only in
terms of the method
per se
and serves by definition as the only method for establishing the
measurand. (Sometimes called “defining method of analysis”.)
Rational Method of Analysis
:
A method that determines an identifiable chemical(s) or analytes(s)
for which there may be several equivalent methods of analysis available.
3.
PROCEDURES FOR ASSESSING RECOVERY
3.1
Recovery Information from Matrix Reference Materials
In principle, recoveries could be estimated by the analysis of matrix reference materials. The
recovery is the ratio of the concentration of analyte found to that stated to be present. Results
obtained on test materials of the same matrix could, in principle, be corrected for recovery on the
basis of the recovery found for the reference material. However, several problems potentially beset
this use of the reference materials, namely: (a) the validity of any such recovery estimate depends
on the premise that the analytical method is otherwise unbiased; (b) the range of appropriate matrix
reference materials available is limited; and (c) there may be a matrix mismatch between the test
material and the most appropriate reference material available.
In the last instance the recovery value obtained from the reference material would not be strictly
applicable to the test material. The shortfall applies especially in sectors such as foodstuffs analysis
where reference materials have to be finely powdered and dried to ensure homogeneity and stability.
Such treatment is likely to affect the recovery in comparison with that pertaining to fresh foods of
the same kind. However, matrix mismatch is a general problem in the application of recovery
information, and is treated separately in Section 3.3.
5
[ Pobierz całość w formacie PDF ]